Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
VOLUME 40
R.M. Chisholm (Brown University. Rhode Island); Mats Furberg (Goteborg Univer-
sity); D.A.T. Gasking (University of Melbourne); H.L.A. Hart (University College.
Oxford); S. Komer (University of Bristol and Yale University); H.J. McCloskey (La
Trobe University. Bundoora. Melbourne); J. Passmore (Australian National Univer-
sity. Canberra); A. Quinton (Trinity College. Oxford); Nathan Rotenstreich (The
Hebrew University. Jerusalem); Franco Spisani (Centro Superiore di Logica e Scienze
Comparate. Bologna); R. Ziedins (Waikato University. New Zealand)
The titles published in this series are listed at the end o/this volume.
Kotarbinski:
Logic, Semantics and Ontology
edited by
Jan Wolenski
Jagiellonian University, KrakOw
and The Technical University, Wrodaw, Poland
..
KLUWER ACADEMIC PUBLISHERS
DORDRECHT / BOSTON / LONDON
Library of Congress Cataloging in Publication Data
ISBN-13:978-94-010-7442-1
Preface vii
Notes on Contributors lX
Tadeusz Kotarbinski
Philosophical self-portrait
Kazimierz Ajdukiewicz
Review-article: T Kotarbinski's Elements of the Theory of
Knowledge, Formal Logic and Methodology of the Sciences 7
Alfred Gawro~ski
Psychologism and the principle of relevance in semantics 23
Peter Geach
Names in Kotarbinski's Elementy 31
Andrzej Grzegorczyk
Consistent reism 39
Henryk Hiz
A note about reism 47
Janina Kotarbinska
Puzzles of existence 53
Czesraw Lejewski
On the dramatic stage in the development of Kotarbinski's
pansomatism 69
Marian Przer~cki
Semantic reasons for ontological statements: the argumentation
of a reist 85
Tadeusz Pszczolowski
Philosophical and methodological foundations of Kotarbinski's
praxiology 97
vi
Vito F Sinisi
Kotarbinski's theory of genuine names 107
Vito F Sinisi
Kotarbinski's theory of pseudo-names 119
Barry Smith
On the phases of reism 137
Klemens Szaniawski
Philosophy of the concrete 185
Jan Wolenski
Kotarbinski, many-valued logic, and truth 191
BogusJaw Wolniewicz
Concerning reism 199
Ewa Zarneck"a-Bialy
The voice of the past in Kotarbinski's writings 205
References 213
Jan Wolenski
NOTES ON CONTRIBUTORS
ix
PHILOSOPHICAL SELF-PORTRAIT
Tadeusz Kotarbinski
convictions, and this is the kind of ethics that I advance despite the
opposi tion of both the traditionalists and the revolutionaries. The
former accuse me of free-thinking which supposedly severs ethics from a
belief in justice ruling the world and in the redemption of guilt and
merit within eternity; the latter, on the other hand, reject any system
of universal ethics and consider moral judgements to be .intrinsically
relative: nothing can be permanent or universal as ethics is believed to
stem from the needs of social groups while these needs are viewed as
essentially changeable depending on circumstance and historical moment.
The kind of ethics described above, on the contrary, advances an intrin-
sic identity of human conscience in any society. For it seems that
assessment of human deeds against the scale 'commendable-disgraceful'
has evolved through recurrent situations when facing danger it was
necessary for the stronger to defend the weaker. Heroes of defence
action of this kind are worshipped everywhere while cowards and cravens
who cannot be relied upon in such circumstances are universally held in
bontempt. What differs from society to society is the emphasis placed on
various virtues of which the moral equipment of the human being is sup-
posed to be composed. Moreover, the degree to which vices as contrasted
to virtues are condemned is also variable.
Ethics has not'become the subject of my professional scholarly con-
centration even though its problems have always been of utmost signifi-
cance to me. I am too skeptical about the possibility of working out a
detailed system of principles underlying wisdom, a system which would
lend itself to intersubjective substantiation. What is more, teaching
ethics professionally, in my opinion, does not tally with the essence of
its problems unless one cultivates a historical, sociological, explicat-
ive knowledge of styles and conceptions of morality instead of tackling
the essential issues which always demand prospective, prescriptive
answers. Only then the subject taught would no longer be ethics but a
kind of science des moeurs, transcending the specific tasks of this
department of philosophy. I reached the chair of philosophy via logic.
Teaching logic became the field of my activity as a university professor
of philosophy, a member of other humanistic faculties. Emphasis is here
placed on the words 'teaching' and 'humanistic'. For my lectures and
classes were conceived as an organon in the classical sense of the term,
for philosophers as well as for those who, having completed their course
of study, would espouse the cause of disseminating humanistic knowledge
and thinking, particularly future secondary school teachers. Somewhat
later my activity embraced also law students. My linguistic equipment
proved to be very helpful in this respect. For it seems especially im-
portant when the problems of an organon of this kind are conceived his-
torically and is quite crucial when pondering the original Organon of
Aristotle (or to be more cautious, of the peripatetic school) and its
continuators. Conceived in this m~nner logic was by no means confined to
formal logic, but came to'comprise the problems of epistemology, seman-
tics and methodplogy. It is precisely the latter problems - not those of
formal logic - that were of particular interest to my mind. Neverthe-
less, I felt bound to contribute to the study of formal logic. The
feeling was encouraged both by my colleagues at Warsaw University and by
my awareness of the precise phase that logic had reached in its histori-
4 TADEUSZ KOTARBINSKI
cal development. It was precisely the moment when mathematical logic was
triumphantly entering the scene. The names of Frege, Bertrand Russell,
Peano, Burali-Forti, Coutu rat and many others were on everybody's minds.
Mathematical logic was closely allied with the rapidly developing set
theory. The international periodical devoted to the latter, Fundaaenta
Mathematicae was, and still is, published in Warsaw. The distinguished
philosophic-mathematical logicians: Jan Lukasiewicz and his disciple and
my colleague, Stanislaw Lesniewski among many others, were active here.
I only mention the names of those persons to whom my studies in mathema-
tically oriented formal logic are particularly indebted. In this respect
lowe a lot to my close alliance with Professor Lesniewski. I simply
took over his original system of formal logic to suit my own purposes. 1
Relieved thus from the necessity to contribute to formal logic itself, I
could concentrate on the problems I faced as a teacher of logic to be
used by humanists. These centred around the problem of overcoming the
hypostases of linguistic origin, what Francis Bacon referred to as idola
for i. Both our everyday language and the language of the sciences as
well are teeming with nouns or noun-like forms. Hence the tendency to
perceive an object behind them even when the noun is an abstract one,
like for example, 'roundness', 'equality', etc. Once the existence of
the alleged objects of such names is admitted, once we agree to the
existence of such qualities or relations, human thought is made to wade
through a mire of apparent ontological problems. They in turn impose a
literal interpretation of the expressions like 'a quality inheres' in an
object in the same way as a nail is embedded in a wall: whereas, in
point of fact, their meaning is only metaphorical. Leibniz himself was
of the opinion (which he expressed in Nouveaux essais 2, XXII, § 1) that
problems bristling with difficulties can be dispelled as soon as we
stick only to the names of concretes in our discourse. Unaware both of
these words and equally ignorant of Franz Brentano' s similar idea 2 I
formulated in 1929 the principles of the so called reism. In its most
mature formula it declares war against the hypostases of linguistic ori-
gin on the following lines: inasmuch as it is possible try to formulate
statements in a way that would eliminate all names other than the names
of objects, that is, physical bodies or parts thereof. Persons ought to
be regarded as objects, i.e. sentient objects. Sentences may contain
words that are not names, e.g. verbs or conjunctions, etc. The point is,
however, to eliminate naaes other than the names of objects. Let me
hasten with an example of a reistic interpretation of sentences. 'Pru-
dence inheres in wisdom' simply menas: 'Every man who is possessed of
wisdom is prudent.' 'Bonds of brotherhood related Orestes to Electra'
simply lIeans: 'Orestes was Electra's' brother.' A reist by no means
demands that the use of sentences with abstract expressions like the
names of qualities or relations be completely abandoned. Quite the
contrary, the necessity of applying them is fully recognised just
2It. waa he who oit. .. d L.ibniz'. word. in a foot.not.e t.o hi .. (1911) - a.e
'p', 163 in t.he 19:16 edit.ion, voL 193 of t.he Meiner Philoaophiache
8ibliot.hek, compare oleo Brent.ano'a dict-ot-ed works, •. 9. (1916).
PHILOSOPHICAL SELF-PORTRAIT 5
because their presence may often reduce the length of the statement. The
only thing he insists upon is to try to be able to do without nalles
which are not the names of things. I lIay add, by the way, that at
present I prefer to use the term 'concretism I instead of the term
'reism' as my readers were prone to identify 'reism' with 'realism'
while the meanings of the two are totally different.
Thus, reisll, that is concretism (or somatism - as I identify all
objects with bodies and in Greek 'solla' means 'body') proves to be a
certain innovation of my organon. It is however, . highly debatable as a
conception since a number of difficulties inhering in an attempt to
interpret reistically theorems of set theory have not yet been overcome.
If I were to be asked, however, about the main field of my scholar-
ly interest, I would disregard logic for the sake of a different discip-
line of science, namely praxiology. Until quite recently little has been
known about the existence of this branch of science, although as early
as 1890 Espinas delineated its .perspectives in the Revue Philosophique.
There is advised observation of the progress in various practical skills
with a view to improving the efficiency of any purposeful activity. I
approached praxiology via a different route as I was ignorant of Espinas
proposals. What struck me as odd was that concepts like 'activity',
'manner', 'method' or 'product' (which comprehend various elements or
phases of activity) are currently used together with those which convey
a technical, as it were, assessment of the activity itself: 'carefully',
'accurately', 'waste', 'bungler', 'a shoddy piece of work', while none
of the existing branches of science seemed to deal with them all. Yet a
separate science which would ponder the conditions of efficient action
in general, with reference to all spheres of human endeavour seelled
quite indispensable. For also in ethics, 'where the matter concerns among
others the pr9blem of living rationally, one invariably must have resort
to this or other recommendations of economy - for example when the
method of the golden mean is advised, or indolence and inaccuracy in
doing the job are censur~d. Similarly, an appropriate meting out of res-
ponsibility depends upon a rational general knowledge of the relations
between the activity and its intended and undesigned results.
On the other hand, even within the organon such matters as, for
example, economy in demonstrating the truth or falsity of a theorem or
in constructing the system of premises, are often taken into considera-
tion. And once we gain an insight into the wealth of possible modes of
acting, we are bound to encounter irritating paradoxes that call for
explanation, e. g. that one may act by simply doing nothing or that
greater forces can be overcome by means of smaller ones. It was also be-
cause of the~e reasons that I took an interest in praxiology. Praxiolo-
gical problems loomed especially large when the matters of organization
and management came to be reflected upon on a sufficiently general level
with respect to their application in solving the problems of industry
and trade. My interest in these matters dates back to my undergraduate
days at the university, and it was uninf1uenced either by my professors
or my fellow-students. Over sixty years ago I published in Polish my
first book on the subject, but I tackled the problems of praxiology
quite frequently afterwards. During the period when I held the presiden-
cy of the Polish Academy o~ Sciences the first team of scientists in ,the
6 TADEUSZ KOTARBINSKI
Kazimierz Ajdukiewicz
In his preface to the book with which we shall be concerned here the
author describes his work as a textbook for those students of the human-
ities who are preparing for the examination in "the principles of philo-
sophy", which is obligatory for all those candidates for the degree of
Master of Arts in Polish uni versi ties who do not take philosophy as
their principal subject. He also thinks that it may serve as a.textbook
for those who are to sit for an examination in logic, methodology and
epistemology, obligatory for those who major in philosophy and pedagogy.
This presentation of the work as a textbook may give rise to a bias
against its scientific value, since textbooks usually are compilations,
expositions of things already accepted, and as a rule do not contribute
much new to science. In the present case, although the work is called a
textbook, it is not merely that. Professor Kotarbinski's book comprises
a wealth of tersely formulated original opinions which would suffice to
fill a number of monographs. To Polish culture it contributes new
values, not only didactic, but also scientific.
The author in many places presents his own views concerning episte-
Bology, methodology, ontology (in the Aristotelian sense) and semantics.
A.nd where he relates epistemological, metaphysical and methodological
trends in the history of human thought, he does not confine himself to
summarIzlng other people's oplnlons, but. undertakes original and
valuable analytical study in that he offers a thorough and penetrating
analysis of important philosophical issues and clearly describes the
various standpoints on those issues. In its analytical part the book
reaches a level which is approached by very' few works in the world
literature on the subject. Neither does the author confine himself to
interpretative analysis alone, but subjects the trends in question to a
cri tical and always profound scrutiny, in doing which he very often
formulates his own opinion.
1AII I>age-ref'erences are t.o t.h .. English t.ranslat.ion of' t.he Element.y.
t. •. t.o Kot.arbinski (1966).
7
I. Woletfsld (ed.), Kotarbitfsld: Logic. Semantics and Ontology, 7-21.
© 1990 Kluwer Academic Publishers.
8 KAZIMIERZ AJDUKIEWICZ
2. A REVIEW OF CONTENTS
The book consists of five parts. The first entitled Remarks on language,
is concerned with the principal semantic relations, such as expressing,
meaning and denoting, with the kinds, or categories, of expressions,
with defects of language and with definitions as a means of eliminating
such defects. That part also includes the exposition of the author's
basic standpoint, which he calls reism. We call that standpoint basic,
because it is from that point of view that the author settles many phi-
losophical issues which are considered classical. Here, too, the reader
will find an exposition of the problem of universals and the formulation
of the author's nominalist standpoint on that issue. All of the Part One
consists of an original contribution by the author, only broken in a few
places by the presentation of results known otherwise.
The second part deals with epistemology. We find there a profound
analysis of the various shades of idealism and realism, and the exposi-
tion of the author's own standpoint, which he calls radical realism. We
find there also an analysis of the concepts of truth, criterion of
truth, and an analysis of the terms: dogmatism, scepticism and criti-
cism. All these analyses are preceded by a study of the meanings of the
terms 'perceptual image', 'reproductive image', and 'productive image',
by a distinction as between the immanent and the transcedent object of
an image and the content of an image. The chapter concerned with con-
cepts naturally belongs here also.
The third part, the least original in the book, presents informa-
tion on formal logic, almost entirely known from other sources. It con-
sists of the theory of deduction (an axiomatic exposition of Russell's
system as simplified by Lukasiewicz), traditional logic, and some infor-
mation on what is called Lesniewski's ontology.
An original contribution in the last-named chapter seems to consist
in bringing out the fine relationship between the formulas of the sen-
tential calculus and the formulas of the calculus of terms.
The fourth part is concerned with a general methodology of sciences
and falls into three chapters. The first, on reasoning, works out the
classification of reasonings known from Lukasiewicz's article (1912).
REVIEW-ARTICLE: T.KOTARBINSKI'S ELEMENTS •.• 9
3. REISM
the sentenre ~hich is thus ohtainerl may he true or false but is never-
theless meaningful.
Now jf we say that a class of expressions is a closerl semantic cat-
egory (cf. p. 56) we have to add in whirh language it is so. The author
sp<~ms to claim that. it must be so in every language. The antinomies of
'self-mpmbership', sur;h as Russell's antinom~-, the antinomy of rela-
t i nilS, the ant i nomy of prop<"rt i es, etc., haH' prov ioed arguments for the
assert ion that if ~e inclurle in the same semantir category expressions
which perform different funct.ions in the sentence, for inst:mc-" some-
t.hing which is a funr;tor and something which may be its argument, we run
the risk of contradictioll. The rule prohihiting, in any language,
inclusioll of such types of expressions in the same ~emantic r:ategory, is
t.hus to be observed under the penalty of contradiction. Now it does not
s(>em that tlwre are "'1ually strong arguments in favour of the prohibi-
t iOll to includic' in the same semllntic category, for inst:mce, the wor-d
'tllol,,' alld such words as 'death', 'paill', etc. In other words, I do not
think that ";e 1'110 the risk of an antinomy if we consider meaningful the
spntpllces ·'something is a pain' aorl 'sompthillg is Il tllble', if thp mean-
Irlgs of the other words occuring in such sentencPH are rptained.
The fact Ulat sPIlt.encec; including 'appal'ent terms' can he replacpd
hy e(lIlisigni ficant ones ~hich do not include Ilppllrent terms proyes only
that, Ilt IF-ast theorf'fically, ~"ca.n do without apparent terms. This
suggpsts ()ne more illtt"rpret.!1tiO)1 of semantic· reism, an interpretlltion
which is not a. thesis hilt a progrilmme. This programme might postulate
t.hllt ilpparent terms should not be lIsed unless the sentences in which
they oreur CRn, by mean~ of appropriate definitions, he reformulated as
e~lisignifica.nt sentences free from apparent terms. This would result in
the arhitrary Ildoption of the principle that in the language we use, the
h'rms which dpnote things form a. closed spmantie category, which means
that only tprms which denote things may meilningfully be substituted for
terms ..-hieh denotp things. I t must be admi tted that the advantage of
such a 1anguage cons i sts in the fad that it el i mi nates a number of
philosophical issues as fictitious.
Tn such a language the principal thesis of real reism, which we now
procepd to analyse, stating that eyery object is a thing (p. 56) turns
out to bf' a truism. 'All A is B' means 'for any x, if x is A, then x is
B, and some x is A' (p. 194).' Hence the thesis of real reism takes 0)1
the form: • for any x, if x is an ohject, then x is a thing, and some .Y
is an object'. It is a truism that some x is an object. And since in a
reist's language \ can take on as its values only names of things (the
variable x is limited to that semantic category hy the context in which
it occurs), hence any value of the variahle x which veri fies the antece-
dent also verifies the consequent of the implir;ation formulated above,
so that that implication must always be true. But real reism does not
confine itself to the positive assertion that every object is a thing,
hut pmphasizes a number of negative theses, such as 'relations do not
exist', 'properties do not exist', etc. Now these theses, if formulated
in -a reist's literal'language, entail an inr;onsistency on his part,
since they are meaningless in that language, and if formulated in a me-
taphorical sense, they are false.
REVIEW-ARTICLE: T.KOTARBINSKI'S ELEMENTS ••• 13
salle semantic category; we would rather think that the words 'whiteness'
and 'looks white' are words of the same semantic category. But we do not
agree on the place in which the author draws the boundary, among the
nominal phrases, between the various semantic categories. We also do not
agree with the postulate of reducing all the sentences that include
nominal phrases to sentences containing nominal phrases of one and the
same semantic category. Neither do we clai .. that such a division of
expressions into semantic categories is found in the (Polish) language,
nor that any other division is faulty.
4. SEMANTICS
of the two terms connected by the copula 'experiences'. For that purpose
such a sentence is interpreted so: 'John experiences so and so', where
that 'so and so' plays the role of an adverbial qualifier which deter-
mines more closely the kind of the experience involved. Now the author
so imagines that adverbial determination that he recommends the follow-
ing manner of speking: 'John experiences this: Dh! It hurts!' In this
way the sentence in question breaks into two parts: the announcement,
before the colon, and the information, after the colon.
By means of the connective 'this', which does not occur as a term,
but, as the reviewer thinks, is to be interpreted as an adverb, the
author continues to explain, "it is implied that the speaker will him-
self in some way experience something imitatively, in the same way as
John experienced something, and that the phrase which he utters will be
a formulation of that experience". (p. 346)
Now the following interpretation of the role of the connective
'this' suggests itself to us: that connective is an occasional adverb.
And the occasion determining the use of that adverb would be the imitat-
ive experience stated in the phrase that follows the colon. In a state-
ment like 'John experiences this: the sun shines!' only that which pre-
cedes the colon would be a complete sentence, while that which follows
the colon would play the sallie role as pointing one's finger in accompa-
niment of sentences in which the occasional demonstrative pronoun 'such'
is a (general) term standing for a physical object.
The second interpretation of the role of the connective 'this'
seems less satisfactory to render the author's intentions; it refers to
the second part of the explanation quoted above that "by means of that
connecti ve it is implied that the speaker will himself in some way
experience imitatively, taking as his model how John experienced, and
that the phrase which he will speak will be a statement of that experi-
ence." Now i f Peter by saying 'John experiences this: bats are probably
birds' implies by the use of the connective 'this' that the phrase which
he will speak after the colon will be a statement of John's experience,
then Peter's statement might be interpreted as: 'John experiences some-
thing which is (indirectly) stated by the phrase "bats are probably
birds'" (and not 'experiences the same as is now directly stated by that
phrase') •
Let us now consider both interpretations. Suppose that a person
says 'John has a suit like that' and points to a certain suit. Now in
the situation described above that sentence may be interpreted in
various ways. It may be that John's suit resembles the suit pointed to
in its cut, and perhaps also in its colour, and perhaps also in some
other respect. Likewise, when I say 'John experienced this: bats are
probably birds' and understand this statement in accordance with the
first interpretation as 'John experienced so', with that occasional 'so'
determined by the situation in which I say 'bats are probably birds',
then it is not known in what respect John experienced just so. Was it in
respect of modality, or in respect of content, or in respect of the
naivety of that belief?
Also in the second interpretation, by which the statement 'John
experienced this: bats are probably birds' turns out to be the same as
'John experienced something which is (indirectly') stated by the phrase
18 KAZIMIERZ AJDUKIEWICZ
'bats are probably birds' the same ambiguity remains, for we may ask,
'what does it state in what respect?'
Let us now concentrate attention on the relationships between the
definitions we are concerned with. Now "states as to content" is defined
by the phrase 'John thinks that q'. The phrase 'John thiks that q' is
explained by 'John experiences this: q'. Now we see that the last phrase
is ambiguous for both interpretations described above, and it cannot be
made univocal by the phrase "as to content", for a vicious circle would
.result. Hence, for instance, the above explanation of 'John thinks that
q' by 'John experiences this: q' cannot be corrected by the replacement
of the second part by 'John experiences as to content this: q' without a
resulting vicious circle.
