Anda di halaman 1dari 420

UNIVERSITY

OF FLORIDA

LIBRARIES

Digitized by the Internet Archive


in

2010 with funding from

Lyrasis IVIembers

and Sloan Foundation

http://www.archive.org/details/introductiontoseOOscha

INTRODUCTION TO SEMANTICS

INTRODUCTION
TO SEMANTICS
by

Adam

Schaff

Translated from Polish by

OLGIERD WOJTASIEWICZ

Pergamon

Press

Book

THE MACMILLAN COMPANY


NEW YORK

PERGAMON PRESS
122 East 55th Street,

1404

New York Avenue

Headington

&

New York

Rue des

N.Y.
5.

D. C.

PRESS LTD.
Oxford.

Hill,

London W.

5 Fitzroy Square,

PERGAMON

PRESS

1.

S.A.R.L.

Ecoles,

Paris

PERGAMON PRESS

G. m.

24,

22,

N. W., Washington

PERGAMON
4

INC.

Kaiserstrasse 75, Frankfurt

V*.

b.

H.

am Main.

Copyright
1962

by Paiistwowe Wydawnictwo

Naukowc

Warszawa
Library of Congress Card

Number

61

18879

Printed in Poland

by Wrociawska Drukarnia Naukowa

CONTENTS
Page

IX

Preface

Part

One

I^SEARCH PROBLEMS OF SEMANTICS


Chapter One. Linguistics

Chapter Two. Logic

22

Chapter Three. Semantic Philosophy


1

The

53

alleged "turning point" in philosophy.

Language as the only

subject matter of research


2.

Language as a product of

59
76

arbitrary convention

90

Chapter Four. General Semantics

Two

Part

SELECTED PROBLEMS OF SEMANTICS

Chapter One. The Philosophical Aspect of the Communication Process


1.

f2.
3.

The essence of

the

113

problem of communication

The controversy between

the transcendentalist

118

and the

naturalist

130

conception

Foundations of a Marxist formulation of the problem

Chapter Two. The Sign: Analysis and Typology


1.

The communication process

as

the

starting

140
155

point in the

analysis of the sign

155

2.

Husserl's tentative typology of signs

162

3.

Definition of the sign

173

4.

General foundations of the typology of signs

176

5.

A. Signals

181

B. Substitutive signs

185

The

194

specific

nature of verbal signs

fv]

Contents

VI

Chapter Three. The Meanings of "Meaning"


1.

On

2.

Meaning

212
215

the sign-situation
as a real or ideal object

228

A. "Meaning" as the object denoted


B.

The conception of

intentional meanings

3.

Meaning

as a relation (1)

4.

Meaning

as a relation (2)

230

247
263

A. Meaning as a relation between men

B.

228

who communicate

with one another

264

The

268

origin of

meaning

C. Meaning and notion

274

D. The mechanism of the


5.

The

linguistic

approach

links between sign

to the issue of

and meaning

meaning

Chapter Four. The Communicative Function of Language

295
301

311

1.

Language and "languages"

312

2.

Language and

reality

321

3.

Language and

effective

Bibliography

Index of

Names

cominunication

352
366
391

THE AUTHOR'S FOREWORD


On

submitting this book to the readers I wish to thank all those

who have helped me

to bring

I wish

of
who had been kind enough
First

all,

and penetrating

to

to its present

it

form.

thank Professor Tadeusz Kotarbinski,


to

read the manuscript. His profound

criticism helped

me

to

add

many

pas-

Wojtasiewicz

and

precision to

sages of this book.


also

Mr

thank

the

translator

Dr

Olgierd

George Bidwell who checked the translation.


Quotations from non-English-language books are either given

in existing

In

refers

due

English version, if available, or translated into English.

three cases only they are given in the original version. This

to

to quotations

from

Husserl, Jaspers,

the specific character

of these

and

Vossler,

and

texts.

A. S.

[vi.]

is

PREFACE

Authors'

prefaces

tend

my

case,

paid to usage. In

explain to the reader

book. Hence

my

to

be

mere conventions, tributes


I do really feel a need to

however,

intentions before he starts to read this

regard this Preface not as an appendix

to,

but

component part of, my work.


The point is that the subject matter of my interests is extremely comphcated and very vast. I lack, unfortunately, such
competence and erudition as would be necessary adequately
as a

to treat
that

all

aspects of that subject matter; indeed,

no individual can master that

satisfactory degree.
cases,

it is

to be feared

subject matter in a completely

Consequently, here too, as in

many

other

a conscious restriction of research interests should be the

I do not know whether I can satisfy that requirement properly, but I should like to explain what tasks I have
set myself, and why. It is up to the reader to decide whether,
and to what extent, I have achieved them.
First of all, I wish to explain, both to myself and to others,
what we mean by semantics. As a discipline, semantics is now
so intricate, and its name so ambiguous, that semantic analysis
must be applied to the very term "semantics", if deplorable misunderstandings and errors of logic are to be avoided.

token of mastery.

Such

is

the objective aspect, not to mention the attitude

towards semantics adopted in Marxist


present author

would

like to

literature.

And

yet the

approach and handle the issue of

semantics from the Marxist point of view, both in the sense of


making Marxist philosophy assimilate as far as possible the
real research problems of semantics, and of subjecting to critic-

ism any possible philosophical abuses of such problems.

[ix]

Preface

In

present form of a logical and philosophical theory,

its

semantics

a comparatively young discipline. Setting aside the

is

embryonic form of
ed at the time of
in Lenin's time,

no such discipline existwas already taking shape

linguistic semantics,

Marx and

but

its

Engels. It

philosophical implications were revealed

only towards the end of his

when he was

life,

otherwise absorbed.

Even an incurable "quotation maniac" could find no authoriMarxist statement on

tative

in Stahn's

Marxism and

the

emphasized the importance of

warned against abusing

rightly
in

except for a paragraph

issue

this

Problems of Linguistics which rightly

and no less
from adequate

linguistic semantics,

But that

it.

far

is

view of the enormous scope of the problems involved.


In recent years Marxist literature has, of course, included
pubhcations on semantics, but

critical

that they were not

The

philosophical
day,

this

show

it

can be

definitely affirmed

performances.

problem in Marxist encyclo-

brief formulations of the

paedias,
to

creditable

dictionaries,

that semantics

which

etc.,

are

was understood

available
exclusively

and that the latter was


somewhat peculiar interbook The Tyranny of Words (New York

as the idealistic semantic philosophy,


in turn identified with Stuart Chase's

pretation offered in his


1938).

The

credit

literature belongs,

for
it

having introduced Chase into Marxist

seems, to B. Bykhovskyi. Chase has since

then stereotypically haunted

all

Marxist publications in which

the term "semantics" has appeared.


It

is

characteristic that

it

is

not only Marxist publications

which have interpreted semantics as a pseudo-science devised


to blur class struggle, a theory implying that a

mere removal
etc.) would

of certain terms (such as "capitalism", "socialism",


eliminate the corresponding social issues.

pressed in anti-Communist propaganda. In

similar view

his 1984,

is

ex-

Orwell has

a scene in which Syme, the editor of a dictionary of "Newspeak",


explains to
1

Winston that a reduction of the stock of concepts

"MapasM coBpeMCHHOM 6ypHcya3HOH

Contemporary Bourgeois Philosophy],

in

ct"HJioco(})Hn"

EojibuieeuK,

[Marasm

1947, Xe. 16.

in

Preface

xi

can eliminate dangerous social issues which

will

become

in-

conceivable after an appropriate reform of vocabulary.

do not mean that there

is

absolutely

no

ascribing all these absurd ideas to semantics.


is

supported by the opinions of

circles close to

be discussed later on in this book. But

will

"in general"? Is

it

fair to identify

semantics and thereby deny to


see

my

first

it

justification for

Such an attitude
is

Korzybski, as
that semantics

such opinions integrally with


all scientific

significance?

task in answering these questions. Both the

various fields concerned with semantics, and the various meanings of the
will

term must be reviewed.

A sui generis semantic analysis

be devoted to explaining the ambiguities of the term

Thus an inquiry must be made

into

what

is

itself.

the significance and

the subject matter of semantics as a branch of linguistics; what


is

the

difference

between

semantics

linguistic

and semantics

connected with logic, semantics born of the specific needs of the


disciphne which

has

shown

language

is

not only

an instrument but also an object of research; what

is

semantics

latter

as the specific philosophical trend

that

which sees in language treated

as a convention the sole subject matter of research (so-called

semantic philosophy)

and

finally,

what

of so-called general semantics which

is

for

semantics in the sense


all its

pecuUarities

approaches the issue rather from the social and sociological


point of view.

Of course, my

object

analysis, to extricate

by the

single

the actual

is

not only to give a passionless semantic

and to present

term "semantics".

all

I shall

the meanings concealed


also endeavour to

real scientific

problems

it

formulates.

Such an approach does

not, obviously, exclude a critical appraisal. Yet, in

my

opinion,

pride of place should go to a clear understanding as to


are the

show

spheres of interest of semantics and to clarify the

new and

what

philosophically significant issues raised by seman-

tics.

believe the emphasis

on the philosophical

significance of

these issues to be extremely important in view of the wide dia-

Preface

xii

pason of the problems covered by the term "semantics". As

al-

ready indicated, such problems include both specifically hnguistic


matters and specifically logical issues, connected with the tech-

nique of logical calculus. Such problems will be dealt with in this

book only from


we

tions, since

the point of view of their philosophical imphcaare concerned with the philosophical aspect of

semantics and the philosophical problems

of that disciphne.

Thus, the problems will be treated broadly, and no absolute

importance

will

be ascribed to

sided, analyses, as

is

partial,

and consequently one-

sometimes done by exponents of formal

objections against such a procedure are raised even by

logic:

people of the same "profession", such as Russell and Wittgenstein.


In this connection, the very

title

of the present work,

Intro-

duction to Semantics, so fondly used by those semanticists

who

on a deliberately provocative tone.


Starting from these assumptions, and basing myself on the

represent formal logic, takes

results of the analysis to


I shall

is

is

in the first part

of the book.

endeavour to expose a number of main problems and

them

solve

be made

effectively in the fight

to

of Marxist philosophy. That

my second task, to be dealt with in the second part of the


Such is my programme and such are my intentions.

book.

There

a strong impulse to point out changes in the method of present-

ing and criticizing non-Marxist trends in philosophy, by Marxists,

and to analyse the sociological foundations of the

distortions

revealed in that respect in Marxist literature. Yet perhaps sufficient

lates

has been written on that point in Poland, enough postu-

have been formulated, maybe, and declarations made. More-

over, to promise too

much at the outset

involves the risk of disap-

pointing the reader. Consequently, such matters are to be handled

way simply by giving effect, as far as my abihand competences permit, to the notion of a scientific analysis

in a different
ties

of the various trends and

issues.

Part One

RESEARCH PROBLEMS OF SEMANTICS

Chaptek One

LINGUISTICS
Semantics (semasiology)
with what

does

a branch of

is

is

found in contemporary

To begin with
French

The ques-

and in what
and the semantic problems

that branch of linguistics concerned,

see the distinction between itself

it

linguistics.

interest in this connection are

which are of particular

tions

logic^.

the term

BreaP and

linguist

is

itself:

it

comes from the eminent

genetically connected with linguistics.

In the late 19th century Michel Breal published his Essai de


semantique.

Science des significations, in which he writes:

"My intention was to


division and, as

give a general outline, to sketch a general

were, a provisional plan of a

it

domain

that

has not been studied so far and which should be the result of

work

many

for

requested to
science

consider this

which

The reader is therefore


book a simple introduction to the

generations of linguists.

propose to

call semantics".

In a footnote, Breal explains the meaning of the term "semantics":

Tsxvv)

"E7)[i,avTt.x7)

word

o7)[Jt.aLvco

the

'denote', as

science

of meanings, from the

opposed to phonetics, the science

of speech sounds"3.

An

its

outline of the history of semasiology

interests

bolischen Fonnen, Pt. I:

"Die Sprache" BerUn 1923; H. Kronasser, Handbuch

der Semasiologie, Heidelberg 1952; S.

Oxford

and the varying aspects of

can be found in such works as: E. Cassirer, Philosophie der sym-

1957;

B.

A.

SBcrHHueB,

UUman, The

Ce^^acuoJlo^uH

Principles of Semantics,

[Semasiology],

MocKBa

1957.
2

Kronasser (op.

cit.,

p.

29),

who

denies that,

comes out against the

generally adopted opinion.


3

M.

Br6al, Essai de semantique. Science des significations, Paris 1904

p. 8.

\i\

Research Problems of Semantics

the

What
new

did Breal imply by that term, and


discipline

defined

by the

how

in general

was

linguists?

Breal defined the subject matter of semantics by detaihng

appropriate research which ought to be undertaken.

"In that second part

we propose

to investigate

how

it

hap-

pens that words, once created and endowed with a certain meaning,
extend that meaning or contract

it,

of notions on to another, raise

value or lower

bring about changes.


semantics,

It

is

its

transfer

it

from one group

second part

this

it,

in a

that

word

constitutes

science of meaning"^.

i.e.

This shows that, for Breal, semantics was the science the
subject matter of

which was study of the cause and structure

of the processes of changes

meanings of words: expansion

in

and contraction of meanings, transfer of meanings, elevation


and degradation of their value, etc.
Such a delineation of semantics as a branch of
is

maintained to

this day, for all the differences

schools in linguistics. Such degree of uniformity


to the definition of semantics alone^.

Not

linguistics

between the various

all

is

not confined

authors give such

a definition; some of them approach the issue from a different point


of view as regards general classification

(e.g.,

de Saussure,

who

develops the concept of semiology as the science which studies


the functioning of signs in society,

of such a general science of

and treats

signs),

but

all

linguistics as a

branch

schools of linguistics

engage in the study of the meanings of words and their changes.

Thus

all

of them, in one way or another, engage in semantics

by Breal.
For instance, Darmesteter conceives of the science of the
meaning of words, i.e. semantics, as a branch of the history
as understood

of psychology, but appreciates

1948,

its

responsibilities in the investi-

Ibid., p. 99.

Cf. A.

No.

W. Read,

4, pp.

78-97.

"'An

Account of the Word 'Semantics'",

in

Word,

Linguistics

gation of the history of meanings and the causes of their changes^.

Vendryes disagrees with Darmesteter and denies that the general


laws of the evolution of meanings of words should be inherent
in

words themselves
problems of

al

"?.

He

too, however, engages in the tradition-

linguistic semantics,

and even postulates the

foundation of general semantics, to be based on data concerning


changes in meanings in

all

languages 8, In Polish linguistic

erature, similar issues are dealt with

lit-

by Zenon Klemensiewicz^.