Let us now confront the definition of (simple) (indirect) statement
wi th the definition of statement as to content. The forRler, in an ab-
breviated and modified form, would be: the sentence A states indirectly
John's thought to be such and such (for instance, the thought that the
sun shines) if the hearer who understands the language in question knows
from that sentence that John had in some respect such a thought as the
thought described (for instance, the thought that the sun shines). The
latter would be: the sentence A states as to content John's thought that
q (for instance the thought that the sun shines), if the hearer who
understands the Janguage in question knows from that sentence that John
thought that q (that the sun shines).
What has been said above seems to indicate, when the phrase 'John
thought that q' is intepreted according to the scheme 'John experienced
this :q', that (simple) statement and statement as to content become one
and the same.
For in both intepretations 'John experienced this: the sun shines'
is more or less the same as 'John experienced in some respect so (as I
shall experience in a moment when saying): the sun shines', or 'John had
a thought which in some respect is the same as the thought that the sun
shines' •
In this way, statement as to content cannot be used in a defini-
tion of meaning. For if, for instance, the phrase 'certainly!' indirect-
ly states (in respect of modality, and hence in some respect, and hence
states simply) every firm conviction of mine, hence the conviction that
war is not a factor of progress, and the conviction that 2 x 2 = 4,
etc., and hence it would also state them as to content. Consequently
'certainly!' would mean in Polish both that war is not a factor of
progress and that 2 x 2 = 4. And not only would it mean this on one
occasion and that on another ocasion, but it would always mean both.
The situation might be saved if the phrase 'John experienced this:
bats are probably birds' were replaced by 'John experienced in respect
of content something like that: bats are probably birds'. But this is
impossible in view of a vicious circle.
Still another interpretation of the phrase 'John experienced this:
the sun shines' might suggest itself to us. It might be said that the
phrase used here is 'this', and not 'such'; and when a person says 'John
experienced this' and adds the sentence 'the sun shines', he does not
want to say that John experienced an identical (in the logical sense of
the term) thought with that which is directly stated by the phrase 'the
REVIEW-ARTICLE: T.KOTARBINSKI'S ELEMENTS 19
sun shines' used in the context described above, but he wants to say
something more than that John experienced a thought which is the same in
some respect. He wants just to say that John experienced a thought which
in every respect is the same as that stated in the phrase in question,
though not identical with that thought. This will hereafter be called
the third interpretation.
Without raising the difficult problems connected with the differen-
tiation between 'the same in every respect' and 'identical','and inter-
preting intuitively the difference which it is difficult to bring out,
we shall point to other difficulties resulting from that third interpre-
tation of the definition of meaning now under discussion.
In such an interpretation of the scheme of psychological statements
as given by Professor Kotarbinski it would follow, it seems, that it
would be possible to say truly 'A means in language L that q' if and
only if q would have the same form as the sentence A. In other words, if
the sentence A runs: 'John is a teacher', then only the following state-
ment about the meaning of that sentence will be true: the sentence 'John
is a teacher' means in English that John is a teacher. Any Qther state-
ment about the meaning of that sentence will be false. For instance, it
will be false to say: the sentence 'John is a teacher' means in English
that John is a man whose profession is to teach.
According to our analysis of Professor Kotarbinski definitions:
'John is a teacher' means in English that 'John is a man whose pro-
fession is to teach' is the same as: "a hearer who understands English,
on having heard from the person X, who thinks that John is a man whose
profession is to teach, the sentence 'John is a techer', will know from
that statement that X experienced this: John is a person whose pro-
fession is to teach." Now a hearer who understands English, on having
heard from the person X the statement 'John is a teacher', will know
thereby that X experienced this: John is a teacher. He will not at all
know that X experienced this: John is a man whose profession is to
teach. To experience this: John is a teacher, is something quite other
than to experience this: John is a man whose profession is to teach, if
it is assumed that 'to experience this: John is a teacher' is the same
as to experience what is the same in every respect as what has been said
above in the sentence quoted after the colon.
An experience which is in every respect the same as the experience
stated directly by the sentence 'John is a teacher' is not in every
respect the same as the experience stated by the sentence 'John is a man
whose profession is to teach'. The characteristics of the experience of
the former kind includes this, that they have as an element an image of
the word 'teacher', which fact is not a property of the latter experi-
ence. Vice versa, the characteristics of the second experience include
the fact that when we have that experience we in someway think of a
profession, whereas in the case of the first experience no thought of a
profession comes into our head.
Hence whoever hears X say 'John is a teacher' will thereby know
only that X experienced this: John is a teacher. The same applies to
other examples. When X says 'this is a case of cancer' no one will
impute to X that X experienced this: this is a case of carino.a, since
everyone realizes the difference between the experience: this is aces
20 KAZIMIERZ AJDUKIEWICZ
'A is B' referring to that object (with the primary understanding of the
copula I is' )." Now it is not known whether this is a formal definition
of denoting, or a manoeuvre for teaching purposes, intended to secure a
fairly good understanding. From the point of view of teaching purposes
it would probably be advisable to precede this quasi-definition with a
number of examples showing the use of the word 'denotes'. And if this is
to be treated as a formal definition, the following objections arise.
First, I fear that the definition "The name N denotes in language L the
object 0, is the same as: the name N can be used in language L as a
subjective complement in a true sentence of the type tA is B' concerning
an object d', may lead to the antinomy of heterosemantic words.
Secondly, what does the formulation "a sentence of the type tA is B'
concerning an object d' mean? If the author's intention is to define it
as equisignificant with the formulation "a sentence of the type tA is B'
with the subject denoting the object d', then these two definitions
would form a vicious circle. And I do not know whether that formulation
could be defined in any other way. In any case, the formulation in
question is not defined by Professor Kotarbinski at all. Hence the
objection of a vicious circle is not to the point, but we have to raise
the objection of obscurity instead. The formulation "a sentence of the
type tA is B' concerning an object d' is not clear at all, for it is not
known whether, for instance, the sentence 'Galileo is Copernicus' is a
sentence concerning Galileo only, or concerning Copernicus as well.
Still worse, it is not known whether the sentence 'Kosciuszko's heart is
buried in Cracow' is a sentence concerning Kosciuszko, or Kosciuszko's
heart, or perhaps the body of King Casimir the Great buried in Cracow,
etc. In view of its obscurity that formulation requires a definition,
which, however, in my opinion, can be given in such a way as to lead to
a vicious circle.
So much for Professor Kotarbinski's semantics, although many other
remarks suggest themselves to the present writer. [ ••• ]
PSYCHOLOGISM AND TI£ PRINCIPLE OF RELEVANCE
IN SEMANTICS
Alfred Gawronski
this question in Ele.enty, and so we can make the assumption, that the
author used the expression 'as to content' in order to mark the
difference between stating and other kinds of expressing, if we accept
that stating is a special case of expressing.
II
III
The problems discussed above have so far been pushed aside by most phil-
osophers of language as belonging to the field of linguistic pragmatics.
However, if we conclude that the primary function of uttering sentences
and texts consists in selecting from a vast number of perceived or known
or possible facts, only (and necessarily only) those which enter through
mutual relations into a system of relevance(s), and that the second
function of language consists in drawing attention to those, and only
those facts at appropriate moments, with a presupposition substantiating
their relevance, then it is unlikely that these functions fail to have
a bearing on the structure of language.
Indeed, this is the case. If we observe the presently evolving
scientific discipline of systems of classification of objects in various
ethnic languages, we find that in the Sixties and Seventies a different
approach to verbal classification emerged. Until recently, including the
period of structuralism, the thesis proclaiming arbitrariness of the
linguistic sign was predominant; however, the new discipline, called
taxonomy (Robert Sokal, Eleanor Rosch and others), forced scholars to
reconsider the simplistic formulation of the problem. If we examine, for
example, the structures of the system of lexical terms in natural
languages, one can easily see that the fundamental factor delimiting
extensions and the degree of vagueness of sillple lexical teras is the
relevance of the type of object and of its delimitation in the
border-line cases in the total system of our knowledge about the world.
First of all, we do not naae all the objects that we perceive (and so,
in some sense, that we think about), but we extract from anonymity only
sOlie of them, to a different degree, by means of various types of naaes
or terms, ranging from singular to more and more general. Singularity of
a term and extensions of general terms are by no means accidental and do
not originate arbitrarily. As is well known, there is a close connec-
tion between the extension of a term and its intension or connotation.
If in aaking a statement I draw someone's attention to fact F, and I am
forced to eaploy a term whose extension is too narrow and intension too
rich for the purpose of my couunication (for there is no other, more
appropriate, terll in the language), then my message is fraught with
PSYCHOLOGISM AND THE PRINCIPLE OF RELEVANCE IN SEMANTICS 29
Peter Geach
In many cases the unity of a phrase that seems to compel one's re-
cognition is demonstrably not a logical unity; and I think an alterna-
tive analysis, which would not recognize the binding together of words
into a complex name, is always possible. This holds even for the
simplest case, of a qual j fying adjective attached to a noun. 'Only a
shameless woman smokes' is not of the form 'Only an A !/Is', coming out
trne just in case '<j) s' is truly predicable of nothing not covered by the
shared flame '.-1'; rather it means 'A woman - i.e. an,Y woman - smokes only
if she is shameless', and here the unity of 'shameless woman' has been
broken lip. T will not di late upon thi.s, because as T argued in various
papers j n my co llect.ion Logic Matters (1972), many examples show how
brHtl~ is the intuitive foundation whim 'complex terms' arE' taken to be
]ogkal unities. r hasten to acid that I do not hereby doubt the exist-
ence of logieaJ ly eomplex predicates. As I indicated, Aristotle may at
Oil€' time have dreamt of analysing our discourse in terms of complex
hondings of simple one - onoma one - rhema. propositions; but there is 110
slIch symmetry as he might suggest between names and predicahle expres-
sions. A name c['tn be simpl€', in a sense, it must be simple, because any
inner complexity is alien to its naming role. But a predicable may need
1:0 be complex; having in mind some named objPct. or objects, no limit can
be drawn to t.h€' complexity of what we want to say about it or them.
The cat€'gory difference that I recognize between names and predi-
cahles j'3 not denied by the logic of Lesniewski, which Kotarbinski
follows in Elementy; for a name wi II still be of different category from
a functor taking one name as argument t.o produce a sentence; and as
Plato says in the Sophist, a pair of names, or of such functors (Plato's
'rhemat:3.', my 'predicables'), will produce no logos, no coherent dis-
course. In Ajdukiewicz's scheme of categories, two expressions of cat-
€'gories sin and n yield one of category s, that is a sentence, but two
expressions both of cat.egory n or both of category sin will form a syn-
tactically incoherent string. Of course I am not saying that the Les-
n.iew!'lkian system overrides t.his distinction, only that the distinction
is in practice obscllred. In using the form 'A est B' it is hard to sup-
press a feel i ng as if one t€'rm had subject ro Ie, t.he other predicate
role; now either 'A est B' and 'B est A' are both well formed, or
neither is; since a predicable may certainly be complex, as I have said,
it is then hard to see why a name in subject role may not be. But a cri-
tiquE' of the Lesniewskian apparat.us would take me too far.
I turn to Kotarbinski's valiant battle against onomatoids, na.zwy
pozorne. It is a worthy battle, but sometimes appears like the Irish
hero's fight with his sword against the sea waves; for Kotarbinski is
fighting against a strong tendency of natural languages, at least of the
Indo-European family. How readily we turn from 'When a man marries his
troubles begin' to something like • A man's marriage makes the beginning
of his troubles'! (Even in 'his troubles begin' there is of course an
onomatoid, but one that we can get rid of immediately by saying 'he
begins to be troubled'). And in many years of teaching logic I found a
strong resistance of my pupils' linguistic habits to the use of letters
for any category other than names. There was some resistance even to the
use of a letter (schematic letter or bindable variable) for verbs or
predicables; there was great resistance to this move for the category of
34 PETER GEACH
'Because Smith did not unload the gun, 8mi th was wounded because
Smith's little boy played with the gun.'
'The car crashed because the driver was not careful and attentive.'
'The crash of the car was caused by the driver's lack of due care and
attention. '
To hypostatize individual car crashes may come natural to us, but
hardly: to hypostatize individual lacks of care and attention! It is
precisely here that theories treating causality as a proper relation
often go manifestly wrong - when they have to deal with negative
elements in causation. To mention one well-known example, John Mackie's
accounts of cause factors are open to this criticism.
36 PETER GEACH
all six areas in aggregate form one square, but the number of squares is
six (one big square, four small squares and one square formed by the
diagonals). When people jeer at Frege for rejecting the simple solution
that numerical predicates attach to aggregates, they simply stick to
cases when it is easy to think of an aggregate and forget cases when t.he
attachment of numbers cannot be thus explained. But although numbers
NAMES IN KOTARBnisKI' S "ELEMENTY" 37
Don't you ever find any difficulty in your theory that nUllbers are
objects?
Sometimes I seem to see a difficulty, but then again I don't see it.
Andrzej Grzegorczyk
x is a /\ y is x ---+ x = y
x is a ---+ x is x
and one axiom-schema:
3 an (x is an - - x is x A ••••• )
which is a version of the axiom of comprehension (ausonderungs-axiom).
Models for ontology are of course complete atomic Boolean algebras
(see Grzegorczyk, 1955).
Kotarbinski was greatly influenced by Lesniewski. They both
considered ontology as a theory of the word 'is' in its fundamental
sense, like e.g. in the sentence: 'Socrates is man' or 'Socrates is
Greek', but not in the sentences: 'Greek is man' nor 'Horse is animal'.
(I omit the prepositions which are of course necessary in normal English
speech patterns.)
Lesniewski's ontology is an elegant theory, but I doubt if it would
be suitable for a very reistic philosopher.
'Socrates' is of course the name of a thing, but 'man' and 'Greek'
are not. Hence in ontology the names for things and the names for sets
of things fall in the same type. This is a good formalization of the use
of the word 'is' in everyday speech. But for a philosopher there should
be a big difference between an individual thing which is the fundamental
enti ty of the universe, and a set of things which is an abstraction.
This distinction should be especially relevant for a reistic
philosopher.
Hence I claim that the appropriate ontology for reistic philosophy
is simply first order functional calculus with identity. The unique type
of variable may be interpreted as the names of things. Thus the quant-
ifier binding a variable: 3 x ( •.. ) can be read as: there is such a
thing x that •••• Then we can prescribe the meaning of real existence to
the existential quantifier, which fits well with the general ideas of
reisll. In Lesniewski's ontology we can not read the existential as:
there is such thing x that ••• , because x can be not a thing. It may be
a set like horses or animals.
One can say that there are several notions of existence or of
quasi-existence. Then the existential quantifier in other theories
(which are interpreted) may express the other notions of existence or
quasi-existence. Thus the simple type theory may of course be understood
as an ontology, in which only the existential quantifier binding the
first order variables express real existence while the existential
quantifiers binding variables of other types express only a kind of
quasi-existence. Such interpretation seems to be suitable for a reistic
philosopher. One can go farther and assume a set theory (and the whole
of mathematics) as a quasi-ontology or as a sci ence-fi ct ion-ontology. It
seems to me quite convenient to assume that mathematics is not a theory
about something really existing; but then perhaps we should not call
mathematics a science, if by science we mean a discipline providing
knowledge about reality. Mathematics may be comprehended as a useful
fantasy, which facilitates our manipulation of things. It becomes a
little mysterious why speculations about non-existent objects can enable
us to construct real existing things or to predict their functioning.
CONSISTENT REISM 43
There are no doubt about some connections between the reistic philosophy
of Tadeusz Kotarbinski and some formal endeavours of Stanislaw Lesniew-
ski (hi s sJ'stems of ontology and mereology) and of Alfred Tarski
(geometry of bodies). 3 In this section I shall go farther in this
direction.
Interpreting geometry or· topology as descriptions of the relations
in real space a reistically educated topologist should of course dislike
speaking about points· as fundamental entities of real space. The real
components of the space should have the full dimension of the whole
space. Such components may be called geometrical bodies. (Topologically
they may be def ined as interiors of the i r own closures, i. e. open
domains in the topological terminology of Kuratowski (1948). E.g. ge-
ometrical spheres (in the sense of whole globes) are geometrical bodies.
Tarski (1956a) sketched a geometry having the notion of sphere and the
inclusion between bodies as the two unique primi ti ve notions Tarski' s
definitions of other spatial relations are very ingenious.
To take this further I would suggest founding a topoiogy without
points which may precede Tarski's geometry of bodies as the classical
topology precedes classical geometry. The topology of bodies may be
developed in two steps. The most general topology of bodies may be a
theory having two primitive notions: 1. the relation of inclusion
between bodies 'e' and 2. the predicate of connectedness 'cd'. I have
the candidates for two axioms: .
AI. A, B cd /'. A rI B '" 0 ----t A I) B cd
A2. X cd A C comp Y .', X - Y cd ----t X - C cd
where 'c comp Y' means: C cd /\ C c Y A 'II D (D cd ACe Dey ----t C = D).
Al and A2 are written in Boolean algebra notation. Using mereological
notation (which seems to be appropriate for this mode of thinking)
instead of An B'" 0 we shall write: 3X (X c A A X e B).
Al and A2 are known as very general properties of connectedness.
The notion of connectedness is very intuitive and in classical topology
may be considered as more general than the notions of closure and
interior, which are traditionally assumed as primitives of topology. One
can prove that the closure operation is not definable by means of the
notion of 'cd' on the ground of all sentences true in all topological
spaces. This observation suggests as the second step a theory having one
primitive notion more (Grzegorczyk, 1960).
Even using higher types one can profit from some reistic intuitions in a
classification of definitions.
Suppose that we have a number of well founded empirical notions
which denote some properties of things or some relations between things.
Let us call them: empirical primitives. Starting from empirical primi-
tives, by means of logical operations one can define some other notions
concerning things.
These newly defined notions may be called more or less empirical
·according to the logical means applied in the definition. The classical
Jloolean operations of the calculus of propositions are most empirical.
The quantification limited to some empirically well defined sets is more
empirical than quantification which is not limited. If we pass from
quantifiers binding thing-variables to quantifiers binding higher types
then there is only one kind of definition which may be thought of as not
exceeding empirical definability. It is a definition with one existen-
tial quantifier at the beginning of the definiens:
aRb +-+ 38 (F(S,a,b»
S may be a variable of higher type, but the formula 'F(S,a,b)' should be
empirical. Some important empirical notions have such a shape. E.g. the
definition of numerical equivalence of two sets of things. It is
interesting to realize how many notions of science are very far from
empirical verifiability if we assume their speculative theoretical
definitions.
Our life runs in the world of things. We can imagine more. But we
should begin by the analysis of our thing-full environment. Then from
the methodological point of view the reistic philosophical programme of
Kotarbinski looks most attractive.
A NOTE ABOUT REISM
Henryk Hiz
Some authors accept the principle that the expressions that jointly
occur as arguments of a functor must all be of the same order. This
principle resembles the simple theory of types. Others allow arguments
of mixed order (in which case a changed is needed in (3), to the effect
that m is the highest order of an argument of G).
Kotarbinski mentions explicitly only the Nj,l where Nj,l is called
the subject. It is a relic of Aristotle's .notion that an elementary
sentence is composed of a name and a predicate. There is no good reason
for a reist to insist on the theory of singularity of subject, or singu-
larity of argument. In the sentence 'Johny and Jill got married' the
words 'Johny' and 'Jill' ~re names of things and can be considered argu-
ments of 'got married' which is a first order sentence-forming functor
from two arguments (in our culture). In the sentence 'The teacher gave
the book to a boy' the words "the teacher', 'the book', and 'a boy' are
names of things while 'gave' is a first order functor froll three argu-
ments. (The case indicators like 'to' show that these tlouns are argu-
ments.) However, according to Aristotle, and presumably according to
Kotarbinski, the functor of the first order in this sentence is 'gave
the book to a boy'. This functor can, in turn, be decomposed into a
functor-forming functor 'gave the book' and its argument 'a boy'. There
is no good reason to debate whether the grammatical theory of Aristotle
and Kotarbinski is reistically more correct than the theory shown in
points (1), (2) and (3). Each of them is compatible with reism.
Kotarbinski wants all sentences to be reducible to sentences which
have names of things as arguments. This reducibility can be understood
in two ways: either loosely, as in the theories just sketched, or more
rIgorously as the requirement that every sentence be translable into a
sentence in which the arguments are names of things exclusively. The
more rigorous understanding is suggested in Ele.enty and in Kotarbin-
ski's other writings. But it cannot be taken quite literally. In the
sentence 'Jill plays tennis when she has free time' the expression
'when' is a functor of two sentences, not of two names. The sentence
'For some F, F(the teacher, the book, a boy)', or lIore simply 'Something
took place between the teacher, the book and a boy', cannot be trans-
lated into a sentence in which there are only names of things and a
functor of them; at least not easily. The rigorous 'demand of reducibili-
ty leads to an artificial language difficult to study. Usually, it is no
longer English (or Italian, or Polish or the like). But sentences are
sentences of English (or Italian, or Polish). It is therefore worthwhile
to limit the discussion to the looser understanding of reducibility.
There are' functors of zero degree, which can for. a sentence with-
out requiring any arguments. In other words, they can constitute'senten-
ces by themselves. When one uses only functors of zero order, the hier-
archy given in (1), (2) and (3) is simplified. Zellig Harris (1982) pro-
poses to use only two kinds of functors in grammar: functors of argu-
me'nts that are nouns and functors of arguments at least sOlie of which
are sentences. Nouns are for him only those, expressions that can be
names of things. Harris is a reist among grumarians, and not without
Kotarbinski's influence.
A NOTE ABOUT REISM 49
P.S. There is a difference between the use of the pair of English words
'noun' and 'name' and the pai r of roughly corresponding Polish words
'nazwa' and 'illlie:. The difference arises frolll the fact that English
'name' has the verbal 'to name' and 'noun' does not have a verbal form.
In Polish, on the other hand, 'nazwa' has the verbal fora 'nazywac' and
'imi~' does not have one. There are also disributional differences; e.g.