In fact, the majority of works on general linguistics treat semantic

problems as being their focal problems.

I shall later

endeavour

what the general programme of linsemantics stands for. But at this point I should like to

to analyse in greater detail


guistics

on certain general formulations.


The fact that the general definition given in a new textbook
based on Marxist principles does not in any degree deviate from
the old definition as formulated by Breal, shows that the traditional meaning of semantics has already stabilized. An Introdwell

duction to Linguistics

by L. Bulakhovsky begins with the

defini-

tion of semantics as one of the important branches of linguistics

"Semantics (semasiology) as a branch of


cerned with the meaning and the changes

and

in

linguistics

is

con-

tneaning of words

expressions''^^.

All this

on general

is

not to imply that, within that universal agreement

definition, there are

no

differences of opinion, in

some

cases quite considerable, as to the essence of the problem. Nei-

mean

ther does

it

definition

and

<5

that linguists rest content with such a general

direct their controversies

A, Darmesteter, La

vie

and

differences of opinion

des mots etudiee dans leurs significations, Paris

(without date).
7 J,

Vendryes, Le language, Paris 1950, pp. 228-229.

8 Ibid.,

' Z.

pp. 240-241.

Klemensiewicz, J^zyk polski [The Polish Language],

Lw6w-War-

szawa 1937 pp. 10-14, 22-24, et passim.


10 Jl. A. ByjiaxoBCKHM, Beedenue e H3biK03HaHue [An Introduction to
Lingmstics],

MocKBa

1953, Pt. 2, p.

7.

Research Problems of Semantics

'6

to the field of special issues concerning the essence of meaning,

and concrete forms of such

the causes of changes in meanings,

changes.

For instance, the attitude of Witold Doroszewski has obvious


philosophical implications.

For him,

root

the

at

of semantic

analysis lies the philosophical issue of the relationship between

the general

and the

particular, the starting point being the anal-

of the function of the copula

ysis

Doroszewski analyses

"is".

the problem of meaning as closely linked with denotation.


is

It

in that question that he sees the focal point of semantics.

"The potential

in every

inherent

conflict

embodiment of a general concept,

ual

understood as a part of

tics

is

word, and finding

word

expression in the fact that the use of every

is

an

individ-

the focal point of seman-

linguistics

that

is

as a science

of the meanings of words and the history of such meanings"' i.

According to Doroszewski, the history of meaning consists

growth of a "gap" between the sign and its designatum.


and the cause of changes in meanings lies in the conflict between
the general character of the sign and the need for its being made
to rise to the occasion whenever it is concretely embodied.
The attitude of Doroszewski, a representative of linguistic
in the

semantics,

is

of interest not only because he draws concrete

conclusions from the general definition of semantics, but also

because of his reflections on the relationship between semantics


as pursued by linguists, and semantics as pursued by logicians.

By touching here upon

that issue,

anticipate further analysis.

This does not advance the clarity of exposition, but

is,

tunately, often inevitable. In

in so far

as

it

tliis

case,

is

it

justified

helps us to realize better the specific traits of linguistic se-

mantics and

A
guists,

its

research objectives.

colloquium on semantics, attended by logicians and

was held

in

Warsaw

in 1955.

The

object

was

the gap between certain points of view, and to agree


'1

unfor-

W. Doroszewski,

zagadiiien leksykografii polskiej

lems of Polish Lexicography],

Warszawa

1954,

p.

93.

to

on the

lin-

narrow
subject

[Selected Prob-

Linguistics

matter of future research. The discussion begot Doroszewski's

"Comments on Semantics", pubhshed

paper

Mysl

in

Filozo-

There, he defined semantics as "the science of the meanings

ficzna}'^.

of words", the central issue of which

liimself against

is

"the problem of the rela-

words and designata". Doroszewski declared

tionship between

psychologism in the interpretation of semantics

and referred to the objective hisiory of words, as linked with


the fact that a given word reflects a fragment of reality. For
should approach the meanings of words

this reason, the linguist

from the

and

social

Doroszewski

that

historical point of view. It

sees

between

distinction

is

in that issue

linguistics

and

logical semantics.

of

"Not only the

tools of

human thought

words

manual labour, but

development. The history of the meanings of words

ical

side the area of interest of


fully studied

formal

it

logic,

by the methods of that

"The history of language


is

also the tools

are subject to the laws of histor-

in

what

is

and could not be

outfruit-

discipline.

is its

most

essential content

the history of language as a social instrument of thought;


historical epistemology

is

which cannot be studied within the

scope of any other discipline.

"The

linguist

is

of necessity only marginally interested in

conventional terminology,
logic

are

linguistics

inchned to

investigate

and even

some

to

all

whereas certain votaries of formal

domains which are

alien to

extent in contradiction to

its

basic

assumptions"!^.

12

in

W. Doroszewski, "Uwagi o semantyce" [Comments on

Mysl
13

Filozoficzna, 1955,

That

No.

Semantics],

3 (17).

confirmed by the standpoint adopted by the admethod in semantics, and expressed by them explicitly
e.g., F. R. Blake in "The Study of Language from Semantic Point of View"
in Indogermanische Forschungen, 1938, 56, p. 242 (quoted by 3BerHHueB, op.
Ibid.

thesis is

herents of the logical

cit.,

in

p.

12), says that as

long as semantics engages in the study of changes

meanings of words at the various stages of

their existence,

historical semantics. It is necessary to originate static semantics,

it

is

in fact

which would

RKStARcn Problems

Thus

Semantics

oi

the characteristic feature of linguistic semantics

deals with meanings

and relationships

that

it

and

their designata, but that

as between

not

words

concerned with the history

is

it

is

of meanings, their origins, their changes, and the laws according


to

which such changes occur. This focuses the

specific feature

of linguistic semantics.

While laying

stress,

in

studies,

its

on the

of hnguistic entities and their meanings,

historical aspect

linguistic

semantics

(semasiology) does not overlook the other, systematic, aspect

of the problem. The division into diachronic and synchronic

by de Saussure, is being continued by approwhich in its developed form becomes the theory

analysis, initiated

priate research

of the semantic domain, to be found in certain linguistic schools


(e.g.,

Jost Trier)i4.

Let us consider

now what

guistic semantics so

in the Preface,

we

research problems arise from

Remember

interpreted.

are interested above

all in

that,

lin-

as indicated

those problems which

have philosophical implications and are in one way or another


dealt with

by the various schools of pliilosophy.

Linguistics

is

devoted to the study, in some way or another,

of language, linguistic expressions, and their meanings.

When

approaching relationship between

linguistic

designata they denote,

semantics comes up against

the
tics

problem of

may

been, to

linguistic

hnguist engaged in the study of seman-

not disregard the theory of signs. That theory has

some

interests.

signs.

expressions and the

extent, for a long time within the sphere of linguists'

For example, Wilhelm von Humboldt raised the ques-

tion as early as 1820,

when he made a

as a reflection of reality

distinction between language

and language as

a system

analyse and systematize the meanings that always exisi

in

the

of signs^s.

minds of users

of language, and would study pure meanings regardless of their form and
development.
'4

On

that point

IS

W.

V.

cf.

Ullman, op.

Humboldt, Vber das

cit.,

p.

152

ft'.

veigleicheiule SpiaclisiiuUum"

gaben der "Philosophischen Bibliothek",

1,

p.

Taschenaus-

22.

Linguistics

^)

The problem of the sign often appeared subsequently also, especially in works on the philosophy and psychology of language^^.
of place

Pride

that

in

respect

Saussure, both in view of

of

work of de

by the

held

is

linguistic

its

character and because

wide theoretical horizon. At the time when semiotics as

its

a general theory of signs has acquired such importance in the


literature

of the subject, attention ought to be directed to the

De

of the problem.

linguistic aspect

Saussure wrote on this mat-

ter as follows:

"Thus we may found the science for the study of the


background of

signs against the

social life;

it

life

of

would form part

of social psychology, and consequently of general psychology;

we

shall call

science

it

semiology (from Greek asfxetov

would explain

to us in

laws they are governed. Since

is

it is

existence,

its

place

its

That

'sign').

and by what

a science which does not yet


it

has, however, a reason

allocated in advance. Linguistics

is

only a part of that general science

will discover, will

signs consist of

we do not know what it will be like;

exist,

for

what

be applicable also to

the laws which semiology


linguistics,

which in turn

be linked with a domain clearly defined throughout the

will

human

entirety of

De

affairs"i7,

Saussure openly protested against the interpretation of

his intentions in

terms of individual psychology. The sign should

be analysed as a social phenomenon, and moreover as such


a

phenomenon
'6

as docs not

depend on our

The following works may be

cited

will,

whether

by way of example: K.

indi-

Biihler,

Sprachtheorie, Jena 1934: E. CassLrer, Philosophie der symboUschen Formeti,

and An Essay on Man.

New York

An

Introduction to a Philosophy of

Human

Culture.,

1954; O. Jespersen, Language, Its Nature, Development and Origin,

London 1954; A. Marty, Untersuchungen zur Grundlegung der allgemeinen


Grommatik und Sprachphilosophie, Halle a. S. 1908; A. v. Meinong, Vber Annahmen, Leipzig 1910; C. JI. Py6HHuiTeHH, OcHoeti odufeii ncuxojio2uu
[Principles of

guage,

General Psychology], Chap. XI, MocKBa 1946; E. Sapir, Lan-

New York

1921;

W. Wundt,

Volkerpsychologie, Vols.

&

2,

Die

p.

33.

Sprache, Leipzig 1911-1912.


'^

F.

de Saussure,

Cows

de

linguistique

generate, Paris

1949,

10

Research Problems of Semantics

vidual or social. In this

linguist

proposed to build a gen-

i.e.,

semiology (semiotics), a matter which

different point of

view was approached by logicians and

eral theory of signs,

from a

way a

philosophers.
Linguistics as such has not developed semiology, but

indeed developed a theory which

problems

The

the

linguists

of meaningly.

theory
first

has

it

linked directly with linguistic

is

of

all

Answers to that questions

ask themselves what

Some of them

vary.

meaning?

is

are quoted below

by way of example.
In Polish linguistic literature
versy

found

is

between two conceptions:

first

of

associationism

a contro-

all

and the op-

ponents of associationism.
Stanislaw

Szober

meaning

interpreted

as

the

association

of a linguistic representation with an extra-linguistic one

(i.e.,

a sound image with an image of an object or attribute).

of

"The meaning of a word is established by the association


its
sound image with
an image of some object or

attribute ..."19.

Such an interpretation led Szober to the assertion that the

word becomes a sign of an extra-linguistic image^o, in such a


way that it makes possible to include in the content of the
meaning of the word not the full extra-linguistic image, but
only some of its details, that is, a simplified image.

quite

different

opposing

position,

associationism.

was

represented in Polish linguistic literature by Henryk Gaertner,

who engaged
18 Cf.
19 S.

in direct polemics with Szober^i.

G. Stern, Meaning and Change of Meaning. Goteborg 1931.


Szober, Zarys j^zykoznawstwa ogolnego [An Outline of General

Linguistics],

1.,

Warszawa

1924, p.

5.

20 Ibid., p. 6.
21

H. Gaertner, Gramatyka wspolczesnego jqzyka polskiego [A Grammar

of Contemporary Polish Language],


of Szober's

associationism

is

to be

Pt. 2,

Lwow

found

in

1933, pp. 96

an

article

flf.

criticism

by M. Ossowska,

"Semantyka profesora Szobera" [Professor Szober's Semantics].

11

Linguistics

should like

now

pay more attention

to

to the interpretations

of meaning by de Saussure and by Bulakhovsky,


these two authors because of the differences

towards the

have chosen

in their attitudes

issue.

With de Saussure,

concept of meaning

the

is

inseparably

connected with his concepts of sign and language. The problem


can, of course, only be introduced here;

greater detail in the

According to de Saussure, a

will

linguistic sign is

with two aspects: sound image


is

it

be analysed in

second part of the book.


a psychic whole

Thus the sign

and notion22.

a specific combination of these two elements. Following from

de

Saussure's

analysis,

sound and notion

is

that

meaning. This

is

between

relationship

why de

Saussure suggests

term "sound image" should be replaced by the term

that the

"signifiant"

(that

which means), and the term "notion" by the

term "signifie" (that which

is

the object of meaning).

The

sign

function only by virtue of that relationship of meaning,

fulfils its

the

two-sided

members of which

are connected inseparably.

The breaking

of that unity would result in the destruction of the sign.

But not of the sign alone.


is

a system of signs.

relationship of
sign,

fully

is

and sound

is

What

It also refers to

language, which

has been said here concerning the

meaning which conditions the existence of the


apphcable to the relationship between thought

in language.

"Language can be compared to a sheet of paper: thought


its recto and sound its verso\ one cannot cut the verso without

simultaneously cutting the recto.


language,

Similarly,

in

the

matter of

one can separate neither sound from thought nor

thought from sound; such separation could be achieved only

by abstraction, which would lead either to pure psychology,


or to pure phonology "23.
It

is

interesting to

observe the contrast between the views

of Bulakhovsky and of de Saussure on the problem of meaning.


22 F.
23

de Saussure, op.

Ibid., p.

157.

cit.,

p.

99.

12

It

Research F*roblems of Semantics

is

interesting because

what

is

it

not only enables us to understand

the subject matter of linguistic semantics, but

shows us the

possibility of

ferent

ways which depend, among other

ual's

philosophical and

here two scholars

who

it

also

approaching the same issue in


things,

on the

We

background.

methodological

dif-

individ-

are interested in the problem of

see

meaning

from the linguistic point of view and in the light of linguistic


method. And yet both of them resort to a definite philosophical

and methodological background much broader


Saussure

is

de

kind:

in

obviously influenced by the concepts of Durkheim,

and Bulakhovsky by those of Marx. Hence the


opinion concerning the same research problems.
Bulakhovsky

opposes

the

association

difference

which

theory,

of

sees

the essence of meaning in the association between the representation

and the sound aspect of the word. The function of meaning

him inseparably connected with the function of denotaA word above all denotes some real fact or phenomenon
about which the individual wants to communicate something
to others. And meaning is the content of the word, revealed by
connections with reality. The proper meaning of a word is shaped
is

for

tion'^^.

by the history of

its

connections with

reality^s.

Let us extract from that statement what


to the conceptions analysed above.
respect: emphasis

Two

is

new

in relation

points emerge in that

on the connection between sound image and

a fragment of reality (understood as the world existing objec-

which appears

tively),

on the

in the function of denotation,

role of the history of a given

word

in

and emphasis

determining

its

actual

meaning.

To be exact, I must add that this is by no means


among the Soviet linguists. For instance A.

attitude

rejects the

opinion which holds that meaning

is

the dominant
I.

Smirnitsky

connected with

the relationship between sound image and the denoted fragment

24 ByjiaxoBCKHM, op. cit. pp.


25 Ibid., p.

13.

12-13.

13

LrNGUlSTICS

of reality, and understands by meaning


tation (reproductive or productive)

tiie

concept or the represen-

which

reflects the

of reality in question. Thus, for him, meaning


link

between sound and

must

the reservation that

confine myself

here to information; the lack of a critical appraisal


interpreted as indicating that
I

fragment

the intermediary

reality^6,

make

explicitly

is

am

in

is

not to be

agreement with the opinions

record.

In connection with the problem of meaning, the linguists


also deal, within the scope of their semantic pursuits, with certain

derivative issues.

What

have meaning as their attribute?

linguistic entities

That problem

analysed by, for example, Vendryes

is

when

he introduces a diiferentiation as between the word, the semante-

morpheme. He understands by semantemes "hnwhich express ideas of representations {les idees


des representations)", and by morphemes, such linguistic elements
as "express the relationships which human mind establishes
between semantemes"^?; he endeavours to define what is to be
understood by the word and by the expression's. This is a lin-

me and

the

guistic elements

issue

guistic

of extreme significance, with a semantic aspect.

Vendryes' view has been quoted here just by

because the problem

is

often to be

met with

way of example

in linguistic litera-

ture.

The same holds

for the differentiation of

meanings and

at-

tempts to classify them. For instance, Kurylovich distinguishes


general
his

meaning,

numerous

which

an

is

predecessors

and following
meaning (independent

abstraction,

principal

of the context), and speciahzed meaning (with elements of the


context added).

26

separate question

that of caique

meaning

A. M. Cmmphhukmh, "SHaneHHe cjiOBa" [The Meaning of a Word],

in Bonpocbi H3biK03HaHUH, 1955.


27

is

Vendryes, op.

28 Ibid.,

pp.

cit.,

103-104.

p. 86.

JVq

2,

pp.

82-84.

14

Research Problems of Semantics

which develops

actual vocabulary or

the

in

\
from

neologisms

loaded with the content of foreign words29.

The second
in

great issue of linguistic semantics concerns changes

meanings and the causes of such changes.


Linguistics has since

very inception been concerned, in

its

some form or another, with the

and the

history of language

etymology of linguistic expressions. One might even risk the

ment that

it

is

guistics are to

state-

precisely in those problems that the roots of lin-

be found.

And

it is

from such roots that

linguistic

semantics has developed as the science of meanings of words

and of causes of changes


The
ly in

in meanings.

of linguistic semantics, then, consists precise-

specific trait

The causes of the variaby semantics either in language

the study of the history of meanings.

bility

of meanings

are

itself

or in factors

to language.
either

seen

psychological

The laws of

or

sociological

external

that variability are accordingly treated

autonomously or heteronomously. Credit, however, must

be given to the research realism of the linguists

who

in principle

take both factors into account and interpret them as reacting

one upon the other.


Breal explicitly connected the evolution of linguistic meanings

with the evolution of social

life in

the broad sense of the term.

"In modern society, the meaning of words changes much


more quickly than it did in antiquity or even in the recent past.
This arises from the intermingling of social classes, the struggle
of interests and opinions, the struggle of political parties and
the variety of aspirations and tastes

The

social

and

class factor

was

..."30.

stressed probably the

most

strongly by A. Meillet, Breal's successor in the College de France.

In particular, in his work


29

E.

P.

KypHJiOBHM

Meaning of Words],

in

Breal, op.

31

First published in

in the

book by A.

Paris 1948.

les

mots changent de

"SaMCTKH o 3HaHeHHH

Bonpocti

.H3biK03Hanu.H,

cjiOBa"
1955.

sens^^.

[Remarks
3,

pp.

on

78-79.

pp. 105-106.

-'0

cit.

Comment

Annee Sociologique, 1905-1906. and then reprinted

Meillet, Linguistique

historique et linguistique generate

16

Linguistics

Meillet

emphasized the co-operation of three factors

velopment of meanings

of words:

autonomous

language in question, the development of the


class (group) factor. In conclusion

and the
"...

Changes

in the de-

laws of the

object

denoted,

he says:

meanings must be treated as phenomena

in

which

the basic condition of

is

differentiation in the

component

elements of society"-^2.

Vendryes,

who

autonomous laws gov-

studied in detail the

erning changes in meanings, protested, as

Similarly, Klemensiewicz,

from the point of view of


importance

the

The

factors

reflected in

ing dead

we have

seen, against

words themselves33.

these laws being treated as inherent in

who investigates changes in meanings


laws,

linguistic

of the extra-linguistic

operating are, above

all,

does not overlook

factor in that

changes in

respect.

human

life,

language as the disappearance of certain words (denotinstitutions

and forms of

life)

and

as formation of

new institutions and forms of life). The


same factors also account for changes in meaning of existing

neologisms (denoting

words

(cf.

the

common term "pen" for


common term "lamp"

tain pen, or the

a quill pen and for a founfor

lamps). Further, the psychological factor

oil,
is

petrol,

and

electric

also involved: this

concerns the emotional attitude of the speaker, with consequent


variations in the emotional shade of
result,

we avoid certain words by

meaning of a word. As a
them euphemisms

substituting for

(which later in turn also deteriorate); in certain other cases, we


look for the extraordinary,

Bulakhovsky,

who

etc.34.

concentrates on the inner problems of

language, also clearly sees and emphasizes

the

the social factor in the development of vocabulary

significance of

and meanings.

"In connection with changes in the social system, with the

development of the various branches of production, with the

32

Meillet, op. cit., p. 271.

53

Vendryes, op.

34

Klemensiewicz,

cit.,

op.

pp. 229-230.
cit.,

p.

21

ff.

16

Research Problems of Semantics

and

development of material
technology,

science,

art

disappearance of

many

changes

with

relations with other countries,

culture

spiritual

we

in

notice,

words, which lose

the

in

general

character

of

on the one hand, the


meanings as the con-

and on the other, much


more frequently, the emergence of word-signs used to denote
new ideas and concepts, worked out in practice by the users of
cepts they denote pass out of circulation,

the given language"35.

The understanding of

the

and

social

historical

character

of linguistics, expressed in von Humboldt's lecture quoted above,


is

also clear in the

let,
it

works of such scholars as de Saussure, Meil-

Vendryes, Marcel Cohen, and others. As regards semantics,

appears with particular force.

Even

Breal,

when

explaining his theses by

of the etymology of names of

"We must

Roman

realize the extent to

which

knowledge of language be based on

way of the example

magistrates, wrote:
it is

necessary that our

history.

Only history can

impart to words that degree of precision which we need in order

them well''^^.
Hence he concluded that semantics belongs

to understand

to

historical

disciplines.

"All the more, therefore, should semantics be included


the historical sciences.

words and no

among

There are no changes in meanings of

peculiarities of syntax

which can be regarded as

anything other than minor historical events"37.

From

that point of view, the study of languages of primitive

peoples must also be viewed as a contribution to a historical

35

BynaxoBCKHH, op.

cit.,

36 Breal, op. cit., p.

112.

37 Ibid., p.
38

p.

88.

256.

L. Levy-Bruhl, Les fonctions mentales dans

les

societes

inferieures.

and Race, Language


and Culture, New York 1949; B. Malinowski, The Problem of Meaning in
Primitive Languages (as an Appendix in C. K. Ogden & I. A. Richards, The
Paris 1912; F. Boas, Kiiltur unci Rasse, Leipzig 1914,

Meaning of Meaning, London

1953).

17

Linguistics

interpretation of semantics. I refer here to the

scholars as Levy-Bruhl, Boas, Malinowski


abstract here

like to

the

which

at

others^s. I should

from such controversial

issues as that of

character

prelogical

works of such

and

of the thinking of primitive people,

one time evoked a wave of accusations by Soviet authors

(accusations which, in

understanding)

my

opinion, were largely based

and

Levy-Bruhl

others

were

on a mis-

charged

succumbing to the pressure of raciahsm, imperialism,

with
etc.39.

Certain results of the relative researches are beyond contention,


irrespective of this or that

interpretation:

those re-

I refer to

searches which indicate a greater concreteness in primitive lan-

guages, the inability of such languages to express general notions,

and the mode of


would be difficult to

and the connection between those

facts

hfe,

and the needs, of primitive peoples.

It

find

more convincing demonstration of the historical character


of vocabulary and meaning.
A separate chapter on this subject was provided by the historical semantics of N. I. Marr and his school. It has never reached
the linguists in the West, and in the Soviet Union it was annihilated following a discussion in the late 1940's. The manner in
which Marr's theory was then rejected must be considered detrimental to science; such methods in scientific discussions must
be regarded as inadmissible, in the same way as the previous
a

sanctification of that very theory as the only true

one

is

inadmissible. It

and in particular
contrast

many

to

the

fantastic

is

and correct

being said at present that Marr's theory,

his analysis

based on four elements


comparative

historical

elements.

It

is

method

in

maintained that

also

sharp

contained

Marr's

concept of the class character of language, and the development


of language by stages, as well as his concept that languages developed

39

only by crossing,

DlenflKHH

<t>.

constituted

co4)HH,

in

0u/toco(/)CKiie

MocKBa

vulgarization

H., "Teopnji JleBii-Bpiojia Ha cjay)K6e

lecKoa peaKUHH" [Levy-Bruhl's Theory


action],

1950, p.

janucKu,

148-175.

Bbiu.

in
5,

the

of the

wMnepHajiHCTH-

Service of Imperialist

AH CCCP,

HHCTHxyT

Re-

cJphjio-

18

Research Problems of Semantics

Marxist theory. This

is

a tenable objection. But

Man's

theory,

even apart from concrete results in the study of Caucassian


languages, included, beyond doubt,

many

interesting

and valu-

able ideas of general theoretical significance. I refer in particular

"manual" language as the proto-language,


and the related hypothesis of the development of human speech

to the concept of the

from the

pictorial concrete to the abstract. Further, I refer to the

hypothesis

connecting

development

the

of language-thinking

with the process of production. These are issues of no small

importance for a historical interpretation of semantics, and they


cannot be rejected without adequate arguments.

guments were not always to be found


said

and wrote, even

in

And

such ar-

what Marr's opponents


more detailed

as regards the criticism of the

theses of his linguistic theory^o.

As

already indicated, concern for the historical and social

factor in the development of

meaning

companied by the study of the inner

is

with

all linguists

linguistic laws

ac-

of such de-

velopment.
If

we

again, by

way of example,

refer to the textbooks of

Vendryes and Bulakhovsky, we find that the course of reasoning


is

in principle the

same

fessional analysis, to be

in

both

met with

cases. Since this involves pro-

in all textbooks,

copiously with examples drawn from


the generalization

40 Cf.

B.

and seasoned

languages,

quote

attempted by Vendryes.

"The various changes


are sometimes

many

in

meanings to which words are

subject,

reduced to three basic types: contraction, ex-

B. BHHorpaflOB

"CBoSojlHaa

BonpocaM a3biK03HaHHii H ee SHaiemie ana


BexcKOM HayKH o flSbiKe"

[A Free Discussion

AHCKycciia b 'IlpaBAe'
^aJibHeflraero
in

'Pravda'

no

pasBHTHH co-

on

Linguistic

Problems] in Bonpocbi H3bmo3HaHu.H e ceeme mpydoe H. B. CmaiuHO [Linguis-

Problems in the Light of the Works of J. V. Stalin], MocKsa 1950;


M. PaMHOIBHJIH "HenpHCMJieMOCTb XeOpHH nepBHMHOCTH fl3bIKa )KeCTOB"
[Inacceptability of the Theory of the Priority of a Language of Gestures], in
M3eecmufi AKodeMuu neda20cni'4ecKux HavK PC<t>CP. Bonpocw iicuxoaotic

Jl,.

^uu MbiuiJieHU.H

ii

Pemi, Bun.

81,

MocKsa

1956.

19

Linguistics

pansion, shifting. Contraction occurs

from a general to a

meaning

specific

from a

there

is

a transition

on the contrary, there is a transimeaning (for instance, arrachei\

or traire); expansion, when,


tion

when

(for instance, pondre, sevrer

specific to a general

gagner or triompher); shifting,

when both meanings

either are

equal in scope or in that respect indifferent to one another (for


instance, chercher, choisir or mettre),

one meaning to another


designata

(e.g.,

when

contains to what

is

and when the

meaning of a word
contained, from cause to
the

to designatum, etc., or vice versa). It


tions

shifting

from

caused by the neighbourhood of the

is

is

shifts
effect,

from what

from

sign

obvious that contrac-

and expansions of meanings are in most cases the results


and that such shiftings of meanings take on

of shiftings,
different

forms to which the grammarians give special terms

(metaphor, synecdoche, metonymy, catachresis,

etc.) ..."4i.

These are the problems dealt with by Breal, Darmesteter,


Nyrop42, and also Klemensiewicz43,
sky.

The last-named author has

the subject, supported by extremely

This

is

why

Zvegintsev and Bulakhov-

a very detailed exposition of

numerous

linguistic data^^.

I refer interested readers to his textbook^s.

hke to dwell on the

Finally, I should

linguistic

aspect of

ambiguity of expressions and on the need to be precise with


them. These are matters directly connected with the method
of semantic

analysis,

used in philosophy

a fact

which can

only add to our interest in them.


Breal, the father of hnguistic semantics, also dealt with the

problem of ambiguity and


itself

has a

much

to Aristotle).

It

its

sources (although

was

in this context

that Breal

'!

Vendryes, op.

42

K. Nyrop. Das Leben der Worter, Leipzig 1903.

cit.,

Klemensiewicz, op.

44

ByjiaxoBCKHH, op.

45

Very interesting and comprehensive data

Ullman and

H.

saw the proper

p. 235.

'-^

of S.

the problem

longer history and can be traced back at least

cit.,
cit.,

pp.
p.

49

Kronasser,

10-14.
ff.

will

quoted above.

be found in the work

20

Rfsearch Problems

remedy

against

meanings46.

the

The

danger

of

oi

Semantics

the interpretation

slips in

of Vendryes was

attitude

ferentiated between the primary

similar-*?.

He

of
dif-

meaning of a word (independent

of context) and the secondary meaning (bound up with context).