'the name of the city', *'the noun of/for the city', *' imi~ miasta',
'nazwa miasta'. Kotarbinski uses 'nazwa' for all the cases of nallles,
nouns and noun phrases.
PUZZLES OF EXISTENCE
Janina Kotarbinska
1. The controversy over universals has been revived, the parties being
the same that clashed in Antiquity and The Middle Ages: realism, con-
ceptualism, nominalism. The difference is that universals referred to at
present are neither "forms" nor "essences of things", nor are they ·Pla-
tonic ideas of concrete objects, but abstract entities of a special
kind: sets of individuals, sets of sets of individuals, sets of sets of
sets of individuals, etc., interpreted as objects which are essentially
non-perceivable, extra-temporal and extra-spatial. Just as centuries
ago, the realistic standpoint is that universals exist; the conceptual-
istic standpoint is that they exist in mente, but not extra mente.; and
the nominalistic standpoint, in opposition to the other two trends, is
that the existence of universals is firmly rejected. Conceptualism has
few adherents at present, but the other two movements are very strong.
Their fortunes are, however, variable. Until not so long ago we
might have been inclined to think that the nominalistic trend had final-
ly won the upper hand, and that realism, which was almost universally
declared to be metaphysical and unscientific, would never recover the
ground it had lost. Yet even a cursory knowledge of the current litera-
ture of the subject shows that the situation is otherwise. Recently, it
is nominalism which has been under fire. It is criticized for being at
variance, as far as its basic principles are concerned, with the funda-
mental assumptions. of contemporary mathematics, which is being pursued
in a realistic spirit, on the basis of set theory and its conceptual ap-
paratus. As we know, the important concepts of all mathematical disci-
plines have been defined by reference to the concept of set, and the
theorems accepted in those disciplines include existential theorems
which assume the existence of sets of specified kinds. The situation is
similar in other disciplines, particularly in logical semantics. Matters
have thus cOile to a head. It is obvious that mathematics cannot be the
loser in this unequal conflict •. It is only nominalism which can be the
loser, as it certainly will be unless it succeeds in demonstrating that
its inconformity with mathematics is merely apparent or can be elimin-
ated in a manner that will not require far-reaching compromise. There-
fore, it is not surprising that in these circumstances nominalists con-
centrate their efforts on finding a way out of the present crisis.
53
J. Wolerfsld (ed.), Kotarbirfsld: Logic, Semantics and Ontology, 53--67.
© 1990 Kluwer Academic Publishers.
54 JANINA KOTARBINSKA
lL"~ni .. " .. ki
<1927 - .. specially pp. 166-9 and 186-20:3>, and (1928 - es-
pecially pp. 261-3). Similar ideos or .. t.o b .. ·found in Leonord • (foodmon
(1940); in Goodmon <1961>; ond in Slupecki <i96tO; olso Lusch .. i (1962).
PUZZLES OF EXISTENCE 55
~ot.arbinski <194-9, p.4-:!4-). It. may be helpful t.o explain t.hat. reism is
based on Lesnievski's ont.ology and uses t.he concept.s charact.erist.ic of
t.hat. syst.em. In t.hat, ont,ology t,he only primit.ive t,erm is ,~, ('U'),
int,roduced by t,he axiom V A, B A ~ B '" V X X ~ A --+ X ~ B /\
3 X X ~ A /\ V X Y. X ~ A /\ Y. ~ A --+ X e·M y. . The hrm 'e/.X'
('~') i . defined t.hus: V A e/.X A '" 3 X X ~ A . This obviously
yields t.he t.heorem: V A B A ~ B --+ e/.X A , by which a st.at,ement, of'
t,he t,ype A ~ B has an exist,ent.ial consequence e/.X A vhich applies 1.0
t,he designat,um of' A.
PUZZLES OF EXISTENCE 57
copula 'is' is intended to make sure that the words usable as subjective
complements are genuine names. 6
The interpretation of this reservation is, however, a source of
additional problems. The examples and comments given by Kotarbinski seem
to indicate that the copula 'is' is used in its primary sense if and
only if it stands between names and if, moreover, it satisfies the
formal comditions resulting from Lesniewski's axiom. Yet, as has been
noted by Ajdukiewicz, that combination of criteria appears to be a vi-
cious circle (Ajdukiewicz's original formulation was ~All that system of
definitions looks like a circle, though riot a vicious one.,,)7 and at any
rate it does not provide a method which would make it possible to dis-
tinguish, in concrete cases, expressions which are (genuine) names from
those which are not. This is why, in practice, when the reist has to
decide whether a given expression is, or is not, a (genuine) name, he
makes use of criteria which do not refer to the syntactic properties of
the expressions in question, -but to their semantic properties. In gen-
eral terms, he will class N as a name only (and probably_ always) if it
follows from the very meaning of that N that N is such and such a thing,
or simply that N is a thing. 8 It is perhaps worth mentioning here that
for such an interpretation the primary thesis of ontological reism is no
longer a hypothesis, but an analytic statement which is true on the
strength of the rules of language themselves, since in accordance with
these rules it is only about a thing that we can meaningfully say, given
a primary understanding of the copula 'is', that an object is precisely
that thing, or, that that thing does exist. '
As regards an apparent name we are told in turn that it is "any
word (or phrase) which may meaningfully stand for B in a structure of
the type 'A is B', but only if that structure plays not its prima~y role
but the role of a substitutive and an abbreviation.~ We are further told
that an apparent hame, when inserted in a singular sentence, yields a
meaningless whole if the fundamental, non-metaphorical sense of the cop-
ula 'is' is retained, and that those sentences in which apparent names
occur are substitutive expressions to which the structural criteria of
truth, used for sentences with genuine names, cannot be applied; in
particular, we cannot infer from them sentences of the type 'A exists'.
Finally, it is said that ~ apparent terms (onomatoids) [ ... ) are certain
nouns, adjectives, noun phrases and adjectival phrases which because of
their appearance are taken to be terms, but are not terms; they can be
used meaningfully only in substitutive formulations, and can always be
eliminated in favour of genuine terms alone.~9
.When all this is taken into consideration, we have the possibility
of the following interpretation of the terms in question, which allows
us to avoid the objection of a vicious circle (we offer these formula-
tjons with no claim to maximum precision).
9 Ko t.arbinskl <1966, pp,- 9 and 406); see also (1958a, pp, 74 and 232-3).'
PUZZLES OF EXISTENCE 59
l~ormulation. (3), (4-> and (5) are, f'or the .ak. of' .implicity, given
only for .tatement. of' the type 'A is ~'. How.ver, sinc. in ontology all
.imple .tat.m.nt. or .. def'initionally r.ducibl. to .tat ..... nt. of' that
typ., and .inc. all compound stat •• ent. or. combination. of' .imple on •• ,
the .am .. f'ormulations can easily be mad .. to cover all other .tat ••• nt •.
11"ot .. that, if' this analy.i. i . corr.ct, th.n of th. two .tat.ment.,
(01) and (Cl>, which in the rei.tic th.ory or. con.ider.d to be .ynony-
moue with (02) or (c2> (cf. footnote 5 above>, (02) alon . . . voke. no
doubt. or objection •. In the ca ... of' .(C2>, it. ....m. to diff'.r in content
from (01) and (cl>; in particular, it do.s not . t a t . - not directly any-
vay - that tran.latability into the language of' thing. i . a necessary
condition. It point. rather to the motives which account f'or th. adop-
t.ion of that assu .. ption.
60 JANINA KOTARBUiSKA
"But it is only apparently the case that what one is aiming at here
is to describe the usages and conventions of such and such a linguistic
system, and it is also only apparently the case that what one is aiming
at here is to suggest such and such linguistic conventions. In my inner-
most intentions, when I call the words 'relation', 'property', etc.,
apparent names I mean a certain condition of the truth of sentences in
any language in which, according to the intention of the speaker, those
words do not denote things. Now in any such language any utterance which
formally implies the existence of designata of such words can be true
only in so far as it is substitutive or non-literal in nature - if it
has a secondary interpretation in which such a proof of existence would
not be possible. Consider, for instance, the sentence 'Whiteness is an
attribute of snow'. Formally, the existence of whiteness follows froll
it, on the strength of the formula: 'A est B ~ ex A'. Now we say that
in this sentence the word 'whiteness' is an apparent name. By saying
this [ ••• J we claim that if this sentence is to be true in the speaker's
language, it must be interpreted in a substitutive, secondary sense
(e.g., as a substitutive of the sentence 'Snow is white'), for in its
literal interpretation we could prove the existence of whiteness on the
strength of this sentence. lfl 4-
When we consider this and similar statements we can hardly avoid
the impression that the motive by which the reist has been guided while
formulating his se.antic doctrine has been an endeavour to ensure to
sentences with abstract terms as subjects (briefly: sentences about ab-
stract entities) an interpretation for which those sentences would have
existential consequences incompatible with the main principle of ontolo-
gical eism. This motive seems to have been decisive for the imposition
on the concepts of name, the primary understanding of the copula 'is',
a meaningful sentence, etc., of definitional conditions which have
resulted in difficulties which are already familiar to us.
We may ask, however, whether - from the point of view of the said
intentions - all these conditions are really necessary, in particular,
whether it is necessary to refer to assumptions which are mainly respon-
sible for such untoward consequences.
The procedure adopted in the reistic system is basically as fol-
lows. Assumptions are adopted as a result of which:
(a) a sentence of the type 'A is B', with A standing for an abstract
term, is meaningful only on condition that the copula 'is' occurs in it
in its non-primary, metaphorical sense;
(b) a sentence of the type 'A exists' follows logically from a sentence
of the type 'A is B', with A standing for the same subject in both sen-
tences, if and only if in the latter sentence the copula 'is' is used in
its primary sense.
As a result:
(c) consequences of the type 'A exists', which would affirm the exist-
ence of abstract entities, cannot be deduced from the sentences de-
scribed under (a).
16Th .. last. issu .. is c .. rt.ainly t.h .. most. cont.rov .. rsial of all. It. do .. s s .... m
how .. v .. r, t.hat. making t.h .. s .. mant.ic 1'01 .. of t.h .. t. .. rms 'is' and ' .. xist.s'
d .. p .. nd on t.h .. met.hodological prop .. rt.i .. s of t.he st.at.ement.s in which t.hose
I. .. rms are used is fairly nal.ural. Moreov .. r, I.his is not. specifically
linked 1.0 t.he reisl.ic st.andpoinl.. For insl.once, t.his ossumpl.ion is
adopt.ed by ·Ajdukiewicz (1961>. II. is also in harmony wit.h t.he I.endencies
discussed by Cornap (1960) and wil.h Reichenbach's views as fOl'mulat.ed in
<1936, Sec. 24. et. passilll).
PUZZLES OF EXISTENCE 63
Having made these explanations we can revert to the issue which has
become the starting point of the present considerations. The question
was to satisfy two requirements which appear to be discordant. One of
them calls for agreement with the guiding ideas of ontological reism,
and the other, for avoiding conflicts with statements which are accepted
in science. Now it seems that, contrary to appearances, a solution can
be found, at the price, however, of a number of radical changes in the
present form of the semantic doctrine of the reistic system. Such
changes would result in a far-reaching liberalization of the criteria of
meaningfulness so far adopted.
The basic method would be to extend the concept of name to cover
both those expressions which the reist accepts as genuine names, i.e.,
"names of things", and those which they class as apparent names (onoma-
toids); to make the primary understanding of the terms 'is' and 'exists'
depend, as a necessary condition, on their occurence together with
names, and the fundamental understanding of these terms, on their occur-
rence with names of things; to make the meaningfulness of statements
about abstract entities independent of their translatability into
reistic language.
This can be done in two ways. They differ from one another above
all by the fact that in one of them the same syntactic functions - us-
able both as the subject and as the predicative word in any sentence of
the type 'A is B' - are ascribed to all names, and hence all names are
classed in the same syntactic category: whereas in the other, names are
treated as expressions in the various syntactic categories, since it is
held that every name is usable as the subject or predicate only in such
a sentence of the type 'A is B' as satisfies certain additional condi-
tions, which are different for different kinds of names. The variety of
these conditions accounts for the variety of categories of names.
Without going into details, we shall briefly describe each of these
methods separately.
The multicategorial concept of names was developed by Ajdukiewicz
(by referring to the suggestions of Aristotle and Johnson) for use in
the controversy over the universals. 16 His point, however, was not con-
sistently to carry into effect the requirements of the reistic system,
but, on the contrary, to demonstrate that it is possible consistently to
defend the realistic standpoint against reistic cri ticisRl. This idea
came close to Russell's theory of logical types. In accordance with the
assumptions adopted here we can single out among names an infinite
number of separate syntactic categories which form a certain hierarchy:
the lowest category - that of the zero order - includes names which are
usable only as subjects in sentences of the type 'A is B' (names of in-
dividuals); names of the first order are those which are usable as pre-
dicates in those sentences of the type 'A is B' which have names of in-
dividuals as their subjects; names of the second order are those which
are usable as predicates in those sentences of the type 'A is B' which
have names of the first ord~'h as their subjects; etc. The general
principle is : names of the k order (where k is a positive integer)
are those which are usable as prediq,aJes in those sentences of the type
'A is B' which have names of the k-l order as their subjects. The mul-
ticategoriality of names implies, obviously, the multicategoriality, and
thus the systematic polysemy, of all functors of name (term) arguments.
For instance, the copula 'is' may have the category of a sentence-
forming functor of two name arguments of zero and first order, respect-
ively; of first and second order, respectively; of second and third
order, respectively; etc. To use Ajdukiewicz's notation, its categories
could be symbolized thus:
z z z
no m m m m n3
. " ., etc •
about abstract entities, and in fact our genuine intentions are to speak
about things; (b) the advantages of such an interpretation in philoso-
phical reflections: problems which have been the subject of endless
sterile discussions often vanish immediately as being ill-posed if we
stick to the principle that in our ultimate formulation we should use no
names which are not names of concrete objects; (c) the psychological
naturalness of the reistic guideline: we grasp the intuitive sense of
statements about abstract entities only when we are able to translate
them into statements about concrete objects. 20
The strongest emphasis is usually placed on this last argument.
This stress largely refers to the sometimes glaringly non-intuitive
nature of the concepts of set theory and af the disciplines derived from
it. There is nothing strange in this fact, for it would be difficult to
treat as intuitive such concepts as that of existence, for which in at
least some cases existence depends exclusively on the syntactic proper-
ties of language;21 or that of the distributive interpretation of sets,
since in this interpretation existence is ascribed to sets precisely
because of the somewhat peculiar criteria of applicability of that term;
or the concept of reality, which identifies reality with the totality of
not only all concrete objects, but all abstract objects as well (the
latter being sometimes termed ideal or "non-real"), Le., sets; or the
concept of semantics as that sphere of research which, while intended to
link language with extra-linguistic reality, links it, inter alia, with
that of "non-real reality" which consists of sets of all possible kinds,
including - let us note - sets of non-existent objects. The non-intui-
tive character of set theory is well known, being cited even by those
who cannot be classed as supporters of the reistic approach. 22
It would seem that the arguments to which reists refer are a suf-
ficient justification of their prograa.e. Even if it were to turn out
that this programme can be implemented only on a limited scale, it would
be reasonable to put it into effect wherever possible. To turn something
which is less comprehensible into something more comprehensible and to
eliminate at least some clashes with intuition is always an intellec-
tually prof!table undertaking.
In conclusion we might say that unlike Berkeley, who once recom-
mended that we should think like scholars and talk like the common
people,23 the reists strive to talk like scholars and to think like the
common people, whereas the realists want to think like scholars and to
talk like scholars. The remaining possibility is to think like the
com~n people and to talk like the common people. To complete the
lOCf. t.h .. t.vo pap .. r. on concr .. t.i •• in t.h .. Suppl ..... nt. t.o (1966).
Czesiaw Lejewski
(3) The terms • property', • relation', • event', and any other would-be
names of alleged objects belonging to an ontological category other
than the category of things are pseudo-names or onoDlatoids (see
Kotarbinski, 1930-31 reprinted in 1958a).
is even worse. In the reistic language one can meaningfully and truly
say that
(10) it is not the case that in the reistic language an utterance of the
form 'a is a property' is a true proposition
72 CZEStAW LEJEWSKI
(11) it is not the case that in the reistic language an utterance of the
form '8 is a property' is a meaningful sequence of words
(13) two expressions belong to the same semantical category if and only
if they can be substituted for each other in any syntactically
coherent context without destroying that context's syntactical
cohesion.
Thus, for instance, and here I follow Ajdukiewicz' s idea but not
his actual examples, if in one's language the words 'the lion' and 'spe-
cies' as used in (15) belong to a fundamental semantical category which
is not the same as the category to which the words •Socrates , and 'phil-
osopher' belong then by assuming that whatever is a species is a univer-
sal, one can prove the existence of universals from the truth of (15).
Ajdukiewicz concludes his discussion of reism in his (1935) by
saying that two philosophers, one asserting that only things exist and
the other maintaining that there are universals which are not things,
may appear to be using one and the same language and propounding theses
that are incompatible. In fact, having idealized ordinary discourse in
different ways, they are speaking two different languages, which are not
translatable into one another. The theses they propound need not be
viewed as contradictory. It may be more appropriate to describe them as
not comparable. Having committed ourselves to a possible idealization of
ordinary discourse, warns Ajdukiewicz, we should not think that those
who prefer a different idealization, are wrong even if what they say in
their language appears to contradict what we are prepared to assert in
terms of the language idealized in our way.
Ajdukiewicz, looking back after more than a quarter of a century,
was rather pleased with the results of his polemic with Kotarbinski' s
reism and with Lesniewski's nominalism. In the preface to his (1963) he
claims that under the impact of his criticism ontological reism was
weakened by its originator to become a programme for constrQcting one's
language in a certain way.
Kotarbinski's immediate reaction to Ajdukiewicz's analysis and eva-
luation of reism can be described as a retreat rather than counter-
attack. But it was a retreat not without a few rearguard battles. In the
early thirties Kotarbinski - see (1930), (1931a) and, in particular,
(1935) - tried to hold on to some of the positions taken in Eleaenty,
but on the question of the negative theses of ontological reism he
admitted defeat and withdrew to a second line of defence. In (1949) he
began emphasizing the semantical claims of reism at the expense of its
ontological aspirations. Retreat on this front continued during the
fifties and early sixties. In (1961) ontological claims of reism reached
their lowest, but in the late sixties, namely in (1968), one can detect
attempts at recovering at least some of the lost ground. Kotarbinski's
war against the use of onomatoids in what he calls final pronouncements,
has been relentless although at times he appears to have foresaken the
most convincing reason for waging it. But let us now turn to some
details of the controversy, and let us begin by considering the distinc-
tion between the ontological and the semantical versions of reism.
Ajdukiewicz's insistence on distinguishing between the two versions
is not surprising. His way of thinking was greatly influenced by the
philosophy of the Vienna Circle, and, for him, metaphysics including
ontology in the traditional sense had been eliminated through logical
analysis long before this was officially done by Carnap in his (1932)
paper. Thus, reism as ontology was bound to come to grief at the hands
of Ajdukiewicz. He could only accord qualified approval to a philosophi-
cal doctrine which concerned itself with problems of language. It is in-
teresting to note that Ajdukiewicz does not really try to clarify the
76 CZEStAW LEJEWSKI
It is true that (32) follows from defintions but (32) does not mean
the same as (22), which is to be understood as equisignificant with
(36). And this last thesis could not be proved without reference to the
axioms. Even (21), trivial as it is, does not follow from definitions
alone.
Among further theses of the system we have
Within the system outlined above, equivalences (37) and (38) cannot
be regarded as definitions let alone arbitrary definition~. They had to
be proved with the aid of axioms among other things.
It is easy to see that presuppositions (25) and (26) between them
are inferentially equivalent to (34). If we delete (26) from the axiom
system of reism we get a weaker ontological doctrine, which as Kotarbin-
ski was to learn soon after the publication of the Eleaenty, was in fact
propounded by Brentano in the later years of his life. By retaining (26)
or by using (34) as an axiom instead of the two axioms (25) and (26) we
get a stronger ontological doctrine, the one favoured by Kotarbinski. It
is to this stronger version of reism that Kotarbinski refers as somatic
reism or, simply, as pansomatism. And this is why this last term occurs
in the title of the present paper.
The problems raised by Ajdukiewicz in connection with the positive
ontological thesis of pansomatism are not as complex or difficult as the
problems surrounding the negative ones. According to Ajdukiewicz, to put
his criticism in a nutshell, the reist's utterance
(41) no utterance of the form 'a is a b' with a genuine name, whatever
it lIay designate, in the place of 'a' and the word 'property' in
the place of 'b' is a true proposition
80 CZEStAW LEJEWSKI
full, and" the conclusions he drew from it, seem to be debatable. How-
THE DRAMATIC STAGE IN THE DEVELOPMENT OF KOTARBINSKI'S PANSOMATISM 81
it can be, and has been, treated under certain conditions by both Kotar-
binski and Ajdukiewicz as a genuine name. The conditions amounted to
assuming that (46) was meant to be understood in its literal sense. And
there was a further assumption that the noun • property , belonged to the
semantical category to which nouns which purport to designate things,
belong. This latter assumption can be weakened. For it is enough to as-
sume that the noun 'property' belongs to one of the fundamental semanti-
cal categories.
We seem to have reached the stage when we can re-state the princip-
al theses of reism in the light of the preceding disquisitions. They
take the following form:
the saDIe statement may be taken by sOlie in its literal sense whereas
others may insist that it should be interpreted as a metaphor. In cogni-
zance of this, the reist tries to accomodate both possibilities and he
is particularly concerned with statements which appear to imply the
existence of objects that are not material things, in the case of unica-
tegorial Platonism, or, if one is allowed to make an illicit generaliza-
tion, the existence of properties and other abstract entities in the
case of multicategorial Platonism. Thesis III sets out the reist's views
in this matter.