Bulakhovsky, too,
of eliminating

"There
all

concerned with ambiguity and the ways

is

it.

are,

however, domains of thought

science

above

such as find the psychological shade of the word of min-

imum

importance.

ed

the task a purely intellectual

-to

In those domains, everything

reality; that explains


cise is

why

the

(logical)

is

subordinat-

understanding of

work of making speech more

of enormous significance there. That purpose

is

pre-

served

by choice of words with a strictly defined scope of application,


words which have each only one specified, especially reserved

meaning such as

no additional

in principle evokes

associations.

what are called terms that are such precise ... words, defined
become directly conventional"48.
Thus linguistic semantics appeals for making speech more
precise, which is the main methodological postulate of semantic
analysis. But the linguists also observe the danger entailed in
It is

so as to

abuse of that method, and warn against


"In the

overwhelming majority of

between the word and

meaning

its

is

word

for

the

connection

purely external, and yet as a

result of frequent use that connection

are inclined to take the

it.

cases,

its

becomes so

close that

meaning, and further,

we
fol-

lowing the objectivization of images, to identify words with


the objects they denote.

We are so easy-going as to

accept counters

and symbols of a subjective reflection of reality.


itself. Under the overwhelming influence of language

for real values,

for reality
ideas,

we push

which

link us with the real world, and, having thus

into

*o

Breal, op.

'^

Vendryes, op.

cit.,

48

ByjiaxoBCKM,

op.

cil.,

p.

background the

the

143

ff.

pp. 232-233.
cit.,

p.

22.

extra-linguistic ideas

broken our

21

Linguistics

direct contact with reality,

we

need to

satisfy the

feel that

with sound symbols that lack any intrinsic value.

tact

in this that the great

ever abuses

it

which sooner or

That

is

clarity

enters the unsure road of

it,

just

danger of language-thinking consists: who-

and thoughtlessly blurs the

associated with

It is

con-

later

must lead

to

of mind"49.

sterility

the linguist's warning. Let us bear

of real images

empty verbalism

it

in

mind

in

our

subsequent philosophical considerations.

This concludes our general information concerning the subject


matter of linguistic semantics.

As we have

seen, the subject matter

is

the meanings of words,

changes in meanings, and causes of such changes. The specific


trait

of linguistic semantics consists in the study of the history

of meanings and the historical approach to language.

As regards the detailed spheres of interest of hnguistic semanwe have succeeded in distinguishing the following issues:
the nature and the function of the sign how it is that signs mean
something; ambiguity in the form of homonymy and poliseray
tics,

and the dangers resulting therefrom,

etc.

Certainly, these are not all the fields of linguistic semantics.


It

seems however that they are the principal ones and that to

have distinguished them should prove helpful


considerations.

49 Szober, op.

cit.,

pp. 20-21.

in

our further

Chaptek Two

LOGIC

giving

effect

our intention to

to

by analysis

investigate,

of the various disciplines, what the term "semantics"

and what are


to logic.

for

it is

We

its

thus

research problems,

come

losophy. This

is

means

pass from linguistics

to the focus of the problems involved,

logic which, since the

the field of semantic

we

end of the 19th century, has become

pursuits

most

why we encounter

with phi-

related

closely

here numerous

difficulties,

especially difficulties methodological in nature.


It

is

obvious that historical

significant

philosophical

(linguistic)

imphcations.

It

semantics too has

obvious

is

plainly

philosophical

(which

positions

they

the

up

quite

sometimes

state

various linguists engaging in semantic research take


definite

that

and openly) that make them solve semantic problems

in specific ways.

But in logical semantics the situation

is

dif-

ferent.

In linguistic semantics, we can easily separate purely linguistic


analysis

from

this or that philosophical interpretation. It

in exceptional cases, for

we have

to

do with a

is

only

example in de Saussure's system, that

close, organic linking of the linguistic the-

ory with a philosophical conception, so that the two cannot

be separated mechanically. But what

is

involved here

is

an

entire

system of the science of language, and not merely studies in the


history of meanings.

Now
is

what tends

we disregard
is

to be

an exception in

a rule in logical semantics. This


the purely

is

technical fields

a philosphical discipline incapable

temology and Weltanschauung.

of logical

first

semantics
of

calculi,

all,

if

logic

of separation from epis-

Secondly,

[22]

linguistic

so because,

it

is

so because the

23

Logic

problems of semantics developed

logical

with
as

it

philosophical

definite

were

trends

connection

in a close

within

framework,

their

the most vivid example of which

called logical positivism)

(also

neo-positivism

is

which for the

last

few decades

Hence the sepaphilosophical background

has been the principal promoter of semantics.


ration of logical semantics
is

not only

difficult,

And

yet

that

we can

or

possible,

is

it

from

its

but also has something

artificial

and even necessary. For

it

is

about

only thus

separate real research problems from their better

worse philosophical

shell;

those

problems have

research

emerged from the need to develop formal logic and are

still

we

consider the place

role of formal logic in the system of

human knowledge,

important for that process; therefore,

and the

it.

they are important for the

if

development of science in general.

The doubts which may be expressed about breaking the


hnks between semantics and philosophy and the objection

genetic

may be

which

work

raised that

certain

present ready results of certain mental

abstraction

an

of

instead

position of data, will be refuted by the well


that the exposition of results of research
sitates

ex-

by no means neces-

reproduction of the entire research procedure. In ex-

plaining his

method of presenting

his

economic theory, Marx

maintained that the path of the researcher

and leads through the study of concrete


order,

objective

known argument

which does not mean that

must repeat

is

always empirical

facts in their concrete

in presenting his theory

his entire route in detail.

The

he

starting point of such

a presentation can, and often must, be the result of the abstraction process already performed.

Of

course, this involves various

and inconveniences, but any other procedure would


involve still more of them. In this connection, let me urge that
this chapter be read and reflected upon in close association with
difficulties

the next, philosophical, one.

full

picture can be obtained only

by linking very closely logical problems with

background

which

is

historically

philosophical interpretation.

their philosophical

conditioned

and

their

Research Problems of Semantics

24

Reverting now,
of the problem,

brief digression,

that

after

can be stated that

it

if

to

the

history

by semantics we mean,

most general way, an analysis of the relationships between


language expressions and the objects they denote, or else re-

in a

'

meaning of expressions, then elements of such


a discipline are to be found even in antiquity, especially in the
works of Aristotle. Throughout the whole course of the

flections

on

the

history of philosophy,

we

who

philosophers

various

find

are

interested in these matters. Yet as a separate discipline logical


semantics appears only in the late 19th century, arising out

of quite

difficulties

specific

and research problems

and mathematics. The understanding of


is

not

only

instrument but

an

emerged mainly from the need

overcome

to

well.

Far be

it

from

me

of research

object

difficulties in laying

to claim that that


in

logic

the logical (set theoretical) foundations of arithmetic, difficulties


which threatened the entire conceptual structure of arithmetic

and logic as

ij

the fact that language:

an

also

in

was

i^

j;

the

language by logicians.

only cause of an increased interest


There was another source of that interest, which from experimental disciplines led to conventionahsm, and hence to the
study of language. But the stimuli due to the discovery of antinomies were specific and particularly strong: they were auto-

genous,

and

connected

logic,

with the particular

and independent

of -

even

needs of mathematics
if

accompanied by

any philosophical interpretations.


Bertrand
It all began with a letter from, young then
sell,

to the great Gottlob Frege

just finishing the

at the

second volume of

time

when

Rus--

the latter was-

his Gruiidsdtze der Arithmetik.

famous paradox of a class


of classes which are not their own elements, and demonstrated
that at the very foundation of Frege's magnum opus there was
In his letter, Russell formulated his

contradiction

which threatened the

great mathematician

and

logician,

entire

now

system.

believed

to

Hence

that

have been

write
the greatest innovator in logic since Aristotle, had to

the

"Appendix"

to

his

work

these

sad words:

25

Logic

"Hardly anything more unfortunate can


writer than to have

befall

one of the foundations of

scientific

his edifice

shaken

work is finished.
"This was the position 1 was placed in by a letter of Mr.
Bertrand Russell, just when the printing of this volume was
the

after

Hearing completion

...

''Solatium (sic) miseris socios habuisse dolorum.


this

comfort,

if

comfort

it is;

for everybody

who

proofs use of extensions of concepts, classes,


position as

I.

What

is

question

in

my

have

at least

made

in his

the

same

sets, is in

not just

is

has

particular

way

of establishing arithmetic, but whether arithmetic can possibly

be given a logical foundation at

all''^.

This shows that Frege was far from underestimating the


nificance

set theory,

with which Frege's system was also related. This was

also the issue to


treatise

sig-

of Russell's criticism, which in fact affected Cantor's

which Principia Mathematica, the main

logical

by Russell and Whitehead was devoted.

In what did the criticism contained in Russell's letter to Frege


consist?
"...

To

use Frege's words again:

Mr. Russell has discovered a contradiction which may

now be stated.
"Nobody will wish
a man.

We

to assert of the class of

that something belongs to a class

whose extension the


cept:

(if

classes that
it

do not belong

First, let

falls

it

now

itself,

it

it

is

say

under the concept

our eyes on the conThe extension of this


is

For

thus the class of


short,

ask whether this class

it

that

itself. I

fix

extension)

to themselves.

now

us suppose

its

falls

we

will call

belongs to

does. If anything belongs to a class,

under the concept whose extension the class

our class belongs to

Let us

is.

we may speak of

the class K. Let us

itself.
it

class

when

class that does not belong to itself.

concept

men

have here a class that does not belong to

is

is.

Thus

if

a class that does not belong to

"Frege on Russell's Paradox", in Translations from the Philosophical

Writings of Gottlob

Frege,

Oxford 1952,

p.

234.

26

Research Problems of Semantics

Our

itself.

Secondly,

then

thus

us suppose our class

let

falls

it

supposition

first

leads

self-contradiction.

to

does not belong to

under the concept whose extension

thus does belong to

itself.

itself,

and
Here once more we likewise arrive
it

itself is,

at a contradiction! "2

paradox achieved

Russell's

came to be formulated in
more or less difficult.

others were discovered later. It also

more or

various manners,

Not

less intuitive,

the paradoxes will be presented here^, but in view of the

all

exceptional historical importance of Russell's paradox


give
in

it

many

although

fame,

special

once more as formulated by Mostowski.

mind

Mostowski the concepts

that for

shall

must be borne
and "property"

It

"set"

are identical.

"Let normal properties be called those properties which are

not attributes of themselves. Thus the properties:


to

be a function,

no man

is

to

be a number,

is

not normal since

Frege, op.

The reader who would be

in greater detail

cit.,

is

The property:

an attribute of

willing to study the

problem of paradoxes

can be referred to the exhaustive literature of the subject.

Principia Mathematica,

Vol.

is

1,

given by: A. N. Whitehead & B.


Cambridge 1925. pp. 60-65; R.

Carnap, The Logical Syntax of Language, London 1937, 60 a. ff.


Ramsey, The Foundations of Mathematics, New York 1931, pp. 20
Hilbert &.
p.

92

W. Ackermann, Gnwdziige

der theoretischen Logik,

F. P.
ff.;

D.

Berlin 1928,

if.

In Polish literature these problems are analysed in detail by A.


ski in

to

itself.

p. 235.

general approach to the problem

Russell,

it

be a man,

to

are normal properties (since

identical with that property, etc.).

be a property

etc.,

Mostow-

Logika matematyczna [Mathematical Logic], Warszawa-Wroclaw 1948,

pp. 204-222

and 308-324. Sea

logiki formalnej

also T. Kotarbinski, Elementy teorii poznania,

metodologii nauk [Elements of Epistemology, Formal Logic

and the Methodology of

Sciences],

Lwow

1929, pp. 138-139;

Lwow-Warszawa
Warszawa 1949, pp.

Granice nauki [The Limits of Science],


136; T.

Czezowski, Logika [Logic],

More

specialist

works by

S. Lesniewski, L.

L.

[1935],

Chwistek,
pp. 125-

13-19.

Chwistek and A. Tarski are

given in the Bibliography at the end of this book.

27

Logic

"If

normal

we

Thus the
of

use the term set instead of property

sets

all sets is

set

has

W^

W there

any property

the following property

W:

be a normal property

(In another formulation:

{W

elements.

of men, functions, numbers, are normal; the

to

(1)

shall say that

own

not normal.

now examine

"Let us

for

we

are those sets which are not their

sets

is

the set of

all

normal

sets.)

Thus,

the equivalence:

is

WJ ^ {W

the property

a normal property),

is

from which by the law of transposition we obtain the second


equivalence
(2)

(W has not the property WJ ^ (W is not a normal property)

"Let us
"If

W^

now
is

ask whether

properties) the property

or

is,

W^

is

is

not, a

not an attribute of

W^. Hence by

has not the property


a

W^

normal property.

a normal property, then (by definition of the normal

(2)

we

normal property, contrary to the assumption that


"If

W^

is

to the

(1)

it

W^

is

W
not

is.

not a normal property, then (by definition of the

normal properties) the property

Hence by

so that

itself,

infer that

we

W^

infer that

assumption that

it

is

is

W^

an

is

attribute

of

itself.

a normal property, contrary

not.

"Thus both the assumption that W^ is a normal property


and the assumption that W^ is not a normal property, lead to
a contradiction in other words, W^ neither is nor is not a normal
;

property although, by the law of the excluded middle, one of these


cases should occur"^.

This reasoning can be

still

more simply presented by means

of formula:

"The

From

definition of the

that definition

Mostowski, op.

W^

is

we obtain
h

'*

property

cit.,

Xe W.^i{XeXy,
p. 209.

28

Research Problems of Semantics

W for the
{p^p')^Bp.p\

whence, by substitution of the constant

and on the strength of the tautology

letter
it

X,

follows

that

it

have dwelt at some length on Russell's paradox because

is

typical as least of the group of paradoxes connected with

theory and with the unrestricted use of the words "every"

set

and

In Principia Mathematica,

"all".

ferentiate

Russell

still

did not dif-

between the various types of paradoxes, but the pubh-

cation of The Foundations of Mathematics


initiated the division into

...

mathematical and

by F.

Ramsey

P.

linguistic (semantic)

paradoxes; that division was accepted by Russell in the introduction to the second edition of The Principles of Mathematics
(1937).

What

is

the issue at stake in the case of this paradox which,

like the

semantic paradoxes of "the liar" type (Eubulides' par-

adox),

appears

involved

is

no

be

to

sophistic

quibble?

The

problem

one, since the demonstration that con-

trifling

tradictions can be constructed within a deductive theory

tamount to cancelling such a theory.


adoxes

thus

gave

rise

to

Russell's

and

is

tan-

similar par-

the necessity of either abandoning

the law of the excluded middle and the logical principle of contradiction, that
it,

is

from the system of formal

logic as

we know

or else seeking ways of eliminating contradictions by cor-

recting the system of formulating one's thoughts. This has di-

rected the attention of scientists to language not only as an in-

strument, but also as an object of research, the

one type of the paradoxes,


liar

e.g.,

the

classical

more

so since

antinomy of the

or the later antinomies of Berry and Richard 6, showed clearly

that the issue

Linguistic

was one of language.


interests

in

logic thus

resulted

from the natural

needs of that discipline and were dictated by the necessity to elim5

Ibid., p. 210.

6 Ibid.,

pp. 315-320.

ji

20

Logic

contradictions threatening the

inate
It

very foundations of logic.

has been found that, in the case of the mathematical paradoxes,

the point lay in

inevitably

an

exposed

word

illegitimate use of the

set

"every", which

theory and related logical theories to the

danger of contradictions; in the case of linguistic paradoxes the


point lay in the confusion of the language being studied with the

language used as concerning the language under investigation.


"In both cases," says Mostowski,

a system

enon:

that

"we observe the same phenoma system

too universal,

is

in

which

'too'

much can be said\ must be contradictory''''^.


In any case, linguistic problems have since become an inseparable part of logical research (more strictly, research into the
foundations of mathematics and logic), creating a stimulus to
the various philosophical interpretations

philosophical trend rests on logic

but

That

why

is

separate philosophy

and the results of logical research,


and direction of in-

chapter,

my

so

difiicult,

logic

as

mentioned above,

these matters.

in

Although

should like to anticipate

my

further remarks by stat-

negative attitude towards the various forms of semantic

philosophy as a

my

so, in

is

it

and

philosophical problems will be dealt with in the next

strictly

ing

That

stricto.

in turn influences the further course

it

vestigations.
to

and speculations, which

with logical investigations sensu

thus intertwine

new

opinion,

variation of subjective idealism. But the


is it

more

necessary to bring out the real scientific

achievements of contemporary mathematical logic, which con-

making us reahze that language

sist in

research. This discovery

is

briUiant in

also
its

is

an object of

elements of a knowledge of that fact can be found


in the history

than

scientific

was raised

much

earlier

of science, but they were brilliant intuitions rather


conclusions.

at the very

self-evident that

Ibid., p. 320.

it

As a developed

theory, the matter

end of the 19th century and

uct of the 20th century.

logical

very simplicity. True,

As a statement

seems almost

trivial.

it is

is

now

But such

in fact a prod-

so simple
is

and

the fate of

30

Research Problems of Semantics

great discoveries, once they are absorbed by the organism of

all

To be exact, it must be added that the latter opinion


many violent opponents as far as regards conclusions from

science.

has

a general statement.

The

logical study of language followed various courses

and

various forms in different periods. They began with an

took

analysis of language

(some

from the point of view of its

different layers

scientists distinguished hierarchical types; others,

Then

semantic

was the subject


came the study of the relationship
between linguistic expressions and the objects they denote, and
also between linguistic expressions and the subject using the
language. These researches and theories bore different names:
type theory; the theory of metalanguage, metascience and metacategories).

of investigation.

the logical syntax of language


Finally,

of language; semantics; semiotics

logic; the logical syntax

(di-

vided into syntax, semantics and pragmatics). Thus, the term


"semantics" seems to be reserved here for a special
search. But

let

us not

fall

into error

field

of

re-

the overwhelming majority

of hnguistic analysis and research belongs to semantics in the

broad sense of the word, since


other, with the relationship

the objects they denote.

a.re

some form or an-

linguistic expressions

thus in

when we speak of

guistic intuition
tics

We

deals, in

it

between

full

agreement with

ical (linguistic)

semantic analysis as of seman-

we

avail ourselves not only of data

from

histor-

semantics, but also from the theory of metalan-

guage, logical syntax and semantics in the narrower sense of

word. This

is

why

it

is

logic,

from semantics

in

all linguistic interests

a narrower sense of the term, con-*

fined to special studies of the relationship between expressions


the

objects they denote. This view

of Alfred Tarski.

the'

correct to distinguish between semantics

a broad sense of the term, which covers

of

lin-

applied in every case of investigating linguistic ambiguity.

In such analysis,

in

and

He understands

analysis concerned with concepts

ships between expressions

is

in

full

and

accord with that

semantics as the whole of the

which

refer to certain relation-

and the objects they denote

(as

example

Logic

31

of such concepts he gives those of denoting, satisfying and defin-

and thinks that such an interpretation of the term "seman-

ing),

narrower than the usual sense of the words. Kotarbinski

tics" is

verbis refers to

expressis

whole of semiotics

the

syntax

(i.e.,

semantics and pragmatics) as to semantics in the broader sense

of the term 9.

But

As

let

us revert to Russell's paradox and

its

consequences.

early as in 1892, Gottlob Frege in his paper "tJber Sinn

und Bedeutung"io, raised the


denotation, that

is

theoretical

problem of meaning and

the focal issue of semantics (Frege undoubt-

on the

edly based himself

between

tradition of differentiating

denotation and connotation),

but that idea, like other ideas

of that great thinker, remained unnoticed for a long time.

we know,

problems came to the forefront

linguistic

following the discovery of Russell's paradox.

The

in

As

logic

possibilities

of overcoming that and similar paradoxes were analysed by


Russell in his The Principles of Mathematics,

on that

idea

was presented

issue

in Principia

and the

full-fledged

Mathematica, which

he wrote together with A. N, Whitehead. This was called the


theory of types.

The operation he recommends is very simple, though by no


means intuitive. As in other similar theories, it consists in so
restricting the "universal" character

Thus,

tradictions.

A. Tarski,

Foundations for
pt.

was

of language as to avoid con-

at first a purely negative operation^^ Tt

"O ugruntowaniu naukowej


Semantics],

Scientific

in

semantyki" [On

Laying the

Przeglqd Filozoficzny,

Vol.

39.

50.

p.

1,

it

T. Kotarbinski, Przeglqd problematyki logiczno-semantycznej [A Re-

view of Problems of Logical Semantics], reprinted from Sprawozdania z czynnosci

Towarzystwa Naukowego za

posiedzen Lodzkiego

No.

Vol. 2,

\.

(3),

Lodz

I polrocze

1947,

1947, p.

und philosophische

1892.

'0

Zeitschrift

'1

In the introduction to the second edition of Principia Mathematica

we read

"It

fiir

Philosophie

Kritik.

should be observed that the whole effect of the doctrine of types

but does not permit any which

would otherwise be valid,


would otherwise be invalid" (A. N. Whitehead

&B.

p.

is

negative:

it

forbids certain inferences which

Russell, op.

cit.,

Vol.

I,

VH).

32

Research Problems of Semantics

was the further development of type theory, and in particular


from what is called the ramified theory of types (Russell)

transition

what

to

the

called

is

Ramsey), that made


it

simplified

theory of types^'^ (Chwistek,

more natural and

it

intuitive

and brought

closer to distinctions of grammatical categories, as especially

manifested

in

theory of semantic

Stanislaw Lesniewski's

cate-

gories, a concept related to the simplified theory of typesi3.

Russells's idea

is

that language cannot be so universal as to

admit of statements about


has not been previously

of the elements of that

the elements of a

strictly defined

words: a statement about

validly only

all

set,

put

a statement about a totality can be


totality.

To

But

let

"An
they

all

circles in

may

made

observe

at the root

lie

This

paradoxes.

is,

fundamental idea of the theory of types.

Russell speak

analysis
result

fail to

to obtain statements that are not wrong, but

is

circle in reasoning, leading to

briefly, the

that set

closed. Or, in other

simply meaningless. Such meaningless statements

of a vicious

if

the elements of a set cannot be one

all

"from the outside" of that

that restriction,

and

set,

for

himself:

of the paradoxes to be avoided shows that

from

The

certain kind of vicious circle.

vicious

question arise from supposing that a collection of objects

contain

members which can only be

means of

defined by

the collection as a whole. Thus, for example, the collection of

propositions will be supposed to contain a proposition stating


that

'all

propositions

are either true or false'. It

would seem,

however, that such a statement could not be legitimate unless


'all

propositions'
12

L. Chwistek,

referred

to

"Antynomie

Logic], in Przeglqd Filozoficzny,

some already

logiki formalnej"

Vol. 24,

collection.

definite

[Antinomies

1921, pts. 3

and

4,

in

Formal

and Granice

nauki [Tiie Limits of Science], Ciiap. V; F. P. Ramsey, The Foundations of


Mathematics.
13 S.

Lesniewski,

"Grundziige eines

neuen

der Grundlagen

Systems

der Mathematik", in Fundamenta Mathematicae, 1929, vol. 14, and

stawach ontologii" [The Foundations of Ontology],


siedzeii
pts.

Towarzystwa Naukowego

4-6.

in

"O pod-

Sprawozdania z po-

Warszawskiego, Section

3,

Vol. 23,

1930.

33

Logic

which

cannot do

it

about

new propositions

if

We

propositions'.

'all

statements about

'all

are created by statements

shall, therefore,

given any set of objects such that,

ally,

have to say that

More generwe suppose the set

propositions' are meaningless.


if

members which presuppose this


total, then such a set cannot have a total. By saying that a set
has 'no total', we mean, primarily, that no significant statement
can be made about 'all its members'. Propositions, as the above
illustration shows, must be a set having no total. The same is
have a

to

total,

we

true,

as

when

these

it

will contain

shall shortly

are

up our

to such as

restricted

argument a given object

of propositional functions, even

see,

can significantly have as

In such cases,

a.

set into smaller sets,

necessary to break

it is

each of which

is

capable of a

what the theory of types aims at effecting.


"The principle which enables us to avoid illegitimate

This

ties

total.

is

may

lection

be stated as follows: 'Whatever involves

must not be one of the

collection';

provided a certain collection had a

total,

it

or,

all

totali-

of a col-

conversely:

'If,

would have members

only definable in terms of that total, then the said collection


has no total'.

because

it

We

shall

this the 'vicious-circle principle',

call

enables us to avoid the vicious-circles involved in the

assumption of illegitimate

way

totalities

...

'All

propositions'

must

becomes a legitimate totality,


and any limitation which makes it legitimate must make any
be in some

limited before

statement about the


Russell

applies

the

totality

it

fall

outside

the totality" i-^.

of paradoxes

analysis

functions and demonstrates that there, too,

to

propositional

without the help

of a hierarchy of types, paradoxes arise which are based on a

vi-

cious circle in reasoning.


"...

We

fallacy at

shall find that

the very outset,

it

is

possible to incur a vicious-circle

by admitting as possible arguments

to a propositional function terms

"

Whitehead

&

Russell, op.

cil.,

which presuppose the function.

pp. 37-38.

34

Research Problems of Semantics

This form of the fallacy

we

leads, as

When we

"...

very instructive, and

is

its

avoidance

shall see, to the hierarchy of types.

we mean

say

that

ambiguously denotes

'cpjsc'

(pa,

96,

means one of the objects 9a, (pb, 9c,


etc., though not a definite one, but an undetermined one. It follows that '9X' only has a well-defined meaning (well-defined,

9c

etc.,

that

is

that

'cpx'

to say, except in so far as

ous) if the objects 9a, 9^, 9c,

say,

values

function
are

of

it is

its

not a well-defined function unless

is

already

well-defined.

no function can have among


poses the function, for

if it

follows

It

tion

was

while

from

is

to

all its

this

that

values anything which presup-

its

had, we could not regard the objects

ambiguously denoted by the function as


definite,

essence to be ambigu-

are well-defined. That

etc.,

conversely, as

function cannot be definite until

its

definite until the func-

we have

just seen, the

values are definite. This

is

a particular case, but perhaps the most fundamental case, of the

the function, since,

is what ambiguously denotes


namely the values of the functions,
cannot contain any members which involve
if it did, it would contain members involving

the totahty, which,

by the

vicious-circle principle.

some one of a
hence

function

certain totality,

this totality

vicious-circle principle,

no totahty can

do"i5.

In order to avoid such dangers Russell suggests a division


of the universe of discourse into "types": individuals,
individuals,
sets

relations

of individuals,

bohzed, which makes


restricts the possibihty

ing

to

contradictions

in the case of

between

individuals,

The "types"

etc.
it

relations

sets

of

between

are correspondingly sym-

possible to distinguish

them and thus

of using them in an improper way, lead-

and

paradoxes.

As we already know,

an improper substitution, a function changes into


is, certain substitutions are meaningless by vir-

a nonsense, that

tue of the linguistic prohibitions


types.

5 Ibid.,

pp. 38-39.

formulated by the theory of

Logic

Thus, the theory of types

35

a result of studies in the language

is

of logical statements and of determination on that basis of a cer-

and

tain definite hierarchy of objects

their

names. That theory,

however, was not intuitive enough and moreover had for techni-

such additional elements as the axiom

cal reasons to introduce

of reducibility.

simphfied theory of types of

It is the so-called

Leon Chwistek which makes it possible to avoid antinomies


and at the same time is natural and intuitive. Its fundamental
idea

is

that in logic

is

it

possible to speak only about objects

of a strictly defined type, so that

not permissible to speak

it is

of a class "in general", but only of a class of strictly defined objects.

In conformity with the principle of purity of types, con-

nected with that theory, propositional functions (which are equivalents of classes) in

over objects

which there

16

fied

one variable ranging

still

more natural

in Stanislaw Les-

theory of semantic categoriesi^, which comes close

"In order to eliminate Russell's antinomy

The theory of

theory of logical types.

Bertrand Russell.
in a

at least

of different types, are meaningless^^.

That idea appears to be


niewski's

is

It is

few words. But

it

suffices to

adopt the simpli-

types has been formulated by

a complicated theory and cannot be formulated clearly

it

can be simplified so that

it

really

can be expressed in

a couple of simple sentences.


"I
I

adopt the so-called universe of discourse consisting of objects which

call individuals.

No

properties or concrete examples of individuals are given.

"Apart from the individuals

adopt classes of

classes of

individuals,

of individuals, etc.
"That is all. It is obvious that here a concept of

classes

sense.

One may speak

class as

only of classes of certain definite

quently the problem whether a class

is its

such has no

objects.

own element must be

Conse-

rejected as

devoid of sense.

"That simplified theory of types has been for the


by

me

tek,

in

my work Antynomie

first

time formulated

logiki fonnalnej, published in 1921" (L.

Chwis-

Granice nauki, p. 129).