The three theses embody the most important part of the reist's hy-
pothetical credo. The term 'hypothetical credo' is particularly apposite
to the third thesis. For there is no general proof that 'metaphors' im-
plying the existence of abstract entities can be rephrased in the manner
required by the reist. In fact it is the problem of possible counter-
examples to thesis III that keeps the reist's attention fully occupied.
But I feel I must leave this problem for another occasion.
As I see it, reism is an ontology with semantical ramifications.
Contrary to Ajdukiewicz' s v iew the reist is not committed to any par-
ticular categorially determined language; he has to use, and is at lib-
erty to do so, the language of whoever happens to be his opponent always
provided that it is a categorially determined langul!-ge. Contrary to
Ajdukiewicz's opinion, the positive ontological thesis of reism, i.e.,
thesis I, is not a tautology; it is the denial of a thesis held by
unicategorial Platonism. Contrary to Ajdukiewicz's insistence, the nega-
tive ontological thesis of reisll, i.e., thesis II, consists of proposi-
tions which, in the light of lIulticategorial idealization of ordinary
language, are meaningful and syntactically well constructed. They deny
equally meaningful and syntactically well constructed assertions of
lIulticategorial Platonism.
SEMANTIC REASONS FOR ONTOLOGICAL STATEMENTS:
THE ARGUMENTATION OF A REIST
Marian PrzeI~cki
Arhe doct.rine "a" first. present..,d in Kot.arbinski (1929); it. "a .. t.hen de-
velop"d an comment..,d upon in some of hi .. lat.er papers, ... g. (1949) and
(1968).
85
J. Wolenski (ed.), Kotarbinski: Logic, Semantics and Ontology, 85-96.
© 1990 Kluwer Academic Publishers.
86 MARIAN PRZEt~CKI
amount to. Reducing the whole realm of objects to a few chosen cat-
egories, they reject thereby the existence of any other kinds of enti-
ties. In doing so, they resort to certain facts about our language - to
its characteristic semantic features. Now, what exactly is the rationale
of that kind of argumentation? What does it consist in? We shall try to
analyze the problem, taking as our example the ontological doctrine of
reism.
The latter does not entail the existence of objects other than things;
it assumes the existence of snow and white things, but not that of
whiteness.
The formulation of the reistic thesis quoted is not what might be
called its 'official' version. This corresponds rather to the following
tenet.:
(Ts') Every statement about entities other than things is only meaning-
ful if it is reducible to a statement about things.
thesis (Ts) speaks about may just be identified with the language of our
knowledge - in particular, of science. The reducibility which the thesis
postulates is meant to hold for any language L in which the body of our
knowledge K is expressible.
Now, what kind of justification is being provided for the semantic
thesis of reism so interpreted? In his later reflection on this point
Kotarbinski (1958, p. 434) states that "the principal foundation of
reism is naively intuitive and based on common induction". And he is
quite right in this qualification. The semantic thesis (Ts) has the form
of a general statement, which is being justified by resorting to its
particular instances. The premises of this argumentation are thus sen-
tences which assert the reducibility of particular non-reistic state-
ments to reistic ones. Given the explication of the concept of reducibi-
lity assumed by us above, what any such premise claims may be rendered
as follows: The proper sense of a given statement S, which apparently
refers to objects other than things, is non-literal and identical with
the literal sense of a statement Sr, which is about things only. Besides
premises referring to concrete non-reistic statements, use is made of
premises which refer to the schemata of such statements. It is clear
that those instances of the semantic thesis which the author actually
resorts to do not exhaust the class of all its instances. This is so in
spite of the fact that the premises referring to sentence-schemata deal
with the whole classes of statements, since the schemata actually taken
into account are far froll exhausting all possible types of non-reistic
statements. And so, the semantic thesis of reism is a kind of inductive
generalization, which goes far beyond the scope of its verified
instances.
The methodological character of this thesis depends ultimately upon
the methodological status of its premises. The last point deserves a few
words of explanation. The premises of the semantic thesis may generally
be characterized as statements about synonymy of certain sentences of
language L. They fall thus under the following schema:
(T:) Every statement about entities other than things is only true if it
is reducible to a statement about things, on some admissible inter-
pretation of language L.
sentence not reducible to a reistic one would then, by the very defini-
tion, be qualified as a meaningless expression. A definition like this
might have been motivated by the ontological thesis of reism - provided
the latter has been accepted without, in turn, resorting to the. semantic
thesis (e.g., as a kind of "hypothetical credo", as the author sometimes
calls it). The meaningfulness referred to in this context is of seman-
tic, or philosophical, rather than purely syntactic nature. Those well
formed non-reistic sentences which have no reistic paraphrases will be
called meaningful in the syntactic, but meaningless in the semantic
sensej as such, they will be devoid of any truth value, taken in its
classical sense. According to this view, the abstract (in particular,
mathematical) part of our knowledge will playa purely instrumental, and
not cognitive role. Cognitive value will be attributed only to the empi-
rical basis of science - here identified with the reistic basis. On this
approach, the function of the abstract superstructure of science will
consist in an adequate forllulation and systematization of the reistic
basis - in accord with a contention that in order to say all there is to
be said about things, one has to speak not just about things.
II
The main object of my analysis is not the problem of truth of the seman-
tic thesis, but its bearing on the problem of truth of the ontological
thesis. Assuming thus, for the sake of argument, the truth of the seman-
tic thesis, I shall discuss the attempts to base the truth of the ontol-
ogical thesis upon the truth of the semantic one. As said before, this
is how the ontological thesis is actually justified by the reist, though
on th'] face of it, the situation might well be judged otherwise. Being
about the world, and not the language, the ontological thesis is clailled
to be justified by reference to "our global knowledge", to "the whole of
experience", etc. On a closer inspection, however, the knowledge the
reist has in mind turns out to be a kind of linguistic knowledge, and
the experiential data he resorts to - certain semantic facts. What they
justify directly is the semantic thesisj the ontological thesis is jus-
tified by them only indirectly. The main argumentation of the reist
amounts to a refutation of some counter-arguments advanced against the
ontological thesis. These counter-arguments point to certain statements
which are commonly regarded as true, and which, literaly understood, as-
sume the existence of objects other than things. The reist's argumenta-
tion aims at showing that the proper sense of these statements is non-
literal and reisticj so understood, they do not entail the existence of
entities other than things and, hence, do not contradict the ontological
thesis. But these are exactly the same arguments which have been em-
ployed in justifying the semantic thesis of reism: they claim the redu-
cibility of certain non-reistic statements to reistic ones. And so, what
they validate directly is the semantic thesis.
The problem then arises how to pass froll it to the ontological
thesis. In order to answer this question, we have to examine the logical
relations that obtain between those two statements. The result depends
SEMANTIC REASONS FOR ONTOLOGICAL STATEMENTS 93
(Ro) One must not accept the existence of any entities unless it is
necessary.
94 MARIAN PRZEt~CKI
The concept of necessity which the rule (Ro) referes to may be expli-
cated in various ways. One of thea, directly relevant to the present
case, relates it to our knowledge, or - strictly speaking - to the class
of all adllissible ways in which our knowledge can be expressed. As
stated before, the language of our knowledge is susceptible of different
interpretations. These interpretations differ as to their ontological
cOlllli tments; there lIay be categories of objects which are assumed by
sOlie of them, but not by others. This suggests the following explication
of the concept of necessity:
The assullption that there are objects of a given ontological cat-
egory is necessary if and only if it is entailed by every admis-
sible way of expressing our knowledge.
Ockham's principle forbids us to accept any existential statements which
do not fulfil the above condition. But what is the rationale of Ockham's
rule so interpreted? Why are we not allowed to accept such statements?
There are two kinds of possible answer to that question: we must
not accept the statements because they are false, or - because they are
groundless. The answers correspond to two different kinds of postulates
that lIay be said to underlie Ockham's rule: an ontological and a method-
ological one. The ontological postulate may be rendered as follows:
(R.) The assumption that there are objects of a given ontological cat-
egory is justified if and only if it is entailed by every admis-
sible way of expressing our knowledge.
Now, what is the bearing of the above postulates in the case dis-
cussed in this paper? It is easily' seen that the postulates find
straightforwar~ application to the situation characterized by the seman-
tic thesis (Ts). As stated by it, the reistic way of expressing our
knowledge assumes only the existence of things. On the supposition that
the alternative ways of expressing our knowledge do not make any weaker
assumptions, it is only the assumption that there are things that is en-
tailed by every admissible way of expresing our knowledge and may, on
that account, be said to be necessary. This fact has important conse-
quences on the basis of either of the postulates mentioned. In virtue of
the ontological postulate (Po), it implies what amounts to the ontologi-
cal thesis of reism:
(To) Only things exist.
The postulate (Po) thus provides a necessary bridge which enables us to
pass from the semantic thesis of reism to the ontological one. The prob-
lem is that the bridge seems to rest on rather weak foundations. The
methodological postulate (P.) appears to be much more firmly grounded.
This postulate, however, cannot yield the ontological thesis of reism.
The only conclusion that one is entitled. to draw from the semantic
thesis of reism in virtue of the postulate (P.) is a statement which may
be called the methodological thesis of reism:
(T.) Only things may justifiably be said to exist.
Either of these statements - the ontological as well as the methodologi-
cal thesis of reism - provides a kind of justification for what might be
termed the methodological rule of reism:
(Rr) One must not accept the existence of objects other than things.
Referring to this formula, we can sum up the argument presented above as
follows. What Ockhall's razor, in fact, secures is a step from can to
ought: froll the fact that one can dispense with objects other than
things (as stated by the semantic thesis of reism) it allows us to infer
that one ought to do so (as required by' the methodological rule of
reism). By being feasible, the reistic paraphrase becomes obligatory.
Let me conclude this discussion with some remarks of a more general
nature. The case discussed above is not restricted to the doctrine of
reism. It may be generalized so as to comprise all ontological doctrines
of the reductionistic type. By these I mean doctrines whih may be char-
acterized as the weakest from a given class of doctrines comparable as
to the strength of their existential assumptions. The simplest case of
such doctrines can be schematically described as follows. Let two ontol-
ogical doctrines yield two different ways of expressing our knowledge K:
Kl and K2. Kl and K2 differ as to their ontological commitments: Kl
assumes the existence of objects of category Cl, K2 the existence of
objects of categories Cl and C2. In this-situation, Ockham's rule (Ro)
forbids us to accept the existence of obj~cts other than those of cat-
egory Cl. In view of the postulate (Po), this is to be so because- only
objects of category Cl exist; in view of the postulate (P.), because
only objects of category cl-may justif~ably be said to exist.
96 MARIAN PRZEt~CKI
Tadeusz Pszczolowski
Implications of reism
acting agent. Action belongs to the category of events, and they are
either changes or states of things. If we distinguish events considering
what happens to things,' considering variability or invariability of
certain features in a given interval, such distinction will also refer
to actions. Above the theory of action Kotarbinski constructs a theory
of events which is IBOre general than the theory of action, which, in
turn, is more general than praxiology, i.e. the theory of action evalu-
ated from the point of view of efficiency. .
The theory of events examines cause-effect relationships,.function-
al sequences and simultaneity of events; the theory of action is con-
cerned with particular cases of causal connection, that is causation
when the cause is an agent.
In his approach to praxiology Kotarbinski does not expose the as-
sumptions of ontological reason or of semantical reason, because he does
not state explicitly that genuine names are only those which have real
designations, and all the others are pseudo-names. Froll those we must
ei ther pass to genuine names or we must find out what the relation
between thell and concrete thing is. In the lectures on praxiology these
postulates are fulfilled implicitly. For instance, following Eugeniusz
Geblewicz (see: 1983), Kotarbinski adopts the definition of goal: John's
goal is a certain event (c) means, that John has made some effort in
order to bring about the event (c). John represents a concrete agent.
And in reality there is no goal without the agent. 'Goal' is. a pseudo-
name, and the same refers to 'event'.
Similarly in Plato's Dialogues the notion of courage is reduced -
it does not exist, there are only people whom we call courageous, i.e.
we ascribe certain features to them, for instance strength or character.
As a consequence of reisll one must accept that there is only one
ontological category - the category of things. However, we have several
semantic categories - events, features, relations, and many others; lan-
guage expressions belong to the same linguistic category if their ex-
change (replacement of one expression by another) does not turn a sen-
sible statement into a nonsensical one.
long to the category of relations and are defined precisely and with as
great accuracy as other terms in praxiology. It is assumed that the lan-
guag~ which praxiologists speak and write is already established and
there is no need of introductory explanations of what the terms which
are used mean or denote. It is also a realization of the postulate of
economic management. In connection with this, on Kotarbinski's initiat-
ive, work on the dictionary of praxiology started in 1959, and it lasted
for twenty five years (see: Pszczolowski, 1978).
In accordance with Aristotelian tradition and Leon Petrazycki's
postulate, his principle of adequacy of constructed assertions, of ad-
equacy of theory in regard to this fragment of reality which the asser-
tion or a set of assertions concerns, is taken into consideration in
praxiology.
A theory is adequate when its subject - predicate theses comprise
all classes of objects - and only them - not omitting those objects
which possess the feature designated by the predicate, and do not
ascribe such a feature to the objects which do not possess it. Mach's
thesis referring to so-called Denkoekonomie is not adequate, because
economical forms are demanded not only from thinking and, scientific
statements, but also from many other outcomes of human action. However,
the economy of human routine and repeated actions in the case of cre-
ation is sometimes replaced by the postulate of the greatest effort. The
search or adequacy of the theory of effective action brought about cre-
ation of more and more general theories - the theory of action, the
theory of events, and of theories which, in comparison with praxiology
are less and less general - the theory of struggle, the theory of organ-
ization of groups of people.
The role of the above-mentioned requirement of adequacy of science
is seldom emphasized. Striving to exhaust all the problems and to shape
a whole led the philosophers to propose a philosophical system composed
of various parts. An instance of this striving to overall formulation of
human action can be Kotarbinski' s proposal to set praxiology wi thin
broadly understood ethics to which felicitology (the principles of
achieving happiness), moral deontology (the principles of fair, that is
good, conduct) and praxiology (the efficient principles of fair achiev-
ing happiness) would be subordinated.
Only aesthetics was missing in the full range of aspects in which
action could be examined. In course of time, however, praxiologists'
became interested in it too, noticing the well-marked interdependence:
eff,icient actions are also beautiful and vice versa. This interdepend-
ence is prominent, for instance, in sport.
specially emphasized the fact that some terms can be treated in two
different ways, according to set theory. Such terms besides ethics are,
for instance, 'philosophy', 'science', 'theory'. Philosophy is a whole
consisting of several branches - metaphysics, epistemology, ethics etc.
But metaphysics, epistemology, ethics are philosophy. Science is a
complex of particular sciences. On the other hand, physics is a science,
biology is a science, and every branch of biology - for instance, micro-
biology - is also a science. It is so because we have two set theories.
The first set theory operates on the notion of a class of M - elements,
a set of M - objects, and then a statement that an object belongs to the
class of M - elements is the statement that it is an M - object. If we
have defined what an instrument is, then the assertion that a given
object is an instrument means that this object is included in the class
of instruments. However, we can never say how many instruments there are
in the world, or imagine the set of instruments. We use 'set' in its
distributive sense - because 'set' is an abstract idea and we have to do
only with the elements of this set. In Stanislaw Lesniewski's mereology
we can find another notion of a set - in its collective sense. There
exist objects consisting of M - elements which constitute a whole M,
also called a set. However, it is a different kind of set. A human or-
ganization can serve here as an example. It consists of sections, which
are also organizations of men. Sections are further divided into depart-
ments, divisions, and these are organizations too. The organization con-
sidered here is then a whole consisting of organizations. To express it
quite clearly we must state that both the organization as a whole, and
its component organizations also belong to a set in a distributive
sense, namely to the set which in this case is human organization.
The reason why I use the predicate 'human' requires additional
explanation. The point is that Kotarbinski formulated a definition of
organization using "variables" like a logical formula. Organization is a
whole whose components contribute to prosperity of the whole. It is a
very general definition and if an agent is substituted for a "compo-
nent", we get a human organization. But these components may also be
bees or ants, and then the organization will be a hive or an ant-hill
respecti vely. Particular parts form a radio set or a car - these are
also certain kinds of organization in this general sense. In Poland
Kotarbinski was a precursor of the general theory of systems, which he
called theory of cOlllplex or theory of compound object. The notion of
organization, as proposed by Kotarbinski, is similar to the notion of
system on the understanding that according to reism an organization must
be a thing. It is so because only things (a cause-effect relationship
exists between material objects) contribute to the occurence of some
positively evaluated states of affairs, that is to prosperity. Since
Kotarbinski proposed a very general ·definition of organization, on the
basis of which concrete objects could be included in a set in its dis-
tributive sense, when discussing organization as a set in its collective
sense he should have added that it is human organization that is prefer-
red to. Nota bene, an organism is also embraced in the notion of organ-
ization, and Kotarbinski in his old age used to joke sadly that more and
more of his components had stopped contributing to the prosperity of his
whole.
PHILOSOPHICAL AND METHODOLOGICAL FOUNDATIONS OF PRAXIOLOGY 101
1. effectiveness:
2. ineffectiveness~
3. counter-effectiveness
The notation may provoke stipulations with regard to the meaning of
negation of a goal, negation of an event, negation of an action or with
regard to the use of the same symbol of negation for an event and for
the symbol of sequence, etc. 1 However, this conventional notation shows
that effectiveness is very close to counter-effectiveness, because:
d --+ c v -'c
We have a great class of actions which are either effective or
counter-effective. In this way we get to the theory of probability which
is included in praxiologieal evaluations. Kotarbinski, however, did not
solve the problem whether the notion of counter-effectiveness is suffi-
cient, or there is a need for enlarging the set of counter-efficiency.
Vito F. Sinisi
While the work of the contemporary Polish logicians is fairly well known
by their colleagues in western Europe and America, little is known in
these countries about recent Polish philosophy. This is perhaps due to
the fact that the logicians generally published in universal languages
while the philosophers published their works mostly in Polish. This is
unfortunate because western philosophers, especially logical empiri-
cists, lIight have profited from a knowledge of the work of such men as
Tadeusz Kotarbinski and Kazimierz Ajdukiewicz, who are probably to be
eonsidered as Poland's outstanding contemporary philosophers. Both have
formulated philosophical views influenced by modern logic: Ajdukiewicz
has ereated radical conventionalism, and, Kotarbinski has' developed
rei sm.
Reism, a philosophical view akin to nominalism, countenances only
the existence of things, i.e. objects which are, as Kotarbinski says,
"extended and resistant". A reist Blight be described as one who system-
aticallyattempts to "reduce" all of the categories (e.g. the categories
of events, relations and properties) to the category of things. He con-
strues sentences purporting to be about events, relations or properties
as fa~ons de parle~, surrogates for sentences exclusively about .things.
For exa.ple, 'roundness is a property of this orange', see_inglyabout
the property roundness, is said to be a substitutive expression for
'this orange is round', which, it is held, is about things exclusively.
Expressions apparently referring to objects which are not things are not
genuine nues but "pseudo-n8.llles", e.g. such expressions as: 'concept',
'disposition', 'llotion', '.ethod',· 'operation', 'event', 'length',
'fact', 'velocity', 'number', and 'function'.
This view, like other ,.ore recent nominalistic views, is essential-
ly programmatic, which is to say that the reductive methods proposed are
i.nco.plete; for eXlUllple, there is no known way to reduce all sentences
prima facie about classes to sentences about things. And it is perhaps
101
J. Wole1lsld (ed.), Kotarbwki: Logic, Semantics and Omology, 101-118.
© 1990 Kluwer Academic Publishers.
108 VITO F. SINISI
;!A similar view of common names is found in Ockham, Hobbes, and more
recE>nt.ly in J.H. Woodger (1961, p.196) and in R.M. Mart.in (1968).
KorARBINSKI'S THEORY OF GENUINE NAMES 109
II
(2 ) 'The man who wrote 29 Waverley novels' denotes Sir Walter Scott if
and only if Sir Walter Scott is the man who wrote 29 Waver lay
novels;
(3) 'Harvey' denotes the discoverer of the circulation of the blood if
and only if the discoverer of the circulation of the blood is
Harvey;
(4) 'A pupil of Socrates' denotes Aristippus if and only if Aristippus
is a pupil of Socrates;
3Element.y, pp. 6,7. Translat.ions of' Polish t.ext.s in t.his paper are by
V,P,S.
4- Ibid ., p.7.
110 VITO F. SINISI
It will have been noticed that the phrase "fundamental sense of the
copula 'is'" occurs in two crucial contexts: in the explanation of a
name, and in the explanation of denotation. In the Eleaenty Kotarbinski
did not explicate this so-called fundamental sense of 'is'; however, an
analysis of various passages in Kotarbinski's works reveals that he con-
strues this 'is' as the single primitive term of Lesniewski's ontology,
which is to say that a sentence of the type 'A is B' (with the funda-
mental sense of 'is') is (loosely) equivalent to: there is at least one
A, and there is at most one A, and whatever is A is B. In order to indi-
cate the meaning of the preceding sentence it will be necessary to say
something about Lesniewski's ontology.
Lesniewski created three systems of logic: protothetic (an extended
propositional calculus), ontology (a calculus of names), and Illereology
(a logical theory of the part-whole relation), which is based on the two
previous disciplines. Ontology is derived from protothetic by the intro-
duction of the primitive term 'E', which may be read as 'is'. The term
is introduced by means of a single axiom, and the theorems of ontology
are stated in terms of this primitive and the functors of protothetic
(material implication, conjunction, alternation, equivalence, negation).
'£' always appears as part of expressions of the form IX £ y'. Such
expressions may be called "singular expressions"; if appropriate con-
stants are substituted for the variables, we obtain a singular proposi-
tion. The meaning of 'E' must be gathered from the single axiom of onto-
logy, which may be expressed as follows5 :
'<:Ix'<:ly (xEy '" (3c ct:x 1\ '<:Ic'<:ld «ct:x 1\ dt:x) -+ ct:d) 1\ '<:Ic (ct:x -+ ct:y»)
6Allowances must. be made for t.h .. fact. t.hat. t.h .. r .. are no d .. finit.e or in-
d .. finit. .. art.icle. in Polish, andt.hat. ont.ological synt.ax is based on t.he
synt.ax of Polish. cr. glupecki (1955), p.13. A suit.able inst.an" .. of t.h ..