"

"In

1922

outlined a certain

which was to replace for


in

my

to

speak meaningfully,

me

conception of "semantic categories'

such or another 'hierarchy of types'

opinion, of any intuitive substantiation


I

would have

and which,

to adopt even

if

there

in

deprived,

my

desire

had been

no

Research Problems of Semantics

3^)

Here too,

to distinctions in grammatical categories.

theory, stress
gories,

is

laid

on

Lesniewski expounded his ideas mainly by word

Since

and

technical

his

we

of short cuts in the course of reasoning,

full

of-

works are very

the excellent presentation of the problem of semantic

category by Alfred Tarski, formulated with his usual classical


lucidity

"For reasons mentioned

we cannot

at

the

beginning of this section,

offer here a precise structural definition of semantical

category and will content ourselves with the following approx-

same semantical
category if (I) there is a sentential function which contains
one of these expressions, and if (2) no sentential function which
contains one of these expressions ceases to be a sentential function
if this expression is replaced in it by the other. It follows from
this that the relation of belonging to the same category is reimate formulation: two expressions belong

flective,

to the

symmetrical and transitive. By applying the principle

of abstraction,

all

the expressions of the language which are parts

of sentential functions can be divided into mutually exclusive

two expressions are put into one and the same class
and only if they belong to the same semantical category, and

classes, for
if

each of these classes

is

called a semantical category.

simplest examples of semantical categories

it

Among

suffices to

the

mention

the category of the sentential functions, together with the categories

which

include

respectively

the

names of

of classes of individuals, of two-termed relations


dividuals,

as

individuals,

between

and so on. Variables (or expressions with

antinomies' whatever.
in

My

in-

variables)

conception of 'semantic categories', while being

a close formal connection with the known 'theories of logical types' as far
its

theoretical consequences are concerned, referred, as for

aspect, rather to

Husser!"

its

the tradition of Aristotle's 'categories', to 'the

speech' of the traditional grammar,

Edmund

a confusion of which leads to meaninglessness.

mouth, in lectures and conversations, and


refer to

as in type|

the principle of purity of semantic cate-

(S.

and

intuitive

parts of

to the 'semantic categories' of

Herr

Lesniewski, "Grundziige eines neuen Systems"..., p.

14).

Logic

37

which represent names of the given categories hkewise belong

same category"i8.

to the

For technical purposes,


ticular natural

order

number

assigned to

is

category. It

if

the

assigned a par-

The same

two functions that they are of the same

number of

free variables

category in the two functions

tical

is

expressions which belong to the given

all

said of

is

semantical type,

to every category

called the order of the category.

of every seman-

the same.

is

Here too, as in the ramified and the simplified theory of


antinomies are avoided by observing certain linguistic

types,

prohibitions: variables in functions cannot range over different


types of expressions or expressions belonging to different seman-

The operation

tical categories.

is

practically identical with that

seen above, the difference consisting, apart


aspect, in a greater comprehensibility

from the technical

and naturality of the theory

of semantical categories'^.
18

A. Tarski, "The Concept of Truth in Formalized Languages",

A. Tarski Logic,
19

Semantics,

Oxford 1956

Mathematics,

Under the impact of those

and

theories,

216.

p.

especially the simplified

theory of types, Russell later changed his argumentation and actually


close to the standpoint characteristic of the simplified theory of types

theory of semantic categories.

The introduction

The Principles of Mathematics (London 1937,

"The

technical essence of the theory of types

ositional function

which
'if

is

is

it

p.

^cpx''

of which

all

is

mortal' are true, and

'x".

came

and the

to the second edition of

XIV)

reads:

merely this

values are true, there

not legitimate to substitute for

a man, x

is

in:

Given a prop-

are expressions

For example: All values of


infer 'if Socrates is a man,

we can

Socrates is a mortal'; but we cannot infer 'if the law of contradiction is a


man, the law of contradiction is a mortal'. The theory of types declares
this latter

values of

of words to be nonsense, and gives rules as to permissible

set
'jr'

in ^(p\

In the detail there are

but the general principle

always been
ary

is

difficulties,

and complications,

merely a more precise form of one that

recognized. In the

older conventional logic,

out that such a form

it

has

was custom-

of words as 'virtue is triangular'


no attempt was made to arrive at a definite set
of rules for deciding whether a given series of words was or was not significant.
This theory of types achieves. Thus, for example, I stated above that 'classes
of things are not things'. This will mean: 'If x is a member of the class a

is

to

point

neither true nor false, but

38

Research Problems of Semantics

In any case,

it

may be

said that the interest in the theory

of language (semantics in the broad sense of the term) was

in-

seminated in logic by the analysis of antinomies, and that the


antinomies connected with the use of the word "every"

antinomies) can be removed by means of this or that form

retical

of the theory of types. This

was the
mies

first

to differentiate

not

sufficiently

lected in Principia

also admitted

is

and semantical antinomies.

remarked, and the fact

Mathematical

two fundamentally

distinct

entirely neg-

is

that these contradictions

we

groups, which

The best known ones are divided


A. (1) The class of all classes which

and

by Ramsey who

between the two groups of antino-

antinomies

theoretical

set

"It is

into

(set theo-

B.

will

fall

call

as follows:

members of

are not

themselves.
(2)

The

relation between

not have

itself to

(3) Burali-Forti's

B. (4)

'I

am

two relations when one does

the other.

of the

contradiction

greatest

ordinal.

lying'.

(5)

The

(6)

The

(7)

Richard's contradiction.

(8)

Weyl's contradiction about the word

greatest integer not

nameable

in

fewer than nine-

teen syllables.
last indefinable ordinal.

The principle according to which

'heterologisch'.

have divided them

is

Group A consists of contradictions


made against them, would occur in

of fundamental importance.
which, were no provision

a logical or mathematical system

mathematical terms such as


there

itself.

class

They involve only

must be something wrong with our

logic or mathematics.

But the contradictions of Group B are not purely


cannot be stated in logical terms alone for they
;

reference to

is

thought, language,

a proposition, and cpx

is

or symbolism,

a proposition,

but a meaningless collection of symbols".

logical or

and number, and show that

tlien 9;a is

all

logical,

and

contain some

which are not

not a proposition,

39

Logic

formal but empirical terms. So they

may

be due not to faulty

or mathematics, but to faulty ideas concerning thought

logic

and language"2o.

The second group requires more particular concentration


of attention on linguistic aspects and different operations than
did the antinomies of the first group. Hence the additional stimwhich

uh

intensified

the

theoretical

of logicians

interest

in

and the problems of metalanguage


and metascience. The concept of metalanguage and the hierarchy
of languages is closely connected with the theory of types and
in a sense results from the latter, a fact already realized by RusselPi. For if expressions of a language belong to different logical
the hierarchy of languages

types (or to semantical categories of different orders), this nec-

makes us think of differences in the hierarchy of languages


as depending on the logical types (or on the order of semantical
categories) of the expressions they contain. The more so since
essarily

conception

this
to

the

connected with certain practical needs relating

is

elimination

of logical

antinomies.

In order to understand this issue better,


at

the

difficulties

let

us take a look

which appeared in deductive

disciplines

in

connection with antinomies of the type of the classical antinomy


of Eubuhdes (the antinomy of the

mum

clarity,

liar).

For the sake of maxi-

quote the Jan Lukasiewicz's version, in Tarski's

formulation, which will enable us to separate the real linguistic

problem from

its

sophistic

shell.

"For the sake of greater perspicuity we


'c'

shall use the

symbol

as a typographical abbreviation of the expression 'the sen-

tence printed

on

this page, line 5

from the

top'.

Consider

now

the following sentence:


c

is

not a true sentence^^.

20

Ramsey, op.

21

This

is

cit.,

pp. 20-21.

stated explicitly in

An

Inquiry into

Meaning and

Truth,

1951, p. 62.
22

That sentence precisely appears

in line 17th

from

top.

London

40

Research Problems of Semantics

"Having regard

meaning of

to the

symbol

tlie

we can

'c',

establish empirically:
(a) 'c is not

a true sentence^

is

"For the quotation-mark name


other of
(P)

its

names) we

not a

is

c.

up an explanation of type

set

not a true sentence'

'c is

identical with

of the sentence c (or for any

a true sentence

is

(2)23

and only

if,

if,

sentence.

true

"The premises

and

(a)

together at once give a contradic-

((3)

tion:
c

is

a true sentence

Commenting on

and only

if,

"The source of
in the

scheme

is

not a true sentence"^^.

why such

we have

substituted for the symbol

an expression which

(2)

term 'true sentence'


be given

this contradiction is easily revealed: in order

to construct the assertion (^)


'jP'

if,

that antinomy Tarski remarks:

Nevertheless,

(...)

no

contains the

itself

rational

ground can

substitutions should be forbidden in prin-

ciple"25.

All this stands out in

Weyl's

antinomy:

This antinomy

still

greater relief in connection with

word

the

consisting

heterological, then

it is

is

is

it is

heterological?

"heterological"

in the fact that

when we say

that

not heterological, and vice versa

also clearly linguistic in nature.

"For the way," says Ramsey, "in which


functions into orders of which

escape the contradictions of

no

this

used to

shown

to result

Group

B, which are

from the ambiguities of language which disregard


reference

may

this distinction,

be made to Principia Mathematica (Vol.

Here

1910, p. 117).

distinction of

totality is possible is

it

may

I,

1st ed.,

apply the method to a


work which is particularly free
mean Weyl's contradiction concerning

be

sufficient to

contradiction not given in that

from
23

irrelevant elements

This refers to an explanation of the statement of the type "x

sentence", which
24

is:

"x

is

a true sentence

if

and only

is

a true

if /?".

A. Tarski, "The Concept of Truth in Formalized Languages",

A. Tarski, Logic, Semantics, Mathematics. Oxford 1956


25 Ibid. p.

158.

p.

158.

in

Logic

41

word 'heterologisch' (Weyl, Das Kontinuum, p. 2), which


must now be explained. Some adjectives have meanings which
the

of the

predicates

are

'short'

'heterological'

predicate

of

that

it;

first,

is,

If
is

it

meaning

its

not

is

it

is,

is

it,

like

word

the

meaning

a predicate of

made

attempts were

Let us

not a

is

not heterological. But

is

word

the

long.

of them,

Now,

its

So we have a complete

heterological.

At

thus

itself;

'long'

heterological.

heterological?

not heterological,
it is

others

autological;

'short',

word

the

whose meanings are predicates

adjectives

call

word

adjective

but

short,

is

if

it

is

and therefore

contradiction''^^.

to avoid antinomies of that

kind by resorting to the operation already described above and

based on the theory of types (or semantical categories).


only later that they

came

It

was

to be separated into a distinct group,

and attention was focused on the hierarchy of languages and the


consequences of disregarding that hierarchy.

The
bear

is

starting point of reasoning

mind the

difference

is

very simple: one has to

between the language which

analysed {object language) and the language


ysis is

made

{metalanguage).

classic

cussion in Polish of grammatical rules of

in

example here
e.g.

is

being

which that analis

the dis-

the English language.

Such a situation may, however, occur within one and the same
language,
to

if

some of its expressions

which they belong. This

denotation, truth, etc.


different layers

being

fixed

We

are discussed in the language

applies, e.g., to analysis of

meaning,

have then to do, so to speak, with

of that language, the hierarchy of such layers

by

their

lesser

or greater "richness".

Of

course,

that can, in a pure form, appear only in formalized languages,

which are consciously and purposefully made to serve the needs


of deductive sciences. In the ordinary language, wliich is universal in character

(i.e.,

includes

all

possible expressions), con-

fusion covers not only expressions belonging to different logical


types, but also linguistic layers belonging to different levels

26

Ramsey, op.

cit.,

pp. 26-27.

of the

42

Research Problems of Semantics

hierarchy of languages. This

and

is

precisely the source of dijfficulties,

of antinomies resulting from the

gives rise to the danger

confusion of expressions of different types, and of languages

of different hierarchies. Yet,

a correct distinction

if

is

made

and metalanguage, the latter will


be "richer", and object language will form a part of it: metalanguage will include, apart from the signs and expressions of
as between object language

object language, also their individual


tive,

names

(structural-descrip-

or other), the names of relations between them and general

logical expressions27.
If such
is

a distinction between languages

made, we may no longer confuse them.

not a true sentence''

If

and then ask whether

of various

we say

levels

that "c

that sentence

is

is

true

or not, then the error consists in the original statement which


is

ly

impossible

if

object language

separate. This

"true",

and metalanguage are kept

so because that statement includes the term

is

which belongs to metalanguage: we have thus subjoined

a metalanguage category to object language.

when

it

comes

to the

comprehensive

is,

to

we observe

language.

language which in

its

when we use a

That contradiction can be

the prohibition

metalanguage

The same holds

antinomy connected with the word "hetero-

logical". Here, too, the contradiction arises

if

strict-

too

avoided

on using too "rich" language,

that

system includes expressions belonging

(names of expressions

and

such

categories

as "to denote", "to be true", etc.).


It is

obvious that none of

ordinary language.
pressions.

this applies to the

It is universal

But for that very reason

and includes
it

is

normal use of

all

exposed to

possible exall

sorts of

dangers, which have been exploited since the time of the Sophists.

Yet what

27 Cf.

in

ordinary language

is

a curiosity and a linguistic

A. Tarski, "The Concept of Truth

in

Formalized Languages",

A. Tarski, Logic, Semantic, Matliematics, pp. 166, 170-171, 173-174^ 212214, et passim; R. Carnap, Tlie Logical Syntax of Language, pp. 153 ff; and
in

Introduction to Semantics,

Cambridge (Mass.), 1948, pp.

3-4.

43

Logic

trick

without any practical significance, becomes a real problem

when

comes

it

to

theories that the


to solve

formahzed

it is

precisely in those

means are connected with

are found there. These

it

But

theories.

problem can be solved, and adequate means

language analysis, and with the discovery that language, too,


is

an object of study.

The

distinction

more advanced

presses for
teria for

between object language and metalanguage


analysis with a view to providing cri-

such a distinction, and in particular for the study of

such metalanguage expressions as refer to the relationship between


object language expressions
this

has given

and the objects they denote. All

rise in logic to interest in the description

en language, in

its

of a giv-

syntax and in semantics in the narrower sense

of the term.

Let us

now

these fields

try to characterize

of research and

the problems with which they are concerned.


Before,

however, giving a general characteristic of logical

syntax and the issues

it

deals with,

it

seems advisable to remove

from our exseem that only the concept of

the danger of misunderstandings that might arise


position of the subject. It might

metalanguage, which was a result of grasping the fact that ordi-

nary language

is

a multi-layer structure, stimulated interest in

syntax and semantics. Such

is

the logical order of things,

which

can be established ex post facto. Actual chronology was


ferent:

all

these problems

and

interests

dif-

were intertwined and

helped one another to develop, rather than to appear successively.

clear sequence in time

of syntactical

and semantical

can only be observed in the case


studies.

The

true

development

when it was shown that


logical syntax does not include a number of notions indispensable for the development of logic, and when Tarski had demonof semantics sensu stricto began only

strated that

the theory

it

possible to introduce the concept of truth into

of deduction without falling into contradiction.

What then
matter?

is

is

so-called logical syntax

and what

is its

subject

44

Research Problems of Semantics

According to Carnap,

it is

such a

field

of

linguistic research]

which we abstract from the users of language and from

in

the

designata of linguistic expressions, and analyse exclusively the'


relations

between

expressions^s.

Carnap

Elsewhere,

defines

logical syntax as a formal theory of linguistic forms, that is such

a theory as

is

interested not in the

meaning of expression, but

only in the kinds and the order of symbols of which those expressions are built29. In such an interpretation, logical syntax

determines the rules according to which such hnguistic entities


as sentences are built of such elements as words. Thus, in such

an interpretation, syntax in the narrower sense of the term lays

down

formation rules for linguistic

establishes

transformation

rules

for

entities,

them.

and formal

logic

Since both can be

formulated in terms of syntax, Carnap combines them into a

system as the logical syntax of language

gle

logic

(in

sin-

such a case formal

becomes a part of syntax and acquires a purely formal

character).

Conceived in
cific calculus.

this

way, logical syntax treats language as a spe-

Exponents of that conception do not claim that

such a calculus exhausts the analysis of language, and do not


to appreciate the importance of semasiological, psycholog-

fail

ical,

to

and

sociological analysis (latterly, the last two have

come

be replaced by the term "pragmatics"). They divide the syn-

tactical study

of language into descriptive syntax, which deals

with the empirical study of the syntactical properties of given


languages, and pure syntax, which includes analytical sentences

of metalanguage obtained from the adopted definitions of such


syntactical

concepts

as,

e.g.,

"sentence

"provable in K", "derivable in K"\


If
ter

we now

ask,

in

[the

system]

K'\

etc.^o.

which concrete problems are the subject mat-

of logical syntax, we find ourselves in an embarrassing posi-

tion,

because in the period of euphoria, especially

28

R. Carnap, Introduction

29

R. Carnap, The Logical Syntax of Language,


R. Carnap, Introduction to Semantics, p. 12.

^0

to Semantics, p. 9.
p.

at the time

45

Logic

saw in

the so-called Vienna Circle

when

and philosophers'

panacaea
to include

not only

stone,

logical syntax a specific

syntax

logical

was made

the logical problems, but also

all

all

that

was considered meaningful in philosophy.


This fact gives rise to the difiiculty already mentioned at the

beginning of this chapter, namely the


logical

difficulty

problems from philosophical ones. In

this

of separating
concrete case,

they intertwine so as to form a single indivisible whole.

The

is-

sue of philosophical interpretation will be treated again in the

following chapter. Yet here

I shall also

point to two philosophi-

cal tendencies closely connected with the

now

being discussed.

First of

all,

there

is

the conventionalist tendency linked with

the problems of logical syntax.


tical

problems of syntax

When

investigating the syntac-

Carnap and others assumed that the


or that language and the laws by which it is

aspect of language,

adoption of this

that

governed

is,

consequently, the adoption of this or that

an arbitrary matter, a matter of convention. That


idea was explicitly formulated by Carnap in the form of the

logic

is

so-called

principle of tolerance^i

choice of rules32.

He

and the principle of a

held the same opinion also in a

free
later

when from recognizing syntax alone, he passed to a broader theory of semantics. At that period, the basing of syntax
and semantics on the conventionalist principles is to be found
in his Introduction to Semantics^^. The same attitude is no less
sharply marked in Carnap's article of 1950, "Empiricism, Seperiod

mantics and Ontology"34, in which he to a large extent pursues


his idea

concerning the freedom of choice of the so-called Welt-

perspektive,

31

an idea that

an element of Kazimierz Ajdukiewicz's

R. Carnap, The Logical Syntax of Language, pp.

32 Ibid.,
p.

"

is

Cf. pp. 13

and

29.

and

24.

Revue Internationale de Philosophie, 1950, No. 11 (reprinted in:


Linsky (ed.), Semantics and the Philosophy of Language, Urbana 1952,
34

L.

XV

52.

pp. 210-212, 223-224, et passim).

46

Research Problems of Sema>jtics


j

radical conventionalism}^. It must, however, be stated explicitly!

no necessary junction between


these issues: the problems of syntax do not inevitably lead to
conventionalism, nor do they lose their meaning if separated
in this connection that there

'

is

from that philosophical background.


The same also apphes to the clearly philosophical tendency
;

endeavours are

the problems of the theory

to impart to logical syntax a universal character

made

to contain within

not only

it

of deduction, constructive in the

and even
to

was

syntax embrace
the

so-called

at that time

philosophy was
guage,

i.e.,

of syntax, but

all logical
;

from

freed

field

philosophical problems in general. Such a tendency

all

logical

let

all

all

the philosophical problems,

pseudo-problems

represented by neo-positivism.

left

Whatever of

it

no necessary junction between

must be emphathe study of the

syntax and such maximalist tendencies. Further, there

no such junction between


tivist

and

analysis

syntactical

is

philosophy.

idealistic trends in

philosophy,

it

def-

must be said that

all

those researches which are connected with the description of


linguistic signs

also

neo-posi-

While rejecting these tendencies, arising out of quite


inite,

.,

coincided with the logical analysis of lan-

with logical syntaxes. Here, too,

sized that there is

{Scheinprobleme),

and

'

expressions, with the study of the rules of

formation of such expressions from more elementary signs, with


the analysis of relationships between those signs,

and of the

rules

of transformation of expressions do deal with real problems of logical

Thus research problems of

syntax.

close to the

problems of syntax

This holds, above

4;

all,

syntax

come

grammatical interpretation.

for the issue of what

is

called descriptive (de-

35 Cf. K. Ajdukiewicz, "Sprache und Sinn", in Erkenntnis, 1934, Vol.


"Das Weltbild und die Begriffsapparatur", in Erkenntnis, 1934, Vol. 4;

'Naukowa perspektywa
in

in its

logical

swiata"

[A

Scientific

Przeglqd Filozoficzny. Vol. 37, 1934,


36 Cf. e.g.

321-323.

R.

Carnap,

The

Perspective of the World],

pt. 4.

Logical

Syntax of Language, pp.

XIII,

47

Logic

tailed) syntax,

The

concerned with the study of individual languages.

issues of pure (general) syntax are quite

formal in nature.

[They involve abstraction from the relations between signs and

on the one hand and their designata on the other,


form of expressions, i.e., their
types and the order in which they appear (which makes it possible to formulate the rules of building sentences and the rules
of deduction). Pure syntax belongs, of course, to the domain

[expressions

and the exclusive study of the

of metalanguage which must include names of expressions of


the object language,

names of

relations

between them

(in partic-

ular structural relations), as well as general logical expressions.

The

practical importance of pure (general) syntax can

seen in connection with the

work on the

translatability

now

be

of con-

languages into an abstract language of digital compu-

crete

ters (a

language which would enable such computers to

calculations in the binary system),

make

and with the construction

of machines for the mechanical translation of languages,

etc.

Thus, despite philosophical contaminations and distortions,


these are real research problems, needed, as

we can now

see,

not only in formal logic. True, a reading of treatises on the subject (e.g., the

now classical work of Carnap on the logical synmay give rise to doubt as to whether the issues

tax of language)

raised there are of

any practical

whether the symboHsm

and

in particular,

there used, so very intricate

and

care-

developed, does not occasionally contribute to obscurity.

fully
I

significance,

my

express here

doubts and mixed feelings experienced when

watching magnificent logical cranes so ingeniously built to Hft


pebbles which can

But

much more

easily

be

moved with

the hand.

dare not grudge, and a fortiori protest. Recent history has

taught us modesty and restraint in that respect.


tential calculus,

Now

that sen-

once seemingly useless and purely speculative, has

proved to have practical apphcations in the theory of


circuits,

now

indispensable

in

are gradually covering semantics

electric

and cybernetics have proved


the construction of digital computers which

that mathematical logic

by way of

their practical needs

48

Research Problems of Semantics

would be difficult to declare with conviction that research of any kind is useless just because we fail,
for the time being, to see its practical significance and apphcability. In that respect I fully agree with the interesting and correct formulation of the problem by Tarski.
"... I do not wish to deny that the value of a man's work
may be increased by its implications for the research of others
and for practice. But I believe, nevertheless, that it is inimical
to the progress of science to measure the importance of any research exclusively or chiefly in terms of its usefulness and applicability. We know from the history of science that many important results and discoveries have had to wait centuries before they were apphed in any field. And, in my opinion, there
are also other important factors which cannot be disregarded
Hence, I beheve,
in determining the value of scientific work
and applications,

it

. .

the question of the value of any research cannot be adequately

answered without taking into account the

intellectual satisfac-

who

tion which the results of that research bring to those

stand
I

to

it

and care for

think that

under-

it"37.

we should endorse

that opinion

when

passing

an analysis of research problems of semantics sensu stricto.


The period of syntactic euphoria passed under the influence

of two factors.
First,
is

not

it

soon became apparent that

sufficient, that it

pursuits.

Such

syntax alone

does not include a number of notions

that are indispensable for logical,


sophical,

logical

and the more so

scientifically

for philo-

important semantic con-

cepts as truth, denotation etc. cannot be formulated in terms

of syntax alone^s. This explains

why

at least a certain

branch

of neo-positivists tried to do away with them.


37

A. Tarski, The Semantic Conception of Truth..., quoted


and the Philosophy of Language, pp. 41-42.

after

L.

I.insky fed.), Semantics


38

M. Kokoszynska, "Logiczna

sktadnia j^zyka,

semantyka

wiedzy" [The Logical Syntax of Language, Semantics and the

Knowledge],

in:

Przeglqd Filozoficzny, Vol. 39, 1936,

pt.

1.

logika

Logic

of

49

Logic

Secondly, Tarski has


of truth

(in

shown

accordance with

that one can use the concept


classical

its

definition)

without

being involved in contradictions; this has broken the resistance

of the propounders of logical syntax to semantic concepts. Tarski


distinguished

in

metatheoretical

the

research

morphology

of

language, over which he has then placed other parts of syntax

and semantics sensu

stricto (i.e., the

problems of truth, meaning,

denoting, satisfying, etc.)

That opened the period of semantics which drove logical


its monopoly position. Semantics also took over

syntax from

the responsibiUty for constructing the theory of deduction in

a formalized way. The introduction of that theory has greatly

widened the scope of interest in

formerly confined to syn-

logic,

and has removed from many issues the brand of "pseudoproblems". In any case, there came a full development of the
study of problems of truth, denotation and meaning. Carnap
introduced corrections to his former ideas expressed in llw
tax,

Logical Syntax of Language in an annex to Introduction to Seman:

tics

he renounced his claim that "a

tensional character

modal

logic^o,

nominaMst

More

logic

of meaning

is

super-

This led to a change in his attitude towards the ex-

fluous'"^^.

thesis

detailed

of logic and to endeavours to formulate

and further to a certain modification of the


propounded by him and his folio wers^i.
of these issues will constitute the

analysis

subject matter of the second part of this book,

such problems as those of communication,

which

sign,

will discuss

meaning and

denotation, and permissibility of general terms in view of their

As

ambiguity and the danger of hypostases.


aspect of the problem,

i.e.,

regards the technical

the building of the various calculi

and a deductive system in semantics,

this is neither

the

task

of this book nor the subjects matter of our analysis. The reader

39

R. Carnap, Introduction

40

R. Carnap, Meaning and Necessity, Chicago 1947.

41 Cf.

to Semantics, p. 249.

Empiricism, Semantics and Ontology.

50

Research Problems of Semantics

who

interested in such matters

is

of the subject, above

quoted above

all

is

referred to the hteratureil

to the two works by Carnap already

Introduction to Semantics

>;

and Meaning and Neces-

sity.

The

theory of models, very interesting

from the philosophical


As compared with

point of view, has developed in semantics.

domain of purely

logical syntax, confined to the

and to the study of formal

ties

made

has

on the

and that which

"decree'"

between them, semantics

a step forward towards recognizing the legitimate

character of research
sion

relations

linguistic enti-

is

relation between linguistic expres-

expressed by them. The neo-positivist

on the elimination of apparent problems {Scheinpro-

bleme) tended to disqualify research of that kind by adopting

hmThe turn towards semantics constituted


linguistic monism and brought to the fore the

Wittgenstein's view that the hmits of one's language are the


its

of one's

world.

a breach in that

question of the object spoken about in a given language, the


question of something which exists outside language.

of models

is

The theory

a further step in the same direction, a step that

philosophically

important since

it

raises

the

problem of

is

the

object spoken about, the object which exists not only outside

language, but also independently of language.

apparatus of language
is

is,

it

The conceptual

were, "added" to the model. This

so at least in certain formulations of the theory of models.

The theory of models

As

as

indicated above,

is

shall

a product of mathematical research.

not deal with the technical aspect

of the problem, but, as in other similar cases, shall only refer


the reader to the hterature42.

The philosophical aspect

will

be

G. Kemeny, "ModeL of Logical Systems", in The Journal of Symbolic


March 1948, Vol. 13, No. 1; J. Los, "The Algebraic Treatment of the

42 J.

Logic,

Methodology of Elementary Deductive Systems", in Stiidia Logica, 1955,


same author, "On the Extending of Models" (I), in Fundamenta
Mathematicae. 1955, Vol. 42; A. Mostowski, "On Models of Axiomatic

Vol. 2; by the

Systems", in Fundamenta Mathematicae, 1952, Vol. 39; H. Rasiowa, "Algebraic Models of Axiomatic Theories", in Fundamenta Mahtematicae, 1955,

'

Logic

in

discussed

51

the next cJiapter; iiere, I shall confine myself to

a few remarks intended to provide general information. In diong


so I base myself

on Roman Suszko's "Logika formalna a niektore


Some Problems

zagadnienia teorii poznania" (Formal Logic and


iof Epistemology), availing

myself of his attempt to popularize prob-

lems of the theory of models, problems which are


their

mathematical formulation and not in the

The

starting

point

is

difficult in

least intuitive.

the following statement by

'"The opposition of a formalized language to

its

Suszko;

models

an

is

extremely abstract formulation of the cognitive relation between


the subject

which

is

and the

object, the relation

between thinking and that

the subject matter of thinking."