KOTARBINSKI'S THEORY OF GENUINE NAMES 111
I have pointed out that while Kotarbinski did not explicitly ident-
ify the so-called fundamental sense of 'is' with Lesniewski"s 'c', there
is evidence in his works to justify the belief that sentences of the
type 'A is B' (with the fundamental sense of 'is') are to be construed
as Lesniewski's. singular propositions. 7 The necessary and sufficient
conditions for the truth of sentences of the type'A is B' are identical
with those for the truth of singular propositions; however,while these
conditions coincide, there is an additional factor present in Kotarbin-
ski's usage of 'is' which is absent in Lesniewski's use of 'c', namely,
a semantical rule, so to speak, to the effect that all the variables in
the axiom range over things (in Kotarbinski's sense) exclusively. Les-
niewski's ontology is not restricted in this way so that under some
other interpretation, for example, wherein the variables are restricted
to some set of non-things (again in Kotarbinski's sense), it is possible
that a singular proposition will be true but will not be true in Kotar-
binski's interpretation.
This explanation of what it means to say that a sentence of the
type 'A is B' (with the fundamental sense of 'is') is true will perhaps
make clear the purport of passages (1)-(6) and the significance of the
concepts of name and denoting name. (Henceforth I shall use "sentence(s)
of the type 'A is B'" to lIean "sentence(s) of the type 'A is B' (with
the fundamental sense of 'is')".) Knowing the truth conditions for a
sentence of the type 'A is B' we can easily distinguish between derioting
and non-denoting nalles. An· expression of a language is a name if and
only if it lIay be used "meaningfully" as the predicate in a sentence of
the type 'A·is B'. Denoting names form a subset of the set of names; an
expression of a language denotes an object if and only if it lIay be pre-
dicated truly of that object. According to this analysis, then, such ex-
pressions as • Zeus 'and 'Xanthippe's unmarried husband' are names if,
for example, such sentences as 'Plato is Zeus' and 'Socrates is
Xanthippe's unmarried husband' are meaningful. While 'Zeus' and
'Xanthippe's unmarried husband' are nalles, they are not denoting nalles
because no sentence of the type 'A is B' is true, wherein either func-
tions as a pre~licate. Furtherllore, this analysis entails that· some
(though not all) definite descriptions (or their analogues in Polish).
come to be classed as names. For exallple, 'The author of Waverley', 'The
author of Sla~kenburgius on Noses'. 'The relation of seniority', how-
ever, is not a genuine name but a pseudo-nalle, according to Kotarbinski,
because it cannot be predicated meaningfully of any thing in a sentence
of the type 'A is B'.
Some descriptive adjectives cOile to be classified as names and sOlie
are classified as denoting names. 'Massive' and 'bright' are, for
example, denoting names because sentences such as 'Gaurisanker is mas-
schema 'XCy' in Polish would b .. 'Jan III Sobieski j .. st. wyba.viciel ...
Wiednia', which would be t.ranslat.ed· int.o English (lit.erally> aa 'Jan So-
bieaki III ia lib.rat.or· of' Vi.nna'. It. might. b.helpf'ul t.o ke.p t.hia in
IDind in diacu~.ion of' t.h. schemat.a 'xC,' and ' . ia B'.
7F.'or example, ..... Element.y, pp. 7-8, 18, a. veli as Kot.arbin.ki'. more
r .. cent. <1955>, pp. 115, 117.
112 VITO F. SINISI
sive' and 'The Evening Star is bright' are true. While many substantives
are recognised as names, many are not. For example, 'wine', 'woman' and
'song' are genuine denoting names leach having multiple denotation) but
'fact', 'fiction' and 'forecast' are pseudo-names.
Let us now consider denoting names more closely. Having explicated
the so-called fundamental sense of 'is', the right side of the bicondi-
tional in (la) has perhaps been made clear. In (2) the left side of the
biconditional is equivalent to the right, and since the right side of
(2) is true, it follows that the left side is also true. Note that in
(2) the sentence of the type 'A is B' has a proper name as its subject
and a definite description as its predicate. In (3) the roles are re-
versed. These two examples indicate that in the analysis of 'A is B' the
uniqueness condition does not pertain to the definite description when
it is used as the predicate in such a sentence; it pertains to it only
when it is used as the subject. The displayed passage (4) differs from
(2) and (3) because the right side of its biconditional contains a
proper name as the subject and a common name as the predicate. (5) and
(6) present cases of names which do not denote any thing. 'The winged
horse captured by Bellerophon' does not denote any thing; specifically,
it does not denote Pegasus. This is so because no sentence of the type
'A is B' is true when 'The winged horse captured by Bellerophon' is sub-
stituted for 'B', and this is so because no individual thing is the
winged horse captured by Bellerophon. The falsity of the right side of
(5) is compounded because a vacuous name, 'Pegasus', has been substi-
tuted for 'A'. Neither 'Pegasus' nor 'The winged horse captured by
Bellerophon' is a name of a thing, though each is a genuine name.
The displayed passage (6) is perhaps the most interesting of the
examples. Previous discussions should make i t clear why the left side of
(6) is false. If the right side of (6) were true, it would follow that
'Zeus exists' would be true. This is made explicit by Kotarbinski' s
statement, late in the Elellenty,8 that "'M exists' for us", i.e. reists,
"is equivalent to 'something is M''', which may be expressed as "M exists
= 3y (y is M) Df.", wherein 'is' functions in the fundamental sense. 9 If
'Zeus is Zeus' were true, it would follow that something is Zeus, and at
most one thing is Zeus, and whatever is Zeus is Zeus. Consequently, 'If
Zeus is Zeus, then something is Zeus'. In virtue of the definition above
the consequent of 'If Zeus is Zeus, then something is Zeus' is equival-
ent to 'Zeus exists', and substituting 'Zeus exists' for this consequent
we obtain 'If Zeus is Zeus, then Zeus exists'. Since it is false that
Zeus exists, it follows that it is false that Zeus is Zeus. 10
III
IV
Denotation with respect to singular and vacuous names does not seem to
present any serious problems, while denotation with respect to common
names requires further explication.
It is sometimes said that predicate expressions denote (or have as
extensions) sets or classes. Does a common name, which denotes severally
the things to which it truly applies, ever denote the set or class of
its designata? If so, and if a set or class is construed as an "ab-
stract" entity, then reism is an ontology which seems to countenance
such entities. This, of course, is to be avoided if reism is to be con-
sistent; reism commits itself only to the existence of "extended" and
"resistant" objects. An analysis of the terms "set" and "class" allows
Kotarbinski to say that some common names not only have multiple denota-
tion but denote the class of their own designata as well. This does not,
however, commit him to belief in the existence of classes as abstract
entities. I shall now try to explain this apparent paradox.
In the Ele.enty (pp. 13-15) Kotarbinski distinguishes the distribu-
tive meaning of "set" from its collective meaning. In the first sense
"set" is said to be a pseudo-name, while in the second it is said to be
a genuine name. Consider "X is an element of the set of M's". In this
expression 'set' is used in its so-called distributive sense, and so
used is a pseudo-name. This expression is said to be equivalent to 'X is
an M'. For example, 'Mars is an element of the set of solar planets' is
equivalent to 'Mars is a solar planet'. In spite of the fact that 'Mars
is an element of the set of solar planets' is true, no object is the set
of solar planets. In the distributive sense 'the set of solar planets'
does not denote any thing because to use the expression 'the set of so-
lar planets' (in the distributive sense of "set") as a predicate in some
sentence of the type 'A is B' would be an absurdity, according to Kotar-
binski; something on the order of 'John is hold on!,13
I think that at least part of what Kotarbinski means may be restat-
ed as follows. "Set", in the distributive sense, can be eliminated from
a sentence without changing the meaning of the sentence. For example,
'Mars is an element of the set of solar planets' would mean the same
thing as 'Mars is a solar planet'. Thus in the distributive sense of
"set" the expression 'is an element of the set of' would be "reducible"
to 'is' (in the so-called fundamental sense).14 'Mars is a solar planet'
would be an instance of the schema 'A is B'; 'Mars' being a singular
name and 'a solar planet' a common name.
In the collective sense of "set" the term is said to be a genuine
denoting name. In this sense "the set of M's" denotes a certain whole
composed of M's - just as a chain consists of links, or a swarm of bees
13Loc . cit,.
14cf . Soboei';ski <1964-6>, p.2. I "ish to thank Prof •• sor Soboei';ski for
an offprint of this artiel •.
KOTARBINSKI'S THEORY OF GENUINE NAMES 115
Alr-__~C~I____~D+I____-=E~I----~B~I
Figure 1
Axiom IV. If some object is a, then some object is the class of objects
a.
Definition III. P is the set of objects 0 if and only if the following
conditions are satisfied:
a) P is an object,
b) for all Q - if Q is an ingredient of the object P, then
some ingredient of the object Q is an ingredient of some
a, which is an ingredient of the object P.
This may be illustrated by the following examples (taken again from Les-
niewski). The segment AC of Figure 1 is a set of parts of the segment AB
because both conditions of Definition III are satisfied. The segment AC
is not a set of ingredients of the segment DB because while the first
condition of the definition is satisfied, the second is not; the segment
AC is an ingredient of the segment AC, but it is false that some ingred-
ient of the segment AC is an ingredient of some ingredient' of the
segment DB, which is an ingredient of the segment AC.
Perhaps these axioms, definitions and examples will help to expli-
cate Kotarbinski's use of "set" and "class". In the collective sense of
"set", then, the term is a genuine name. In this sense a set of M's is
an object P. In the distributive sense a set of M's is not an object.
According to Kotarbinski, 'x is an object' implies 'x is a thing', i.e.
• x is extended and resistant'. This stipulation is not to be found in
the axioms and definitions of mereology, and hence. Kotarbinski's use of
"set" may be thought of as a specialization of Lesniewski's use of the
term.
Let us now consider the term "class". According to Kotarbinski,
"the class of M's" is equivalent to "the set of all M's". In the dis-
tributive sense of "class" the term is a pseudo-name, and it behaves in
the same manner as "set" (in its distributive sense). For example, it is
KOTARBINSKI'S THEORY OF GENUINE NAMES 117
17Element.y, p.16. Perhaps it. should be point.ed out. t.hat. i1' a part.icular
tree is t.he set. 01' it.s cells and also t.he set. 01' it.s,part.icles, t.hese
sets do not. have t.he same number 01' element.s, and hence in t.he set. -
theoret.ic usage 01' "set." are di1'1'erent..
118 VITO F. SINISI
Vito F. Sinisi
119
I. Woletfski (ed.), Kotarbitfski: Logic. Semantics and Ontology, 119-135.
© 1990 Kluwer Academic Publishers.
120 VITO F. SINISI
(2') If some object is older that another object, and the latter
object is older than some other object, then the first object
is older than the third object.
In the abbrev iating sentence (2) , is' does not stand between genuine
names although 'seniority' and 'relation' are substantives. In (2) we
have a sentence of the form 'A is B' "but in a secondary use, wherein
'is' sounds like or look~ like the copula in the sentence" (1) " but it
fulfils another role, it means something different, and the set of words
in the place of 'A' and 'B 'are not n8.lles3 ". These expressions are
called "pseudo-names". Sentence (3) is described in the same way: it is
said to be a substituting-abbreviation for
(3') Whatever is a whale is a mammal.
It may be recalled that ontologically reism countenances only the
existence of things. Semantically the reist countenances among genuine
names only singular, common, or vacuous names. Expressions apparently
(4) The movement of the second-hand was faster than the movement
of the minute-hand
is true, and since (4) is about events it is likely that some object is
an event. The following also seems to be a true sentence about an event:
Since (5) is true one might be induced to believe that there are events,
e.g., the eruption of Vesuvius. Kotarbinski insists that (4) and (5) are
only prima- facie about events; they are not about events because they
are substitutive expressions for
In both (4') and (5') ther.e are no names of events; in (4') and (5')
each nalle denotes a thing. Acording to Kotarbinski, "No object is an
event because only spuriously (by construing a sentence in a substitu-
4 Ibid , p.61.
sElemeni.v, p.61.
122 VITO F. SINISI
(6) The relation higher than holds between Garluch and Giewont.
(7) The relation older than joins John with his son.
It is said that in (6') and (7') there are no names of relations; all of
the names in (6') and (7') denote things. The expressions 'older than
•••• and 'higher than ••• ' are not names of relations; these expressions
are, respectively, names of John and Garluch because the first can be
predicated of John in a true singular sentence of the form 'A is B'
(with the fundamental sense of 'is') 'and the second can be predicated of
Garluch in a similar sentence.
The claim. again, is that the category of relations is "reducible"
to the category of things because sentences which seem-to be about rela-
tions are mere substitutive-abbreviations for sentences which are about
things exclusively.
The category of properties is the last to be "reduced" to the cat-
egory.of things.
Consider the following sentences:
6 Ibid , pp.62-63.
KOTARBINSKI'S THEORY OF PSEUDO-NAMES 123
II
's;:' expresses 'are' and 'is' in contexts like 'Cats are animals' and
'A whale is a mammal'. The copu1a in (3) could be construed as
expressing class inclusion, and its terms interpreted as referring to
classes; we could say that (3) and (3') are, respectively, instances of
the definiendum and definiens of the definition schema (13), and that
(in an appropriate language) (3) and (3') are definitionally equivalent.
In such cases one usually speaks of the definiendum as being an abbrevi-
ation of the definiens, and one also says that the definiendum may be
anywhere substituted for the definiens. It would not be inappropriate,
then, to speak of the definiendum of the definition as being (in an
appropriate language) a substituting-abbreviation of the definiens.
Let us consider (2) and (2'). Suppose that in some language L
"transitive relation" is defined in the familiar way:
(14) R is a transitive relation.:!.!. 'Vx 'Vy 'Vz (xRy A yRz .... xRz).
KOTARBINSKI'S THEORY OF PSEUDO-NAMES 125
If for 'R' we take 'older than', then we could say that in L (2) is
definitionally equivalent to (2'). Hence, (2) could be thought of as an
instance of the definiendum and (2') as an instance of the definiens of
(14); and it would be appropriate to say that in L the instances of the
definiendum of (14) are substituting-abbreviations of the instances of
the definiendum of (14).
I think that the relation between (2) and (2'), and (3) and (3') is
similar (though not identical) to that expressed in (14) and (13), re-
spectively, I shall try to show that a SAS is an instance of a definien-
dum of a definition schema in a reistic language (hereafter abbreviated.
as 'RL').
Consider (3). In some contexts 'whale' is a genuine name, e.g., in
(16) XaY~Vx(xisX-+xisY).
This is offered as a contextual definition of 'a', which may be read as
'is' or 'are'. The expression 'a' is defined in terms of what might be
called the "primitive" terms of RL, Le., the definiens contains vari-
ables ranging over things, occurences of the undamental sense of 'is',
and the usual logical connectives.
This explication, however, goes only part way in explaining the re-
lation between (2) and (2') since neither can be construed as an in-
stance of the definiendum or definiens of (16). If the relation between
(2) and (2') is that of definitional equivalence, it would seem that
what is being defined is the expression 'is a transitive relation'
rather than a use of the expression 'is'. If so, then (2) cannot be con-
strued as an instance of the definiendum of (16), as' (3) was, since (16)
offers a definition of the ,binary predicate 'a'. It would seem that a
different definitional schema is necessary to account for the defini-
tional equivalence of (2) and (2').
Consider the singular sentence
(23') and (24') will not do because Kotarbinski stipulates that "every'
object is a thing, whatever is is a thing,,8, and a thing is extended and
resistant. No object (in Kotarbinski's sense) is divisible by another
object, and no object is relatively prime to another object; divisibili-
8EleDlent.y, p.67.
128 VITO F. SINISI
Kotarbinski has said that (3) is equivalent to and is a SAS of the type
'A is B' for (3'). Both (25) and (3) are of the form 'A is B', and each
has a subject and predicate which are "class-names", and both (25) and
(3), according to Kotarbinski's criterion, are SAS's. (25) differs from
(3) in that 'is' in (3) takes as its arguments expressions which could
(in a different context) function as genuine common names, while 'is' in
(25) does not. (25) seems to affirm a relation of membership holding
between two classes; the class of sperm whales is a member of the class
Species of whale. The class Species of whale has as its members classes
of whales (not individual whales), e.g., the class of sperm whales, the
class of right whales, the class of hump-back whales, etc.
Kotarbinski has said that to say 'x is an element of the class of
M's' is equivalent to saying 'x is an M,.9 He nowhere places any
restriction on the type of name which may be substituted for 'x',
although in examples of 'x is an element of the class of M's' he uses
what are ordinarily called "individual" names as substituends, e.g.,
'Mars'. When we "reduce" 'the sperm whale is an element of the class
Species of whale' we obtain (25). Now, (25) contains as its subject term
a classname (or pseudo-name), and (25) is of the form 'A is B'. Since
sentences containing pseudo-names are SAS's, then (25) wo~ld seem to be
a SAS, and it calls for further reduction. We cannot "reduce" it in the
same manner as we "reduced" 'a whale is a mammal', for if we did, we
would obtain the evident falsehood
(26) Whatever is a sperm whale is a species of whale.
The problem, then, would be to find a meaningful sentence K which is
equivalent to (25), and which is about things.
In the preceding I have attempted to clarify Kotarbinski's concept
of SAS of the type 'A is B', and on the basis of this clarification I
have attempted to explain his concept of pseudo-name. I have also
attempted to show that there are a number of unresolved problems arising
from his concept of SAS of the type 'A is B'. I shall now consider other
types of SAS and pseudo-name which result from Kotarbinski's discussion
of the reduction of the ontological categories to the category of
things.
9Elementy, p. 14..
KOTARBINSKI'S THEORY OF PSEUDO-NAMES 129
III
Kotarbinski 's general claim is that there is but one ontological cat-
egory, and that the putative categories of events, relations, and prop-
erties are "reducible" to the category of things. The basis for this
claim is that certain sentences, viz. (4), (5), (6), (7), (8)-(12), are
SAS's, surrogates for sentences which are about things exclusively.
A more general claim would be that not only these but all sentences like
these are surrogates for sentences about things. And a still more gen-
eral claim would be that liny meaningful sentence which purports to be
about events, relations, or properties is "reducible" to a sentence
about things exclusively.
Although Kotarbinski does not give or even suggest a general method
for translating all sentences purporting to be about events, relations,
and properties into sentences about things, it would seem highly desir-
able to prov ide a systematic method whereby any meaningful sentence
about these bogus entities can be shown to be "reducible" to a sentence
about things exclusively. I shall try to show that such a procedure (at
least for some sentences about relations and properties) -is forthcoming
by showing first that it is possible to formalize (using some elementary
devices of logic) a small fragment of a reistic language. I shall formu-
late a part of this language, and show that it gives the reist a general
procedure for systematically "reducing" at least certain types of sen-
tences, SAS's, about relations, and certain types of sentences, SAS's,
about properties to sentences about things. And I shall try to show that
this allows the reist to say that in certain cases the relation of being
a SAS is the relation 9f definitional equivalence within RL. I shall
suggest, for example, that in RL (6) and (6') are instances, respective-
ly, of the definiendum and definiens of a contextual definition of the
expression 'the relation higher than', and that (7) and (7') are in-
stances, respectively, of the definiendum and definiens of a contextual
definition of 'the relation older than'.
Just as Russell in his theory of definite descriptions showed how
one might use a seeming name without supposing the existence of any en-
tjty named, so too one can use, in an appropriate language, 'the rela-
tion higher than' and 'the relation older than' without assuming the ex-
istence of two relations. This will be possible provided that such ex-
pressions can be so defined in context that all reference to these rela-
tions are eliminated in the definiens.
In the following discussion "relation" is construed extensionally;
two relations are identical if they have the same extension.
We begin by assuming that the schema 'Aa,b' represents sentences in
RL which consist of a dyadic predicate A (e.g., 'is higher than', 'is
older than') and two singular names (in Kotarbinski's sense). (6') and
(7') would be instances of this schema. It may be said that a dyadic
predicate is true of ordered pairs of objects, and that the ordered
pairs of which a dyadic predicate A is true constitute a relation, the
relation of anything x to anything y such that Ax,y. With each dyadic
130 VITO F. SINISI
In light of (27) let us consider the relation between (6) and (6'),
and (7) and (7'). Be making appropriate substitutions into (27) we
obtain a contextual definition in RL, of 'the relation higher than'.
Thus, to say that (6) is a SAS for (6') would be to say that (6) is, in
RL, definitionally equivalent to .(6'): and the same may be said of (7)
and (7'), Le., by making appropriate substitutions we obtain a contex-
tual definition of 'the relation older than'. As Russell might say, we
have defined a use of 'the relation older than' and a use of 'the rela-
tion higher than'. Of course, there may be other uses of relation sym-
bols in RL: for these additional uses further definitions are requisite.
If this analysis is correct, then to say that 'the relation higher
than' and 'the relation older than' are pseudo-names is to say that in
RL these expressions are instances of '(Ax,y)(Ax,y)' appearing in (27).
Generalizing, we can say that in RL any instance of '(Ax,y)(Ax,y)'
appearing in (27) will be a pseudo-name, e.g., 'the relation brother
of', 'the relation larger than'. (27) shows how, in at least one ,con-
text, Kotarbinski may meaningfully use what appears to be the name of a
relation without assuming the existence of that relation. Expressions
apparentty referring to relations are pseudo-names, and are eliminable
by contextual definition.
If we examine (6) and (7), we note that each asserts that a binary
relation holds between a -couple of things (in Kotarbinski' s sense) • .It
should be pointed out that in his discussion of the reduction of the
category of relations to the category of things Kotarbinski does not
consider sentences which assert that a binary relation holds between a
couple of non-things, e.g., 'the relation greater than holds between 9
and 7'. The relations he considers in (6) and (7) are such that their
respective domains and counter-domains are sets of things. It would seem
that he means to confine his discussion to relations whose relata are
things. The definition schema (27) was suggested with this restriction
in mind; it is based on the assumption that the dyadic predicate takes
two singular naaes as arguments. A singular name, it will be recalled,
denotes one and only one thing. If this restriction is lifted, if, for
example, 'a' and 'b' in (27) took vacuous names as substituends as well,
then it would be possible to say that there is a sentence x which is a
SAS for a sentencey, and y is not a sentence about things. It would be
possible to say, for example, that 'the relation taller thBll holds
between Zeus and Apollo' is a SAS for 'Zeus is taller than Apollo".
Let us note that the definition schema (27) is idle for such con-
texts as
KOTARBINSKI'S THEORY OF PSEUDO-NAMES 131
IV
U Quin & (1939). All quot..d pa .... ag ... in t.his paragraph are from Quine's
aJ-t.icle.
KOTARBINSKI'S THEORY OF PSEUDO-NAMES 135
construed in such a way that "its variables admit only concrete objects,
individuals, as values - hence only proper names of concrete objects as
substituends. Abstract terms will retain the status of syncategorematic
expressions, designating nothing, so long as no corresponding variables
are used." Yet, a nominalist, in Quine's sense, is free to use variables
taking abstract terms as substituends provided "that he can explain this
usage away as a mere manner of speaking". Quantification of this kind
"can often be introduced by a contextual definition - a mere convention
of notational abbreviation". If the nominalist can devise contextual
defini tions explaining quantification with respect to entities of an
abstract kind as a mere abridged manner of speaking,
he becomes justified in speaking as i f there were such entities with-
out really forsaking his nominalism. The entities remain fictions for
him; his reference to such entities remains a mere manner of speak-
ing, in the sense that he can expand this sort of quantification at
will into an official idiom which uses only variables having proper
names of individuals as substituends. But if the nominalist can not
supply the relevant contextual definitions, then his nominalism for-
bids his use of variables having abstract entities as values. He will
perhaps still plead that his apparent abstract entities are merely
convenient fictions; but his plea is no more than an incantation, a
crossing of the fingers, so long as the required contextual defini-
tions are not forthcoming.
Barry Smith
iii 1. Introduction
Along with almost all the more important Polish philosophers of the
twentieth century, Kotarbinski, too, was a student of Kazimierz
Twardowski, and it is Twardowski who is more than anyone else
responsible for the rigorous thinking and simplicity of expression that
is so characteristic of Kotarbinski' s work. Twardowski was of course
himself a member of the Brentanist movement, and the influence of
Brentanism on Kotarbinski's writings reveals itself clearly in the fact
that the ontological theories which Kotarbinski felt called upon to
attack in his writings were in many cases just those theories defended
either by Twardowski or by other thinkers within the Brentano tradition.
Lesniewski, too, inheri ted through Twardowski an interest in Brentano
and his school, and as a young man he had conceived the project of
translating into Polish the Investigations on General Grammar and
Philosophy of Language of Anton Marty, one of Brentano's most intimate
disciples. Lesniewski, as he himself expressed it, grew up "tuned to
general grammar and logico-semantic problems la Edmund Husserl and the a
representatives of the so-called Austrian School" (1927/31, p.9).
The influence of Brentanism on Polish analytic philosophers such as
Kotarbinski and Lesniewski has, however, been largely over looked
principally as a result of the fact that the writings of the Polish
analytic school have been perceived almost exclusively against the
background of Viennese positivism or of Anglo-Saxon analytic philosophY.
It is hoped that the present paper, following in the footsteps of Jan
Wolenski's recent work 2 , might do something to help rectify this
imbalance. The paper will consist of a critical survey of Kotarbinski's
development from his early nominalism and 'pansomatistic reism' to the
later doctrine of • temporal phases '. It will be shown that the surface
clarity and simplicity of Kotarbinski's writings mask a number of
:lSee WOlenski (1989), and compare also Schnelle (1982) and the papers by
Schnelle in Cohen and Schnelle (1986).
137
I. Wolelfski (ed.). Kotarbmski: Logic. Semantics and Ontology. 137-183.
@ 1990 Kluwer Academic Publishers.
138 BARRY SMITH
8 See L .. sni .. wski (1914-, 1927/31). It may b.. also that L .. sniewski's
criticism of prop .. rti .. s in his (1913) help .. d to provok .. Kotarbinski's
initiol nominalism.
empty set, merely because they prove expedient for certain purposes 10
Lesniewski's own strictures in this respect are directed in particular
against axiomatic theories of sets such as were developed by Zermelo.
These do not merely lack the sort of naturalness that would dispose one
to accept them; they lack also that intrinsic intelligibility which
would make their meaning clear, so that Lesniewski can in all honesty
assert that he does not understand what is meant by 'set' as this term
is supposed to be 'implicitly defined' by theories like Zermelo's11.
Lesniewski, himself, in contrast, starts not from 'inventions' or
from axioms or hypotheses selected for pragmatic reasons, but from what
he calls intuitions, commonly accepted and meaningful to all, relating
to such concepts as whole, part, totality, object, identity, and so on 12 •
The language of Lesniewski's theories is therefore an extrapolation of
natural language, a making precise of what, in natural language, is left
inart.iculate or indistinct. His work forms part of that strand in the
development of logic, represented also by Frege and by the early
Russell, which sees logic as a descriptive enterprise, part and parcel
of the attempt to produce formal theories adequate to and true of the
actual world 3. Hence he is mistrustful, too, of the model-theoretic
semantics that has been built up on an abstract set-theoretical basis,
and he is opposed also to the work of those formalist logicians who
embrace an essentially abstract-algebraic approach to logic, or see
logic as having to deal essentially with un interpreted formal systems 14 •
St.age 4 in the development of reism, which consists in the
rejection by Kotarbinski of mental images and other 'immanent contents',
again reflects the influence of Twardowski. The status of mental
entities was an issue of particular importance to Kotarbinski, since it
marked one of the very few areas of disagreement between himself and
Lesniewski. For Lesniewski admitted contents and images into his
ontology, remaining in this respect faithful to the heritage of Brentano
and Twardowski 15. Thus Lesniewski, in this sense and perhaps also in
others, is not a reist. Since, however, like Brentano, Marty and
Twardowski, he held that contents are concrete items, tied, in effect,
14In t.erest.ingly, Tar ski, at. least. in his early years, up t.o and including
his paper on '''The Semant.ic Concept.ion of Trut.h",. agreed wit.h Lesnievski
in t.his (see esp. pp.34.2f. of Tarski 194.4.). Tar ski, be it. not.ed, was
never a formalist.: Tar ski and Lesniewski part.ed company rat.her because
Tar ski came gradually t.o accept. t.he Use of set. t.heory and infinit.ist.ic
met.hods in his work.
binski is on the one hand keen to insist that the term 'body', as he
understands it, embraces not only planets, rocks, etc., but also objects
investigated by physics "such as electrons, protons, magnetic fields"
(1966, p.331). On the other hand, however, he stresses that it excludes
for example "immanent coloured patches". Consider, however, the example
of a glass cube that is uniformly red in colour. Is the transcendent
redness of this cube (an individual three-dimensionally extended moment
of colour), a body, on Kotarbinski's view? Certainly it is bulky and
lasting and, perforce, such as to offer resistance. Kotarbinski, it
would seem, was able to ignore such cases in framing his account of
'body' only because his attentions were concentrated on instances of
surface colour, entities which fall short of three-dimensionality and
can be excluded on this count. In order to rule out examples like the
cube of colour, Kotarbinski would have to add the condition that a body
is that which exists (is extended and inert) in its own right (has need
of no other thing in order to eXist)18. As we shall see, a condition of
this sort is very much in the spirit of Aristotle. Certainly such a
condition would capture the sense in which the given example gives
grounds for suspicion - that the cube of colour exists merely as a
dependent moment of the cube of glass, and enjoys no separate existence.
Yet how are we to formulate the condition in question in such a way that
it would not rule out other examples which we would wish to count as
bona fida bodies? Does a human being, for example, exist 'in his own
right', given that he has need, for example, of nourishment, and
processes of breathing and metabolising (to say nothing of parents), in
order to exist? How, moreover, are we to make precise the sense of
'other' in 'has need of no other thing'? Simple non-identity will not
do, since everything may in this sense stand in need of its own proper
parts in order to exist. On the other hand spatiotemporal discreteness
or disjointness will not serve, either, since the cube and its colour
would seem to coincide in space and time. All that can be said here is
that considerations such as this have exercised Kotarbinski (and
Lesniewski, et. a1.) too little, so that the project of a somatist
ontology still leaves much to be desired in terms of a clear statement
of what is meant by 'body'.
18some t.hing .imilar would be r.quir.d t.o exclude from t.he r.alm of t.hing.
01.0 cert.ain .ort.. of' event.s. Con.ider, f'o.. example, a .. ot.at.ion of a
met.al .phe .... Thi . . . ot.at.ion i . ext.end.d in .pace and t.ime and, again, it.
is such as t.o off' ... re.ist.ance. Not.e t.hat. what. i . est.ablished by t.he
a .. gument. in t.he t..xt. is me .. ely t.hat. t.he canonical .. ei.t.ic not.ion of
~ is indet. ... minat.. in it.s applicat.ion. The a .. gument. is not. designed
t.o .how t.hat. on. could not. deal .at.isfact.orily wit.h colou .. s (0 .. t.hree-
dimensional shapes or mas.es of sound or heat.) wit.hin t.he Le8niew.kian
framewo .. k. As Lejewski has .uggest.ed, Ju.t. a. Chronology and St.e .. eology
(t.heories of' t.ime and space) can in p .. inciple be obt.ained f'rom Me .. eology
by t.he addit.ion of cert.ain ext.ra-Iogical con.t.ant.s, .0 it. would be
po.sibl. t.o conceive a di.cipline of" Colourology, obt.ain.d by adding
const.ant. t..rm. such 'r.d', 'blu.', et.c., and a relat.ional pr.dicat.e .uch
ae 4is t.he 8ame colour as".
144 BARRY SMITH
210n t.h ...... rminology of' 'making t.ru .. ' s .... Mulligan, Simons and Smit.h
(19S~>. The t.erminology has a numb .. r of' advant.ag .. s over t.h .. mol''' usual
t.alk of' correspond .. nce. It. is dis .. mbarrassed, f'irst. of' all, of' all
connot.at.ions of' 'copying'. It. does not. s.uggest. t.ha't. t.h .. r, .. lat.ion bet.ween
a sent.ence and t.hat. in virt.ue of which it. is t.ru .. would be a sYmmet.rical
relat.ion. And it. can cop .. wit.h t.h .. fact. t.hat. t.h .. re may b .. mar .. t.han on ..
ent.it.y which makes or h .. lps t.o mak .. a given sen"enc .. t.ru ... Thus, in t.he
simplest. possible case, 'I have a headache' may be made t.rue by my
present. headache ('f'rom .. he beginning t.o t.he end of' it.s exis"ence'), or
by ~ny phase of' t.hi~ h .. adache ov .. rlapping wit.h my pr .. sent. ut.t.erance, or
by relevant. st.a"es of' nervous "issue upon which my headache supervenes.
146 BARRY SMITH
is not therefore 'a "static conglomerate" ("a mere sum") of "rigid and
changeless solids'" j it is a "fabric composed of changing things". in a
new and extended sense of 'thing,25. No explicit criterion is provided,
however, as to what is 'thing' and what 'event' on this more liberal
dispensation. so that one does not know, for example, whether quarks,
neutrinos, or flashes of lightning are to be admitted as (short-lived)
things or rejected as events.
Moreover, even where we are dealing with non-scientific sentences
of the everyday world, the reist's literal renderings are not in every
case so easy to come by. What, for example, is to count as a 'literal'
rendering of a judgment like: 'John's jump impressed the spectators'?
Perhaps: 'John jumped and impressed the spectators'. Yet it is far from
clear that this rendering is even roughly adequate. John's jump, after
all, may have impressed the spectators, but not John himself. Or John
may have jumped, and impressed the spectators, without it being the case
that it was his jump by which they were impressed~.
Kotarbinski's problem here results from the fact that there is as
it were a selectivity of intentional verbs like 'see' or 'think about'
or 'be impressed by'. It seems that such verbs may relate their subjects
to entities such as events, processes, images, contents, meanings, sur-
faces, boundaries, states of affairs, absences, and so on, in ways not
accountable for exclusively in terms of any mere di~ectedness to things.
Such selectivity is characteristic especially of memory, which may as it
were conceal from our present consciousness the things which serve as
supports for events or circumstances remembered. Thus Harry may remember
the intonation of Mary's voice, yet he may have forgotten both Mary
herself and the voice that had this certain quite specific intonation 27 •
Further problems arise for an approach of the sort sketched by
Kotarbinski when we cosider sentences apparently involving quantifica-
tion over events or types of event (John" danced the same jig twice), or
when we consider relational or comparative sentences like Mary's blush
was redder than Susan's, The beginning of John's jump was more elegant
than the end of Jack's, and so on28.
Now there is, cert.ainly, some justice to this, if it means t.hat the
reist is restrained from emharking on gratuitous attempts to reform the
language of his fellows, language which must surely be in order as it
is. What Lejewski has to say should not, however, be interpreted as
implying that we may properly ignore those forms of everyday and
scienti fie language which pose prima facie problems for the would-be
reist translator.
3OS .... e.g. Br .. ntano <187~, pp.12~-32, Eng. pp.88-9~), Twardowski (189~,
p.3, Eng. p.t>, Hofl .. r (1890, §6).
ON THE PHASES OF REISN 149
Suppose, however, that during some given period of time one and the
same sentient organism is both thinking and jumping. The same thing, in
such circumstances, is both a thinker and a ,jumper. Is not the reist
left in such circumstances with no means in his ontology to distinguish
between what are, surely, activities of different sorts? Clearly, he
cannot solve this problem by appealing to the fact that different
(mental and physical) predicates are applied to the thing in question,
for the issue here is precisely that of establishing in virtue of what
such different predications count as true, and to this end the reist has
only things to which he can appeal. The problem cannot be solved,
either, by appealing to any special understanding of the m~terial 'thing
that thinks' (which had been left indeterminate by Kotarbinski himself).
For whichever concrete thing' is fixed upon by the reist as that which
thinks, be it the brain, the central nervous system, or some other
proper or improper part of the organism as a whole, there will 'always be
physical truths about the thing selected in relation to which the given
problem will recur. Moreover, whatever the nature of the material thing
that the reist puts forward as his candidate 'thing that thinks', it
seems not logically excluded that two parallel consciousnesses should be
realised simultaneously within it, after the manner of Siamese twins. We
might then have occasion to assert that consciousnessl is thinking this,
while consciousness2 is thinking that, and then it seems that the reist
- short of assuming special immaterial things - would have ,no way of
doing justice to the presence of parallel thinking processes in the
given case, for there are ex hypothesi no separate bodies which might
here serve as subjects of the respective predications. What is not
logically excluded seells thereby to be ruled out by the fiat of the
reist's linguistic predilections.
Reism has consequences not only for the subjects of mental experi-
ence, however, but for the objects of such experiences also. As already
stated, the reist insists that that to which our experiences are related
is in every case a thing. In everyday perception, as also in hallucina-
tions, dreams and memories, we are typically presented with external
things which seem to us to be coloured and shaped in 'this or that par-
ticular way. And in dreams and memorie.s, as Kotarbinski puts it, we as
it were "observe, though somehow in a secondary manner, things from our
past enviroment, which seem to us to be such or another" (1966, p.431).
This account will clearly face problems in connection with iterated
reference to what is mental - dreams about dreams, for example ~ as also
. in connection with that peculiar selectivity of memory {lnd other acts
discussed in §4 above. Kotarbinski's view, nevertheless, is that our
mental experiences are in every case a matter of our being related in
special ways to things. From t.his it follows that all (third person)
psychological statements must have literal r~adings of one or other of
the forms:
A feels this: B,
A experiences this: B,
A thinks this: B,
and so on - where 'A' stands in for the name of some sentient body and
'B' for words or phrases which answer the question 'what?': 'What does
John imagine'?, 'What does John think?', 'What does John want?', and so
ON THE PHASES OF REISM 151
For this to make sense in reist terms, therefore, it must be that our
utterances themselves are in some extended sense imitations of the very
psychic determinations they bring to expression 39 • Thus in the general
formula of the psychological statement 'A experiences this: B', the 'B'
may be seen as an imitation in this extended sense by the one who makes
the given statement of the relevant experience on the part of A. When I
say, 'John thinks this: 2 + 2 = 4', 'John feels this: they are playing
badly', 'John doubts this: to be happy', then I become a samesayer with
the way John thinks or feels. And we can even
generalize this formula so that not only a sentence, but any phrase
referring to how a given person experiences, could be substituted
for 'B'. It might even be an inarticulate exclamation, SO that a
given psychological statement would be: 'John experiences so: Oh!'
(1966, p.428)
41(1927/31; p.l0)
ON THE PHASES OF REISM 153
44Cf . Brentano (1933, pp.37, 108, 112, 131, Eng. pp.37, 86, 88, 109).
154 BARRY SMITH
(v) They are that which, while remaInIng numerically one and the
same, can admit contrary accidents at different times. (Cat.,.4 a 10)
(vi) They are able to stand in causal relations. (Met., 1041 a 9)
(vii) They are "one by a process of nature". A substance has the
unity of a living thing. Hence it enjoys a certain natural completness
or rounded-offness, both in contrast to parts of things and in contrast
to heaps or masses of things (Met., 1040 b 5-16, 1041 b 12, 1041 b 28-
31, 1052 a 22ff., 1070 b 36 - 1071 a 4; Cat., 1 b 5)45. Hence also, for
Aristotle, a thing is that which has no actual but only possible parts
(Met., 1054 a 20ff.). A part of a thing, for as long as it remains a
part, is not itself a thing, but only possibly so; it becomes an actual
thing only when it is somehow isolated from its environing whole. In
this sense (and also in others) the substance is the bearer of
potentiality, and it is at this point that we should have to list those
marks of substance which flow from Aristotle's hylomorphic theory, and
from his theory of act and potency.
There are further marks of substance, less easily documented in
Aristotle's texts since they were taken entirely for granted in
Aristotle's day. These are above all:
(viii) A substance is independent of thinking, a part of nature -
where no Greek would have understood what is meant by • independent of
thinking' .
(ix) A substance is that which endures through some interval of
time, however small. This means, firstly, that things exist continuously
in time (their existence is never intermittent). But it means also that
there are no punctually existing things, as there are punctual processes
or events (for example beginnings, endings, judgings, decidings, and
instantaneous changes of other sorts)46. A thing is also typically such
as to endure for such a length of time that it may acquire a proper name
for purposes of reidentification47.
(x) A substance is that which has no temporal parts: the first ten
years of my life are a part of my life and not a part of me. As our
ordinary forms of language suggest, it is events and processes, not
things, that have temporal parts.
Even leaving aside, now, the passages where Kotarbinski explicitly
allies himself with Aristotle48 , the focal instances of the concept of
thing made prominent in the Eleaenty' make it clear that he had intended
to follow Aristotle in almost all of the above. A body, as we have seen,
is bulky and lasting and such as to offer resistance. Further marks of
bodies distinguished by Kotarbinski are:
47As is clear from Arist.ot.le's t.reat.ment. of <ii), it. is possible t.hat. t.he
marks of t.he concept. of subst.ance may be est.ablished in part. t.hrough
considerations o:f the language we use "'0 re:fer i-o subst..ances t.hemselves.
See Met.., :1029 b 13.
4,9(1966, pp.327, 4,35, 34,2). Not.e t.hat., like inert.ness, t.hese marks are
held by Kot.arbinski t.o be incident.al; t.hat. is, t.hey do not. aff·ect. t.he
ext.ension of t.he concept. ~ or 6~<ty.
51(1894., pp.88, 105, Eng. pp.86, 100). Cf'. Ingol"den <1964./65, vol.l, p.219).
156 BARRY SMITH
Here 'a', 'b' and 'c' are any expressions belonging to the category
name. This means, as Lesniewski sees it, that they may be either:
(1) ordinary singular designating names or nominal expressions like
'Ronald Reagan' or 'the British Prime Minister';
(2) shared or general names like 'philosophers' or 'apples in
Vermont' j
(3) fictitious or empty singular names like 'Pegasus' or 'the
largest prime number': or
(4) fictitious or empty general names like 'sirens' or 'fates ,.56.
All such expressions belong to a single category, Lesniewski argues,
since whether a name like 'man at the door' is singular or shared or
empty depends on the factually existing state of the world, and so'
cannot be regarded as basic from the point of view of logic57 • From this,
however, it follows also that we must admit as names expressions like
'jumper', whose number is in a certain sense indeterminate.
The axiom of Ontology, now, lays down simply that for 'a is b' to
be true, it must be the case that every a is b and that e~actly one
object is a. It is not difficult to show, on these terms, that if both
'a' and 'b' are singular and designating, then 'a is b' is deductively
equi valent to .' a= b'. But now, applied to what has now become the
Polish-sounding sentence 'John is jumper', this analysis of 'is' tells
now, since John's circle of similars qua jumper is different from his
circle of shilars qua thinker, this- may enable the reist to distinguish
separate truth-makers for 'John is thinking' and 'John is jumping' even
in those cases where the two activities are performed simultaneously.
From this it would follow that the family of jumping things contributes'
in some way to making it true that John,. in particular, is jumping -
a consequenc&which certainly goes beyond what Lesniewski himself had to
say on these matters, but which nevertheless has advantages from the
reist point of view (to the extent that it has been found acceptable by
Lejewski). In particular, it enables the reist to distinguish the truth-
makers of 'John is Jumping' and 'John is thinking' even in those
circumstances where jumping and thinking are simultaneous: other jumpers
contribute to making true the former sentence in a way in which they do
not contribute to making true the latter.
This is not quite all that can be said on Lesniewski's behalf, however,
and before returning to our discussion of Kotarbinski "s own ontological
views it will be useful to look at the Lesniewskian treatment of the
phenomena of verbal tense. Recall that the 'is' in 'John is jumping' is
intended to express a real present tense. The Polish 'jest', on the
other hand, for example in 'Jan jest skaczc!lcy' (John is jumping) -
a form which sounds odd due to the absence in Polish of the continuous
aspect -does not express a present tense, and this holds too of 'Jan
skacze' (John jumps) and 'Jan jest skoczkiem' (John is a jumper).
In and of itself the Polish 'is' is timeless. In order to mark the
fact that the jumping is taking place at the moment, the speaker of
Polish must add an explicit temporal index and say, for example, 'Jan
teraz skacze' (John jumps now) or (more stiltedly) 'Jan jest teraz
skacz~cy' (John is now jumping). This timelessness, we now see, must be
characteristic also of the 'is' of Lesniewski's Ontology. This is first
of all because Lesniewski insisted that the sentences of Ontology should
be absolutely true, Le. true independently of time and occasion of
utterance58 • But it is also because, as already noted, the 'is' of
Ontology is to enjoy absolute universality of scope; it is to be
applicable to abstracta as much as to concreta, to objects past and
future as much as to objects of the present. It is in f.act the same
timeless 'is' as that which we employ when we say, e.g., 'Socrates is a
philosopher' or '3 is a prime number' or 'whales are mammals'.
Ontology is not, however, restricted to 'timeless' sentences of the
given sort. Return, for the moment, to 'Jan je-st teraz skacz~cy' (John
is now jumping). We should normally interpret the temporal index ('now'
or 'teraz') in such a sentence as governing the verb. Given the
69Bolzano, t.oo, in §46 of" t.he Wi .... en .. chaf"t. .. lehre, sees t.ime-det.ermina-
t.ions as part. of" t.he subject., so .... hat., as he put.s it., "a pair of" prop-
osit.ions such as 'Caius is now learned' and 'Caius was not. learned t.en
years ago' t.urn out. t.o have dif"f"erent. subject.s".
62Sinisi <1983,p.66)
64 I t.
should go wit.hout. saying, in light. of our discussion of Ont.ology in
§7 above, t.hat. we do not. share t.his est.imat.ion of t.he nat.ure of
Lesniewski's achievement..
162 BARRY SMITH
67A nat.ural longuage reading of 'John now ' may involve a necessary
indet.erminacy in t.he precise ext.ent. of t.he relevant. int.erval; since t.his
indet.erminacy can in principle be eliminat.ed, however <for example by
ut.ilising t.he resoure .. s of a formal t.heory such as t.he 'Chronology'
proposed by L·ejewski in hi~ (1982>, we can ignore t.he mat.t.er her ...
164 BARRY SMITH
68ThiS analogy is of course very rest.rict.ed (see e.g. Mellor 1981, pp.
66f., 128ff.). The accept.ance of t.he concept. of a world line in a four-
dimensional cont.inuum is moreover fully consist.ent. wit.h a cont.inued
belief in t.he ont.ology of t.hings (Sellars 1962, p.678). One might. indeed
go furt.her and argue t.hat. t.he concept.s of t.he four-dimensional ont.ology
t.hemselves presuppose t.he I-hing onl-ology for t.heir coherent. formulat.ion,
that.., for exomple, t.he idea of a world line makes senSe only if t.here is
some ident.ical ~ t.hat is tracked from one t.ime-point. on t.he line t.o
anot.her (see Simons 1987, pp.126f.). See also Brent.ano (1976, pp.296ff.)
for furt.her criticisms of t.he view of t.ime as a 'fourt.h dimension' of
space.
ON THE PHASES OF REISM 165
~lThis is to exclude, f'or example, the case where a .e,~-na {.u:te bulky and
resistant whole is united with, say, an empty volume of' spacetime.
ON THE PHASES OF REISM 167
72See Simons (1983) for a formal t.reat.ment. of a view along t.hese lines
wit.hin a Lesniewskian framework.
73See Mourelat.os (1981), Galton (1984, Appendix II), Hoeksema (1985, ch.6).
168 BARRY SMITH
events, where we can distinguish on the one hand unitary events which
take time but have no homogeneous sub-events as parts, for example
judgings, decidings and so on, and on the other hand events which are
strictIi' punctual, such as beginnings, endings and instantaneous
changes 4. Lesniewski's Ontology and Mereology, now, have shown them-
selves surprisingly adept at coping in a formally rigorous way with
relations such as this in so far as these are manifested among things.
Truth, however, is a relation which involves not only things and the
names of things. It involves also verbs and that in reality to which
verbs correspond, which is typically an event of one or other sort.
Hence we can begin to understand why it is that the bicategorial ontol-
ogy may be particularly suited to the task of giving an account of what
makes sentences (particularly empirical sentences) true. For it allows
us to take account of just those differences in reality which are
reflected in language in the differences of verbal aspect (differences
which are preserved, incidentally, even if we move over to a language
shorn of tenses, of the sort that was favoured by Twardowski,
Lesniewski, and other proponents of the 'absolute' theory of truth).
74It is above all Ingarden who has invested effort in the ontological
analysis of these distinctions. See his <1964/66, ch.V).
ON THE PHASES OF REISM 169
Among the entities that have parts, there are some whose whole is
not composed of a multiplicity of parts; it appears much rather as
an enrichment of a part, though not as a result of the addition of
a second part. One example of such an entity is a thinking soul. It
ceases to think and yet remains the same soul. But when it starts
to think again no second thing is added to that entity which is the
75See Twardowski <1894-, pp.12f., Eng. pp.l1f.) and also his (1923). Twar-
dowski's t.heory was derived ill t.urn from Brent,ano <1924-/25, vo!.II, p.62
Eng. p.220). See also Husserl <1979, p.309), Mart.y <1884-, pp.179f'.;
1895, p.34-; 1908, pp.518f.) and Lesniewski <1927/31, p.4-8, n.78). There
may be a remnant. of t.his doct.rine of' modifying expressions also in
Kot..al'binski's not.ion or 'subst.itut.ive rendering's' or 'onomat..oids' dis-
cussed above.
77Me t,. 1206 b 16; see also 1024- b 30, 1018 a 2, 1030 b 13. See also
Suarez's discussion of t.he acc-i4e1w CGn.c~ in his Disput.at,iones
Met,aphysicae, XXXIX, s.1,n.l0-12 and cf. Cajet.an, In De Ent,e et. Essent,ia
d. Thomas Aquinat,is, §§ 153f·f'.
ON THE PHASES OF REISM 171
soul. What we have here, then, is not like what we have when one
stone is laid alongside another or when we double the size of a
body. •.• The substance is a thing and the accidentally extended
substance is again a thing, though but a thing not wholly other in
relation to 1;he substance; hence we do not have that kind of
addition of one and one that leads to a plurality.
(Brentano 1933, pp.53f., Eng. pp.47f.)
It is not that the being of A must come into being in order for the
judgment 'A is' to be transformed from one that is incorrect to one
that is correct; all that is needed is A. And the non-being of A
need not come into being in order for the judgment 'A is not' to be
transformed from one that is incorrect to one that is correct j all
that is required is that A cease to be. And if only this happened
and nothing else ••. would there not be in this fact alone, which
relates to what is real, everything that is needed for the correct-
ness of my judgment? Without doubt ••• And thus the doctrine of the
existence of such non-things has nothing whatever in its favour.
(Brentano 1930, p.95j Eng. p.85)
821ngarden put.s t.his view t.o Kot.arbinski around 1936 when t.he lat.t.er came
t.o Lv';w t.o give a t.alk on Leibniz as a precursor of reism. (Personal
communicat.ion of W, Bednarovski)
83He was followed in t.his by his lat.er disciples, above all Oskar Kraus,
Alfred Kast.il and 'George Kat.kov, who t.ook over from Kot.arbinski t.he word
'reism' t.o describe t.he lat.er Brent.anian view. See Kraus (1937, pp.
268ff.).
84 '
Brent.ano <192~/26" vo1.ll, p.162; Eng. p.29~)
174 BARRY SMITH
Aquinas, too, sees the concept of a thing or of what is real as the most
general concept to which reason can attain90 , and a broadly similar view
is present for example in Husser!' s Philosophie der Ari thmetik, where
the purely formal concept of an Etwas or 'something' lies at the basis
of Husserl's theory of number and counting.
For Kotarbinski, on the other hand, 'thing' is a term of material
ontology, to be understood by reflecting on specific sorts of examples
of thing and of the meanings of terms like 'bulky', 'extended', 'inert',
and so on, whose significance is confined to the region of physical
bodies.
Marty is in this respect closer to Kotarbinski than he is to
Brentano, since he pursued energetically the idea that the concept of
thing or ens reale should be confined to those entities which partici-
87<192.t/26, Yol.II, pp.162, 213r., Eng. pp.29.t, 321f.; 1933, p.18, Eng.
p.2.t; 1956, p.38)
890n the opposition between formal and material concepts see Smith (1961>;
on Breni.ano and rorlllal concepts see Munch· (1986); on formal YS. mat.erial
ontology see Ingarden <196.t/66, esp. §9).
91<1908, §66)
99See (1966, pp.369f.). Cf. also <5J~. cU., p.191, "here Kotarbinski denies
the suggestion that 'is' "ould be an abbreviation of 'is no,,'.
101No t.e t.hat. ~ could not. be candidat.es for t.he role of t.hat. which
individuat.es one body from anot.her (as was held for example by Lot.ze);
for as we have seen, everyt.hing t.hat. exist.s is for Brent.ano such as t.o
ex i s t. .j,n lhe ~am.e U1ne.
180 BARRY SMITH
For all the divergencies between Kotarbinski's pansomatist reism and the
later formal ontology of Brentano, there is a clear sense in which they
are proponents of a common approach to ontology. This approach is shared
also by Lesniewski and his followers, as also by Quine, Goodman and
other modern nominalists. It can perhaps be characterised as an approach
which takes as its starting point in the construction of its ontology a
view of things drawing equally on examples of qua.ntities, ma.sses or
homogeneolls coll ecti \'es as on the uni tary substances of the tradition.
Thus it is contrasted with the approach to ontology of Aristotle 103 ,
Leibniz, Twardowski and Ingarden, which takes its cue primarily from the
uni tary substance and from the individual accidents which may inhere
therein.
There are, interestingly, a number of different routes taken by the
philosophers mentioned to the· homogenous collective view of things
('homogeneous' because the distinction between thing and mass is held to
reflect no fundamental ontological division). Thus Quine, for example,
seems to have been inspired particularly by those physical examples
(energy fields, liquids, gases) where arbitrary delineability does
indeed seem to hold, as also by related considerations deriving from the
semantic treatment of mass terms in· natural language. Quine, like
Brentano lO 4. and Kotarbinski I sees masses as full-fledged even though
possibly scattered individuals. Thus he regards as of no importance the
difference between what is spatially continuous and what is spatially
scattered 105 , and indeed his general approach is to view every object as a
four-dimensional section of the world, after the fashion of what Zemach
calls 'events'.
102Some clues as to Brentano's views are provided by the argument 1'or the
a fV1Ac't-iimpenetrability of bodies in his (1976, pp.1801'.). Brentano's
views in this respect seem to have been stimulated by the positivism of
Auguste Comte: see Munch (forthcoming).
106See t.he "Appendix" t.o Bl'ent.ano (1933): "The Nat.ure of t.he Corporeal
World in t.he Light. of t.he Theol'Y of Cat.egol'ies", and also Bl'ent.ano
<1976, p.184).
We could go further, and extend the chart by taking into account opposi-
tions of other sorts, relatin§, for example to the issue as to whether
enti ties are or are not general l 9, independent of mind llO , atomic 111 , or such
as to have temporal parts 112 In addition we could think more carefully
about the different meanings of 'unitary', distinguishing for example
the requirement of connectedness of parts, the requirement of spatial
separateness from other entities, the requirement of functional inter-
dependece of parts, and so on. We could investigate further the extent
to which things may have parts which are themselves things - as an
organism may include cells, chromosomes, genes, etc. as parts113.
Already as it stands, however, the chart will enable us to see the
inadequacy of any simple-minded opposition between 'reism' on the one
hand and 'Platonism' on the other. Thus it would be wrong to go along
108Again, t.he reader should bear in mind t.hat. it. is by no means clear
where Arist.ot.le would have t.o be post.ed on t.his spect.rum if t.he whole of
his ontology ~ere laken into accoun~.
114-Compare Lejewski's assert.ion in his (1976) t.o t.he effect. t.hat. event.s
are non-mat.erial object.s, so t.hat. anyone who admit.s event.s int.o his
ont.ology would be a 'Plat.onist.' on Lejewski's reading of t.his t.erm. This
reading is no doubt. derived from t.he fact. t.hat. event.s, for Lejewski, are
t.he ref'erent.s of abst.ract. nouns like 'swimming', 'f'alling', '\'alking'.
Klemens Szaniawski
The first formulation of the concretism, which then and afterwards has
been called reism, can be traced back to 1929. In Elementy in a para-
graph closing the discussion of ontological categories, we read as
follows:
" ... the entire reduction of categories, as outlined above, took place
precisely to their (Le. things' - K.S.) benefit. That reduction com-
pleted, it turns out that there remains only that category of objects -
that is, there are no objects other than things, in other words, every
object is a thing, whatever exists is a thing. When metaphorical, ab-
breviated, picturesque, in a word, substitutive, formulations are elim-
inated and replaced by the basic formulations, interpreted literally,
the latter will include no phrases which would appear to be names of
something other than things. They will be statements about things only.
But it must be emphasized here that by things we do not mean only inor-
ganic solids. Things are inorganic and organic, inanimate and animate,
and 'endowed with psychic li fe' - that is, they are both things in the
narrower sense of the word, and persons, too. So much for the reduction
of categories of objects to the category of things. The stand taken here
by those in favour of such a reduction might be called reism." (p. 55-6)
Even this first formulation reveals two non-equivalent variants of
the thesis of concretism. One of them is a statement about the world.
The fundamental philosophical question 'What exists?' is answered:
'Things and only things.' The other deals with language; it says that
all the statements made in a descriptive language are reducible to such
sentences which refer directly to things alone, i.e. that the only names
they contain are individual and general names of things.
The evolution of concretism, I think, consisted mainly in replacing
the ontological thesis by the semantical one and in substantial weaken-
ing of the latter; the statement about language has been transformed
into a prescription for its users; a doctrine became a programme.
The ontological variant has been abandoned (what does not mean
giving up the ideas that motivated it) mainly because of certain diffi-
culties involved in attempts to find a consistent formulation for it.
Briefly speaking, the point was that statements denying existence of
properties, relations, etc., and ascribing it to things alone, violated
185
J. Wolensld (ed.), Kotarbinsld: Logic, Semantics and Ontology, 185-190.
© 1990 Kluwer Academic Publishers.
186 KLEMENS SZANIAWSKI
1"N&" Essays on Human Underst.anding", Book II, Ch. XXIII, § 1, in: The
.'hUosophical Works of Leibniz, New Haven, -1890, p. 340.
PHILOSOPHY OF THE CONCRETE 189
If one shares those ideas he will not deny anybody's right to re-
flection upon individual existence; to emotional response, intensified
by conscious efforts, to the most general features of the world. On the
contrary: those experiences and attempts to express them are to be re-
garded as distinguishing traits of human consciousness. It is not there-
by implied that we lIay interpret as genuinely cognitive, statements
which purport to say something about the world, while in fact they are
but expressions of purely emotional attitudes. The above position is
followed in the philosophical work of the descendants of Kotarbinski's
school.
Nothing more can be achieved in science than to become the founder
of a school. Only few succeed in it; some highly distinguished scholars
fail to achieve that aim. Undoubtedly, a specific type of personality is
essential here: a strong individuality, concentrated on a clearly formu-
lated programme and, at the same time, open to others, gaining associ-
ates owing to attention paid to their development and careful observance
of the principle of partnership.
A scientific school creates peculiar values. In Kotarbinski's
school the value that has won the supreme position is intellectual
rigour. The intention of the present attempt to realize this fact is to
repay, if only in part, a debt of gratitude Kotarbinski's pupils owe to
their Teacher.
KOTARBINSKL MANY-VALUED LOGIC. AND TRUTH
Jan Wolenski
191
J, Wolenski (ed.), Kotarbinski: Logic, Setnmltics and Ontology, 191-197.
© 1990 Kluwer Academic Publishers.
192 JAN WOLENSKI
weak; in order to argue for the view that predeterminism is not univer-
sally valid he comes to considerationa about eternality and preeternali-
ty of truth.
The crucial concept, the one of truth, is defined in the following
way
(2) a proposition A asserting an object 0 is true if and only if 0
exists.
Now, every truth is eternal, i.e.
(3) for every A, if A is true at t, then A is true at every tl such that
t < tt.
Thus the problem consists in whether every truth is preeternal. If so,
then
(4) for any A, if A is true at t, then A is true at every tl such that
t > tl; of course, any preeternal truth is, by (3), eternal too.
According to Kotarbinski, some truths are preeternal but others are not.
To see this suppose that something, say 0, may be created by a human
action. Thus 0 does not exist until its creation and, by (2), proposi-
tions asserting 0 are not true. Let A be such a proposition. We know
that A is not true. Let us ask whether A may be false. If A were false,
its negation would be, by (3), eternally true. In that case it would be
impossible to create 0, which is contrary to the assumption. Thus A is
not a preeternal truth; there is a t, such that A is neither true nor
false at t.
Because the negative answer to (4) implies the existence of propo-
sitions which are neither true nor false, the problem of the law of
excluded middle arises. Kotarbinski considers the following forms of
this principle:
(5) for any A, either A or ,A is true;
(6) of any A, A is either true or false;
(7) for any A, if A is true, then ,A is false.
Kotarbinski points out that (5) and (6) assume the completeness of the
division of all propositions into true and false, which yields to the
equation
(8) true = not-false
On the other hand, (7) is independent of (8). Thus one may adopt (7) and
the negation of (8). An ontological counterpart of this combination is
the world with decided and undecided things. Respectively, all proposi-
tions are divided into definite (true or false) concerning decided
things, and indefinite (neither true nor false) which are related to
undecided things. (5) and (6) have to be restricted to definite proposi-
tions; (7) is a general rule. Moreover, we have the forms:
(9) for any A, A is either definite or indefinite.
This completes Kotarbinski's argument that the existence of indefinite
propositions as well as truths which are not preeternal does not violate
KOTARBINSKI, MANY-VALUED lOGIC, AND TRUTH 193
the Rubicon'. The last sentence is timeless and may be uttered at any
moment t. The general interpretative strategy recommended- by Lesniewski
consists in putting the temporal index into the grammatical subject of a
sentence. This device enables Lesniewski to regard the sentences 'Ceasar
will not cross the Rubicon in 49 "B.C.' uttered before 49 B.C. and
'Ceasar crossed the Rubicon in 49 B.C.' uttered today as a pair of
contradictory sentences; on Kotarbinski's view the former is an indefi-
nite proposition while the latter is a definite one. Let us note that at
this point Lesniewski closely follows Twardowski's principle (1). Twar-
dowski himself also disagreed with Kotarbinski in his (1971)2. Kotarbin-
ski subsequently accepted Lesniewski's criticism as a conclusive one and
has abandoned logical indeterminism. 3
tukasiewicz himself did not take part in the outlined discussion
until 1918. On March 7th, 1918 he delivered the lecture at Warsaw Uni-
versi ty in which he gave information about three-valued logic: "In 1910
I publ ished a book on the principle of contradiction in Aristotle's
work, in which I strove to demonstrate that that principle is so evident
as it is believed to be. Even then I strove to construct non-Aristote-
lian logic, but in vain. Now, I believe, I have succeeded in this. My
path was indicated to me by antinomies which proved that ,there is a gap
in Aristotle's logic. Filling that gap led me to a transformation of
traditional principles of logic [ ••• ] I proved that in addition to true
and false propositions there are possible propositions, to which object-
ive pqssibility corresponds as third in addition to being and non-being.
This gave rise to a system of three-valued logic, which I worked out in
detail last summer. That system is coherent and self-consistent like
Aristotle's logic, and is much richer in laws and formulas [ .•• ] Poss-
ible phenomena have no causes, although they themselves can be the
beginnings of a causal sequence. An act of a creative individual can be
free and at the same fact affect the course of the world." (1918, p. 86)
Kotarbinski was, like the majority of young Lvov philosophers,
greatly impressed by tukasiewicz (1910). It is fairly evident that he
read and perhaps even participated in the meeting of the Polish Philoso-
phical Society in Lvov (February 26th, 1910) where tukasiewicz was
lecturing on the principle of the excluded middle. 4 On the other hand,
Kotarbinski (1912) goes far beyond tukasiewicz's remarks on the connec-
tion of the excluded middle and det~rminism. Thus we have every reason
to regard that paper as a very original contribution. Now, what about
the reverse influence, i. e. that of Kotarbinski on tukasiewicz? The
printed evidence is unfortunately very small - especially since Kotar-
binski's name is not mentioned in tukasiewicz' s works on many-valued
logic. Jordan (1963) claims that tukasiewicz in his treatment of strict
determinism as a semantic principle follows Kotarbinski's ideas; it is
not impossible, however, that Jordan is relying here on the oral tradi-
tion of the Lvov-Warsaw School. The only explicit sources are Greniewski
(1957) and (1958). He suggests on the basis of an oral report by Ajdu-
kiewicz that tukasiewicz was really influenced by Kotarbi~ski; according
to Greniewski, Ajdukiewicz was repeating tukasiewicz's own opinion. Gre-
niewski himself, impressed by Ajdukiewicz's report, claims that Kotar-
bi~ski invented a system of three-valued logic. As far as I know this is
all that one may find in the published sources.
Doubtless, some problems involved i.n tukasiewicz (1918) and his
subsequent works on many-valued logic are analogous to those involved in
Kotarbi~ski (1912), for instance, about creative human activities, prop-
ositions about future facts, propositions other than true or false,
possibility and preeternal objective contingencies. Thus it is arguable
that ideas expressed in Kotarbi~ski 's paper helped tukasiewicz to de-
velop three-valued logic. On the other hand, I do not agree with Gre-
niewski's thesis about Kotarbinski's being the inventor of a formal sys-
tem of three-valued logic. Indefiniteness may be taken as a third logi-
cal value but it is not the only possible interpretation. The second way
is to understand indefinite propositions as truth-value gaps. I see no
conclusive reasons supporting the view that Kotarbinski in his paper
prefers the first interpretation. Hence one should be cautious with
formalisation of Kotarbi~ski's ideas. If A is an indefinite proposition,
,A is one too; they both can either have a third logical value or ex-
press truth-value gaps. Even if tukasiewicz was impressed by the idea of
indefinite propositions, he interpreted them from the very beginning
within the framework of many logical values. 5 And this was not only the
interpretation of Kotarbinski's view but an entirely new proposal. One
other point should also be strongly underlined here. Kotarbinski links
indefinite propositions with the refutation of the law of excluded
middle. This is not tukasiewicz's opinion, however, since he points out
that three-valued logic is based on the negation of the law of bivalence
which is a metalogical principle. On the contrary, although (5) - (7)
are formulated in metalogical manner, Kotarbinski seems to work inside
logic, not in metalogic. And this is, in my opinion, a crucial differ-
ence between both of these approaches to indefinite propositions.
It is interesting to ask whether Lesniewski's proof is valid with
respect to tukasiewicz's conception; as a matter of fact tukasiewicz
never addressed himself to this question. As I have already said Les-
niewski's intepretation of tensed sentences is an essential ingredient
of his proof that indefinite propositions are impossible. But tukasie-
wicz could say that he is not committed to such an interpretation.
Now, let us come back to the successive steps of Lesniewski's proof
providing tukasiewicz' s three-valued logic. (15) is a premise for the
reductio ad absurdum proof. (14) should be supplemented by the observa-
tion that 1/2 is the logical value of A. Furthermore, (15) is misleading
even on Kotarbi~ski's assumptions: if A is not preeternal, both A and ,A
are indefinite in Kotarbinski' s sense j in tukasiewicz' s logic both of
these propositions are valued at 1/2. Thus ,A cannot be true. However,
Boguslaw Wolniewicz
(R) Va e: L 3a 'e: Le
,
a Cn a ,
~~
where the formula "a c~ a'" is to say that the proposition a is equi-
valent to the proposition a' under the consequence operation Cn fixed
for L. Hence Cn cannot be simply the logical consequence operation of L,
being based on at least some additional rules of inference or defini-
tions. (Clearly a proposition like "if it's raining, then there is a
rain going" is not a thesis of logic by itself.)
The postulate (R) applies to single propositions of L. But we may
express it also in such a way as to apply to sets of them, i. e. as:
The formulae (R) and (R') are equivalent. Indeed, (R) is implied by
(R'). For take an arbitrary proposition a e: L, and set A = {a}. Then by
(R') there is a set A'c Lc such that Cn {a} = Cn A'. Hence a e: Cn A' and
so - with Cn finitistic - there is also a finite subset B'c A' such that
a e: Cn B'. Now taking the conjunction fl' of all members of B' we get:
fl'e: Lc and a e: Cn {fl'}. But B' c A' c Cn {a}, and so - with each con-
junct of fl' following from a - we have: 8'e Cn {a}. Thus we get Cn {a} =
= Cn {8'}, for some fl'e Le, which is what was to be shown.
The converse implication holds too. For take an arbitrary set Ac L.
By (R) there is for each a e A an a 'e Le such that a c~ a'. Now let
r(A) c Le be the set of all such propositions, i.e.
Taking now one Bi'e: r(A) for each Bi e: B, we get a finite set of
propositions B'c r(A). So let r' be the conjunction of all its items,
i.e. Cn{r'} = Cn B'. By construction, the members of Band B' are pair-
wise equivalent, with the relation Cn being a logical congruence.
Hence the conjunctions rand r' must be mutually equivalent too: Cn{r} =
Cn{r'}. Thus we have: Be: Cn{r'} = Cn B' c Cn r(A) which is what was to
be shown.
The converse inclusion is obtained by a parallel reasoning since in
view of the definition of r(A) we have the formula:
(2) Va'e: r(A) 3a e: A a' c~ a
symmetric to (1).
CONCERNING REISM 201
cate two propositions a' and /3' which are strictly equivalent to the
former, and which do not contain any onomatoids; i.e., the propositions
'John is older than his son' and 'Gerlach is higher than Giewont'. The
latter contain no onomatoids, for - as Kotarhinski puts it - phrases
like 'older than' and 'higher than' are not names of relations. To put
the point in general terms, the difference hetween t.he two couples of
propositions at hand is like the difference between the formulae
'R(x,y)' and '(x,y) E R'. In the latter the symbol 'R' is an onomatoid,
in the former it is not.
The two phrases mentioned are certainly no names, for they are both
predicates. But does that mean that they do not refer to anything, nor
denote it? The proposition 'it is a fact that ,Jack and Jill went up the
hill' is true if and only if Jack and Jill went up the hill, i.e. if the
proposition ',Tack and .Till went up the hill' is true. But why should we
take this as an indication that there are no facts in the world, only
couples of people walking up hills? And how does one justify the view
that participial expressions like 'Jack and Jill walking up the hill'
denote things, and not some syntactically reified facts? Or to put it
generally: why should the circumstances that to each proposition a con-
taining the expression 1'; one may indicate an equivalent proposition a'
not containing t; be regarded as proof that the expression l; is devoid of
the semantic function of denoting?
With regard to reism this very objection was raised already half a
centllry ago by Ajdukiewicz (1930), in his review of Kotarbinski' s Ele-
menty: "the fact that propositions including 'apparent terms' may be re-
placed with equisignificant ones which do not include them proves only
that [ ..• J we can do without apparent terms." (see this volume, p. 12).
This objection seems very important to us, and difficult to meet. The
task of meet iIlg it, however, rests in equal parts wi th the reist and
with his opponent. For it represents their common cause, though
approached by them from opposite quarters.
The variant of reism confined to that rule only might be called moderate
reism. What is demanded now is not any more the complete elimination of
hypostases. We wish them only subjected to some limitations. If you want
to talk about abstracta like 'temporation' or 'the structure of defer-
ring', please do it at your convenience. But we should appreciate it if
you took at least into account that if something is to hold in the realm
of abstracta - if, e.g., the structure of deferring makes it impossible
to reduce temporation to a simple complication of the living present -
then something else should hold in the realm of concreta. What exactly?
For reism answering this question is a test of meaningfulness. And if no
answer is coming forth, then it becomes clear that hypostatiz ing has
already trespassed safe limits.
The reist's test would be effective if we had any general principle
for correlating particular hypostatizing propositions with their con-
crete truth-conditions. No such principles are extant so far, however.
And so the transcedentalists could easily satisfy the reist's request by
correlating his dicta with arbitrary true propositions referring to con-
creta, e.g. with the consecutive items of the Warsaw telephone direc-
tory.
So even moderate reism, as a recipe for doing philosophy, is merely
a programme, a declaration of intent. In fact, all its efforts went in a
different direction: to oppose the abstractions of set theory, not the
products of neoromantic lucubrations. But the former are heavily bridled
by set theory itself, while the latter do not know any bridle at all.
204 BOGUStAW WOLNIEWICZ
5. The glllf between reism and an ontology of facts is wide enough. But
it looks less unbridgeable if one tltkes into consideration that they
have something important in common: their respect for logic. Tndeed,
both were inspired by logic, growing out of it as its metaphysical
extensions: one from the logical systems of Le~niewskj, the other from
those of Frege and Russell. The real philosophical abyss opens up only
where the consensus on logic comes to an end. When we hear war-cries to
the effect that "defining impairs the theoretical efficiency of con-
crepts", so down wi th defining; or t.hat "the ideal of univoci ty is to Iw
rf·jPcled as a mattpr of principlp"; then we see that what one tries to
do herp is replacing the rules of logic by those of euphony. It does not
matter an) more whether something in philosophy is consistent with some-
thing plse, or not. What matters is merely that it should have It good
sound, "'ith 'g,)od' meaning either 'elated', or 'scientific', or 'human-
istic', or just 'swell'.
H"ism is not a barrier massive enougb to withstand the concerted
onslaught of all those - as Professor Kotarbinski said on occasion -
"soulful men forming a trend semi-rational only, and steering towards
goals still less rational" (1958a, p.731). Such a barrier, however was
the Reist himsplf, that grand seigneur of Polish philosophy. And while
he was there, he gave a warrant by his own person that nothing really
bad may happen to that philosophy; the standards will be kept up. But
now he is no more.
THE VOICE OF THE PAST IN KOTARBINSKI'S WRITINGS
Ewa iarnecka-BiaIy
~.g. Kotarbinski (1934) reviewed Tarski (1933). The review was quite
enthusiastic and stressed the need or rendering this book accessible
also to the roreign reader - which was done thorough its German (1936)
LranslaLion (English translation - 1966).
20S
I. Wolerfski (ed.), Kotarbilfski: Logic, Semanlics and Ontology, 205-211.
" 1990 Kluwer Academic Publishers.
206 EWA ZARNECKA-BIAf.Y
6Kotarbinski had high regard f"or Bacon as a Renaisance man and humanist,
representing an epoch which, as he pointed out, originated in England
later than on the Continent but which was to rea~h here a more mature
f"orm. Probably Kotarbinski was f"ascinated by Bacon not only as a philos-
opher but also as a praxiologist.
7 The Polish version of" this work appeared as Kotarbinski (1966), sup-
plemented by the author with some new inf"orma'ion included in the two-
paged Afterword, dated .;July 19th, 1966.
THE VOICE OF THE PAST IN KOTARBINSKI'S WRITINGS 209
I t would appear that Kotarbinski did not use the oldest books or
manuscripts. He made reference only to the accessible literature on this
subject, mainly Struve (1911) - see Kotarbinski (1966b, p.15). On the
other hand, Kotarbinski's conception of Polish logic since the 18th
century (Kotarbinski 1959, p. 3) was based on his own research, e.g.,
concerning the history of a textbook written by Condillac specially for
the Polish schools. 8
The synthesis in Kotarbinski's outline of the history of Polish
logic is visible in e.g. the clear-cut periodization of this history. He
suggested five periods: the Renaissance; the Jesuist period; the
Enlightenment; the pre-logistic period of modern times; and the contem-
porary, characterized by the primacy of mathematical logic. This is a
sufficiently detailed division, especially with respect to the earlier
periods; Kotarbinski here accepts Struve's general scheme. It should be
also not.ed that Kotarbinski's acount of Polish logic begins with the
year 1499 - the date of the publication of the first Polish book on
logic, Exercitium novae logicae, a commentary on Organon written by Jan
of Glogow, and not at the commencement of the teaching of logic in
Poland which would take us further back.9
Let \IS add a digression here. Writing about Jan of Glogow
Kotarbinski crit.icised C. Prantl who had said mentioning one edition of
Exercitium: "die Worte 'Tobie Mily Boze Chwala' am Ende der vorletzten
Seite bedeuten doch wohl der Drucker" (1870, p. 29). Perhaps influenced
by Struve's interpretation of Prantl, Kotarbinski (1959, p. 2) remarks:
"The famous erudite Prantl, while reading a later, posthumous edition of
Jan's Exercitium printed in Cracow in 1511, in Latin of course, found
there on the page before last a couple of words unitelligible to him,
which he thought were the name of the printer. These were the Polish
words 'Tobie Mily Boze Chwala' [Glory to Thee, Good God]. They are
evidence of the Polish nationality of the editor, printer or the
author." Really, Prantl did not realize, as he could have said if he had
known it, that the quoted phrase is meaningful in Polish. Nonetheless
Prantl was right associating these words with the printer: it had been
customary for printers to include their inscriptions in the books they
set. It could be a pious invocation, it could also be a cryptographic
text encoding the printer's name.
10Vasil .. v called f'or 'non-Aristot .. lian' logic, hoy .. ver Yithout abandoning
the id .. a of tyo-yaluedn ..... Kline's int.erpretation i . misleading. On the
other hand, Klin .. is not. the only author respon.ible for this conf'usion.
He yas pr .. ce-ded by ChYist.ek and also by Gr·enieyski. See tarnecka <1986,
1988).
THE VOICE OF THE PAST IN KOTARBINSKI'S WRITINGS 211
6. Final remarks.
Martin R.M. (1958) Truth and Denotation, Routledge and Kegan Paul:
London
Marty A. (1895) "tiber subjektlose Satze und das Verhaltnis der Grammatik
zu Logik und Psychologie", 7th article, Vierteljahrsschrift fur
wissenschaftliche Philosophie 19, 19-87
Marty A. (1908) Untersuchungen zur Grundlegung der allgemeinen Grammatik
und Sprachphilosophie, vol.I (only volume published), Niemeyer:
Halle a.d.S.
Mehlberg H. (1963) "The Present Situation in the Philosophy of Mathema-
tics", in B.H.Kazemier and D.Vuysje eds. Logic and Language. Studies
Dedicated to Professor Rudolf Carnap on the Occasion of His Seven-
tieth Birthday, Reidel: Dordrecht, 69-103
Mellor D.H. (1981) Real Time, Cambridge University Press: Cambridge
Mostowski A. (1965) Thirty Years of Foundational Studies (Acta Philoso-
phica Fennica XVII), Basil Blackwell: Oxford
Mourelatos A.P.D. (1981) "Events, Processes and States", in P.J.Tedeschi
and A.Zaenen, eds., Tense apd Aspect (Syntax and Semantics, vol.14),
Academic Press: New York, 191-211
Mulligan K., Simons P.M. and Smith B. (1984) "Truth-Makers", Philosophy
and Phenomenological Research 44, 287-321
Munch D. (1986) "Brentanos Lehre von der intentionalen Inexistenz", in
J.C.Nylri, ed., From Bolzano to Wittgenstein. The Tradition of
Austrian Philosophy, Holder-Pichler-Tempsky: Vienna, 119-127
Munch D. (forthcoming) "Brentano's Soul", to appear in Grazer Philoso-
phische Studien
Novak M. (1963/64) "A Key to Aristotle's 'Substance''', Philosophy and
Phenomenological Research 24, 1-19
Pelc J. - ed. (1979) Semiotics in Poland 1894-1969, Reidel: Dordrecht/
Boston
Prantl C. (1870) Geschichte der Logik in Abendlande, vol. IV, Leipzig
Prior A. (1967) Past, Present, Future, Oxford University Press: Oxford
Prior A. (1976) "Intentionality and Intensionality", Papers in Logic and
Ethics, Duckworth: London
Przel~cki M. (1982) "The Law of Excluded Middle and the Problem of
Idealism", Grazer Philosophische Studien 18, 1-16
Pszczolowski T. (1978) Mala encyklopedia prakseologii i teorii organiza-
cji (Mini-encyclopedia of praxeology and theory of organisation),
Ossolineum: Wroclaw
Quine W.V.O. (1939) "Designation and Existence", Journal of Philosophy
36, 701-9
Quine W.V.O (1953) From a Logical Point of View, Harvard University
Press
REFERENCES 221
Quine W.V.O. (1960) Word and Object, MIT Press: Cambridge, Mass.
Rasiowa H. and Sikorski R. (1963) The Mathematics of Metamathematics,
PWN: Warsaw
Reichenbach H. (1957) Experience and Prediction, 5th edition, The Uni-
versity of Chicago Press: Chicago
Schnelle.T. (1982) Ludwik Fleck - Leben und Denken. Zur Entstehung und
Entwicklung des soziologischen Denkstils in der Wissenschaftsphilo-
sophie, Hochschulverlag: Freiburg i. Br.
Sellars W. (1962) "Time and the World Order", in H.Feigl and G.Maxwell,
eds., Scientific Explanation. Space and Time (Minnesota Studies in
the Philosophy of Science, vol. III), University of Minnesota Press:
Minneapolis, 527-616
Simons P.M. (1983) "A Lesniewskian Language for the Nominalistic Theory
of Substance and Accident", Topoi 2, 99-109
Simons P.M. (1987) Parts. An Essay in Ontology, Oxford University Press:
Oxford
-Sinisi V. (1962) "Nominalism and Common Names", Philosophical Review 71,
230-5
Sinisi V. (1966) "Lesniewski's Analysis of Whitehead's Theory of
Events", Notre Dame Journal of Formal Logic 7, 323-27
Sinisi V. (1983) "The Development of Ontology", Topoi 2, 53-61
Srupecki J. (1955) "S. Lesniewski's Calculus of Names", Studia Logica 3,
7-70
Srupecki J. (1958) "Toward a Generalized Mereology of _Lesniewski",
Studia Logica 8, 131-54
Smith B. (1981) "Logic., Form and Matter", Proceedings of the Aristotel-
ian Society, Supplementary Volume 55, 47-63
Smith B. (1987) "The Substance of Brentano's Ontology", Topoi 6/1, 37-47
Smith B. (1989) "Kasimir Twardowski: An Essay on the Borderlines of
Psychology, Ontology and Logic", in Szaniawski - ed., 313-~3
Smith B. (1990) "Sachverhalt", in J.Ritter and K. Grunder, eds., Histo-
risches Worterbuch der Philosophie, Schwabe: Bas~l/Stuttgart, vol.
VIII
Smith B. (forthcoming) "Brentano and Marty: An Inquiry into Being and
Truth", in K.Mulligan, ed., Mind, Meaning and Metaphysics. The
Philosophy and Theory of Language of Anton Marty, Martinus Nijhoff:
Dordrecht/ Boston/Lancaster
SoboGinski B. (1954-5) "Studies in Lesniewski's Mereology", Rocznik
Polskiego Towarzystwa Naukowego na Obczyznie 5
Sobocinski B. (1971) "Atomistic Mereology", Nortre Dame Journal of
Formal Logic 12, 89-103, 203-13
222 REFERENCES
225
INDEX OF NAMES 226
Tarski A. 44, 140, 176, 196-7, 205 Wittgenstein L. 24, 31, 35,144,
Taylor E. 103 199
Tegtmeier E. 147 Wole6ski J. vii, ix, 137, 139,
Twardowski K. 1, 137-4D, 144, 148, 156, 158, 161, 183, 191, 196
151,155,157,159,168,170-2, Wolniewicz B. 34, 148, 199
180, 182, 191, 194, 196 Woodger J. 108, 163, 182
Wundt W. 121
Vasilev N. 210
Zarnecka-Bia!y E. vii, ix, 205,
Waskowski E. 210 210
Weingartner P. vii Zemach E. 164-6, 180, 182
Well C. 208 Zermelo E. 140
Whitehead A. 165, 181, 183 Znosko J. 209
Wierzbicka A. 26
INDEX OF SUBJECTS
229
INDEX OF SUBJECTS 230
Volumes 1-8 previously published under the Series Title: Melbourne 1nternational Philosophy Series.
Nijhoff International Philosophy Series
17. i. Dilman (ed.): Philosophy and Life. Essays on John Wisdom. 1984
ISBN 90-247-2996-3
18. J. J. Russell: Analysis and Dialectic. Studies in the Logic of Foundation
Problems. 1984 ISBN 90-247-2990-4
19. G. Currie and A. Musgrave (eds.): Popper and the Human Sciences. 1985
ISBN Hb 90-247-2998-X; Pb 90-247-3141-0
20. C. D. Broad: Ethics. Lectures given at Cambridge during the Period 1933-34
to 1952-53. Edited by C. Lewy. 1985 ISBN 90-247-3088-0
21. D.A.J. Seargent: Plurality and Continuity. An Essay in G.P. Stout's Theory of
Universals. 1985 ISBN 90-247-3185-2
22. J.E. Atwell: Ends and Principles in Kant's Moral Thought. 1986
ISBN 90-247-3167-4
23. J. Agassi, and I.C. Jarvie (eds.): Rationality: The Critical View. 1987
ISBN Hb 90-247-3275-1; Pb 90-247-3455-X
24. J.T.J. Srzednicki and Z. Stachniak: S. LeSniewski's Lecture Notes in Logic.
1988 ISBN 90-247-3416-9
For 'The Lesniewski's Collection' see also Volume 13.
25. B.M. Taylor (ed.): Michael Dummett. Contributions to Philosophy. 1987
ISBN 90-247-3463-0
26. A. Z. Bar-On: The Categories and the Principle of Coherence. Whitehead's
Theory of Categories in Historical Perspective. 1987 ISBN 90-247-3478-9
27. B. Dziemidok and P. McCormick (eds.): On the Aesthetics of Roman Ingarden.
Interpretations and Assessments. 1989 ISBN 0-7923-0071-8
28. J.T.J. Srzednicki (ed.): Stephan Korner - Philosophical Analysis and
Reconstruction. Contributions to Philosophy. 1987 ISBN 90-247-3543-2
29. P. Brentano: On the Existence of God. Lectures given at the Universities of
Wiirzburg and Vienna (1868-1891). Edited and translated from German by
Susan P. Krantz. 1987 ISBN 90-247-3538-6
30. Z. Augustynek: Time. Past. Present and Future. Essays dedicated to Henryk
Mehlberg. 1990 (in preparation) ISBN 0-7923-0270-2
31. T. Pawlowski: Aesthetic Values. 1989 ISBN 0-7923-0418-7
32. M. Ruse (ed.): What the Philosophy of Biology Is. Essays dedicated to David
Hull. 1989 ISBN 90-247-3778-8
33. J. Young: Willing and Unwilling. A Study in the Philosophy of Arthur
Schopenhauer. 1987 ISBN 90-247-3556-4
34. T. Z. Lavine and V. Tejera (eds.): History and Anti-History in Philosophy.
1989 ISBN 0-7923-0455-1
Nijhoff International Philosophy Series