The concept of a model of language

is

meaningful with

erence to formalized languages. Every entity about

may speak

in a given language,

is

a model of that language. In

the case of simpler languages, every


the universe of the

model,

i.e.,

ref-

which one

model

set

consists ot

two parts

of individuals which

may

be objects of an arbitrary type, and the characteristic of the


i.e., a set the elements of which are properties, relations
and other aspects of the individuals of the model. The relation

model,

its model is such


model corresponds

between a language and

that

characteristic of the

to

an element of the

every constant ex-

pression of that language as the value of that expression, and

every element of the characteristic

is

represented by a constant

expression belonging to that language.

language has a number of models which belong to the

same type and about which one can speak in the given language.
But, as Suszko points out, the problem should be viewed from
another angle as well: there
Vol. 41;

is

a family of languages of a given

R. Suszko, "Syntactic Structure and

Semantical

Reference",

in

same author, "Logika formalna a niektore


zagadnienia teorii poznania" [Formal Logic and Some Problems of Epistemology], in Mysl Filozoficzna, 1957, No. 2&3; A. Tarski, "O poj?ciu wynikania logicznego" [On the Concept of Logical Consequence], in Przeglqd
Studia Logica, 1958, Vol. 8; by the

Filozoficzny, Vol. 39,

1936, pt.

I.

52

Research Problems of Semantics

model, connected with that model; these languages can be used


to speak
ture.
is

about that model and

Among them we

all

have the same syntactic

such which under given circumstances (reference to a

group and

its

struc-

meaningful languages, that

distinguish

social

perform the function of communication.

activity)

Such languages are obtained by "joining" appropriate conceptual


apparatus to the given model. That "joining" of a language is
a definite social action.

"We
stances

shall say that the


t,

is

to the family

language

meaningful under circum-

/,

a conceptual apparatus joined to models belonging

RMAt{J)

(i.e.,

to

a non-empty subfamily of the

family of models of a language that


given circumstances

A.

S.).

We

is

meaningful under the

assume that every language J

(meaningful under certain circumstances

t)

has come into being;

as a result of the operation, performed in a

human

group, of

joining conceptual apparatus to certain models of the language

J,

which include the proper model Mi{J)".


Suszko says further that the "joining" of conceptual apparatus to a model can take place in a social group only through
its

direct contact with that "model",

perception and practical action.

As

that

the

is

through sensory

model becomes

better

understood, the language in question becomes further developed.


This

is

a popular exposition of the intuitive ideas connected

with the theory of models. These ideas are closely connected with
the philosophical interpretation
in the following chapter.

of semantics, to

be discussed

Chapter Three

SEMANTIC PHILOSOPHY

" 'In the beginning

was the Word', says our version of

Gospel, and in reading some logical positivists


that their view

is

pithy,

am

St.

John's

tempted to think

represented by this mistranslated text".

(Bertrand Russell:

TfflS

Human Knowledge)

trenchant and excellent characteristic was applied

by Russell to neo-positivism, but in fact


ciple to the entire philosophical trend

it

was addressed

cognitive importance of language, as confirmed

speciahzed disciplines,

raises

in prin-

which, on the basis of the

by the various

language problems to the rank

of the only or at least the focal issue of philosophy. Thus Russell's

remark appUes to most trends and schools

semantics in

As

in

contemporary

the broad sense of the term.

stated earher, the discovery that language

is

not only

an object of research has been a great

Ian instrument but also

achievement in the development of science, in particular logic


sand mathematics. But
\

if

the merits of semantics (in the broad

sense of the word) are connected with definite problems of specialized disciphnes

and

their needs, its demerits are

linked with philosophy.


science does

for the

first

most

closely

time in the history of

poor philosophy prey on great

Such was the


also

Not

scientific discoveries.

fate of Einstein's theory of relativity,

such has

been the fate of semantics. The ideahstic interpretation

suggested by so-called semantic philosophy preyed on the dis-

covery of the role of language in research. The essence of that


interpretation consists of

making an extremely important, though

apparently small, step from the assertion that language also


is

an object of philosophical study to the assertion that language

alone

is

the object of such study.

[53]

54

Research Problems of Semantics

In discussions with representatives

of

the

Lwow- Warsaw

School, the assertion was often heard that such an objection


is

a wrong one, since no one has ever formulated such a

I shall

do

my

thesis.

best to demonstrate that not only has such a thesis

been formulated by various propounders of semantics, but moreover, that

it

constitutes the sense of semantic philosophy in

various shades.

admit that

its

I derived considerable satisfaction

in

opening the present chapter with Russell's words which were

so

much

ophy.

in agreement with

My

thesis is thus

my own

appraisal of semantic philos-

supported not by just a somebody, but

by Russell, not only an eminent philosopher but one of the


itual

contemporary semantics.
is

spir-

trend in research which has given rise to

fathers of that

His

understanding

of such

matters

thus exceptionally keen, and his appraisal, exceptionally

sig-

nificant.

The forebears of that

variation of idealistic philosophy which,

in recognizing language to

be the sole object of research wants

to eliminate the issue of reahty

and to regard the

conflict

between

materiahsm and ideahsm as a pseudo-problem, had appeared


in science long before Russell.

The same apphes

to the inter-

pretation of the role of language in the process of cognition.

necessary to go back to the nominalist tradition, above

It is

to

its

all

English version in the 17th and 18th centuries. I shall not,

my

intention

to write a study of the history of philosophy. In this

book we

however, enter into those matters, since

may

it

is

not

spare ourselves historical analysis, but must dwell on those

elements in the philosophical interpretation of semantics which


are cormected

with modern philosophical trends,

first

of aU

with conventionaUsm and neo-positivism.

But even that


the
at

more so

once facts and


1

do not want

as I have already

I principally

my

to discuss in too

much

had an opportunity

detail,

to present

opinions about them^. In our stormy times,

mean

the chapters dealing with a criticism of conven-

and neo-positivism in my work Z zagadnieii marksistowskiej teorii


prawdy [Some Problems of the Marxist Theory of Truth], Warszawa 1951,
tionalism

55

Semantic Philosophy

so full of political

and

may sometimes amount

ideological upheavals, five or six years

to

an epoch. That

my

in the essential points I support in toto

now under

is

why

and

discussion, published several years ago,

them to be a supplement to what

is

affirm that

criticism of th|| trends

here written.

consider

should like

however, to avail myself of this opportunity and to raise a number


of points which

I believe

to have been at fault in the

manner of

that criticism.
First of aU, in

my

present opinion the very style of that

criti-

cism was wrong.


It

often said of Marxist criticism that

is

empty-worded and should not confine


but that

it

is

it

should not be

to attaching labels,

should be documented and based on a profound knowl-

edge and honest


This

itself

presentation

of opinions

under discussion.

basic to all scientific ideological criticism, although

it is,

unfortunately, not always observed in Marxist literature.

There

are,

however,

regards the style of

other,

more
of

criticism. First

not consist of pure negation. Let

me

subtle,
all,

requirements as

such criticism should

explain that

There was a prevalent opinion in Marxist

more

literature

fully.

that

must disarm and annihilate the opponent; to recognize that some of the views of
that opponent might be correct was tantamount to "objectivism", which distorted analysis, and to succumbing to academic
if criticism is

to be ideologically engaged

it

The fear of being accused of academic objectivism weighed


on Marxist science especially after Zhdanov's well-known
1947 speech in the discussion on philosophy. People used to avoid

vices.

heavily

that objection in a very simple

way

they sought in the opponent's

views only what was false, passing over in silence what was true,

not to mention what, being new, was problematic and controversial.

Obviously, any opponent so processed could easily be

as well as the criticism of radical conventionalism in Poglqdy filozoficzne

Kazimierza Ajdukiewicza [The Philosophical Views of Kazimierz Ajdukiewicz],

Warszawa

1952.

56

Research Problems of Semantics

overhelmed.
there

is

no

It is

equally obvious that, especially in philosophy,

thus processing an opponent: for a ma-

difficulty in

ieriah% the principles of idealism are as absurd as for an ideahst


are the principles of materiaUsm. Thus,

if

the opponent's posi-

tion

is

reduced to such absurd (from our point of view) princi-

ples

and

this

easy triumph.
will

It

is

then we can score an

doubtful, of course, whether such criticism

convince anybody except those

and share the

can always be done

critic's

who

are already convinced

view. Then, the question arises as to

long such a game can continue,

how

remembered that the opwith


equal ease overwhelm
ponent can in the same manner and
his critic. All this has very httle in common with science and
scientific philosophy, but assists an understanding of the duration, hopelessness, and sterility of many philosophical disputes.
The issue would be clear enough, and a remedy for the disease
of criticism by denial not difficult to find, were it not based on
a mistmderstanding, which goes much deeper and is much more
of the essence. The conviction that the falsity of some opinions
expounded in a work cancels the correctness of all other theses
of its author arises from a specffic interpretation of the development of science as following a straight-line course in which
some theoretical systems represent pure truth, whereas others
represent pure untruth. This is nonsense, and can be refuted
by any confrontation with facts. Thus, myths can play a certain
role even in the history

if it is

of science.

In the history of science, progress

is

not a matter simply of

white against black. The notion that certain trends have a

nopoly of truth, and others a monopoly of untruth


acceptable.

It

is

mo-

simply not

would be only too easy to be a sage and

to hold

only true opinions in a world where the interpretation of the


progress of science was thus Utopian.
are

much more comphcated.

But

in

reality,

Truths, even those of great

matters
signifi-

cance, can be found in systems that are otherwise in error and

even

anti-scientific (an

example near and dear to the heart of

every Marxist: the dialectic method

and the Hegelian system).

57

Semantic Philosophy

Falsehoods, even very serious ones, can appear in systems that


are otherwise highly correct.
if we consider not only the results
new ideas and research problems posed
known that in science the placing a finger

All this holds a fortiori

of research, but also


to science. It

on the problem
solution.

well

is

And

often no less important than

itself is

where

pecially in philosophy,

of opponents. This

errors

is

positive

its

can be no monopoly

in that field there

at all, es-

on the

conflicting trends thrive

how new

problems, stimulating

the progress of thought, are born. In that respect, ideahsm certainly

has a great deal to

Does

it

its credit.

follow that the criticism of idealism

is

to be weakened,

and that the ideological engagement of philosophy


doned? Not in the
understood

be

The point

least.

is

adherence

properly:

is

to be aban-

that that postulate should


to

ideological

principles

should not be identified with clamorous words, and ideological

engagement with criticism by annihilation, merely in the


of the ideological engagement and, above
of criticism. If criticism
is

to perform

its

is

not to be a

proper task,

not yet convinced, then

it

rite for

interest

the effectiveness

the initiated, but

who

are

not be confined to denial.

To

i.e.,

may

all,

to convince those

say

"no", especially in philosophical matters where a direct

and

final verification

little,

is

usually impossible,

means very

since the opponent can reply with an analogous "no" and

stick to his
all,

of theses

own

views. Effective criticism should consist, above

in taking up the given issue, explaining its sense

in the

and

its

place

system of our knowledge, and when rejecting a certain

solution as in error

it

should suggest another, true, one. Such

an attitude has nothing to do with "objectivism" and "academism"

it is

simply scientific criticism properly understood.

ability to

The

in-

adopt such an attitude with respect to bourgeois ideology

has been, in

my

opinion, the greatest defect of Marxist

criti-

and therefore also of my own interpretation of conventionahsm and neo-positivism.


cism,

But

let

us

come back

to the point.

Research Problems of Semantics

58

have stressed many a time that when we speak about seman-

we cannot

tics

separate

its

from the

hnguistic or logical aspect

underlying philosophy which in this case

not only a shell or

is

a connective tissue of given research problems, but often the


very blood and bone. Yet efforts can be made, as in the preceding
chapters, to bring out those special research problems of semantics

and take them up regardless of philosophical

attitude.

We

might

and appraise critically those elements of


philosophical interpretation which have manifested themselves
thus far in the development of semantics. I shall attempt to do
also try to proclaim

that now.

What

I call

semantic philosophy, that

is,

all

those philosoph-

views which see in language the only, or at least the principal

ical

and the most important, object of cognition and philosophical


analysis, includes ideas hnked with the various trends of contemporary philosophy, from which semantic philosophy drew
inspiration

its

thesis that

and stimuh. Conventionalism accounts for the

language and

and ultimately

this

its

morphology, the system of

or that perspective of the world,

logic,

may

be

chosen in an arbitrary fashion. This


in the idea, arising

is consistently complemented
from neo-positivism, that language is the

only object of philosophical analysis. Finally, the marriage of


neo-positivism with pragmatism gave birth to the belief that

semantics can be a specific panacea in social matters.

In analysing these three points which, in

most important

my

opinion, are the

do not intend, as I have


already said, to give a full analysis and a complete evaluation
of conventionalism, neo-positivism and pragmatism (which
are made up of different, although interconnected, ideas and
in semantic

research problems).
role

my

am

philosophy

interested here in the concept of the

of language in the cognitive system, and


analysis to that issue.

positivism,
Circle,

since

was the

interested.

its

influence,

especially

greatest as regards the

will also

shall confine

Attention will be focused on neothat

problems

of the Vienna
in

concentrate on the period

which we are

more or

less

69

Semantic Philosophy

before 1936, when, under the influence of Tarski and pragmatism,

Carnap and others began to modify their views and at least


to discard the "pure" form of logical empiricism, so characteristic
of the Vienna Circle.

P
1.

THE ALLEGED "TURNING POINT" IN PHILOSOPHY.

LANGUAGE AS THE ONLY SUBJECT MATTER OF RESEARCH


Moritz Schlick, at one time the recognized philosophical
leader of the Vienna Circle, wrote in the

first

issue of the neo-

under the

positivist periodical Erkenntnis, in the leading article

significant title

"What, then,

"The Turning Point


is

philosophy? Well,

Philosophy":

in
it

not a science, but

is

something very important and great, so that even


considered the queen of sciences, although

down

not anywhere laid

itself

be a science. Nowadays we see in


trait

not

now
itself

it

can be

a science.

that the queen of sciences

It is

teristic

it is

it

and that

is

it is

must

the charac-

of the great turn in contemporary philosophy

not a system of results of cognition, but a system of acts. Philos-

ophy
is

is

an

asserted

activity

or

through which the meaning of statements

explained.

Philosophy explains sentences,

and

sciences verify them"2,

In the same issue of that periodical Rudolf Carnap accom[^


panied Schlick:

"The new trend of

this periodical, initiated

with this

issue,

sets

out to support a new scientific method of philosophical think-

ing,

ing

method which can be characterized most briefly as consistof a logical analysis of theorems and concepts of empirical

science ..."3

Vol.

Vol.

M.

Schlick,

"Die Wende der Philosophie",

in Erkenntnis,

1930-193L

in Erkenntnis,

1930-1931,

1.

1.

R. Carnap, "Die

alte

und die neue Logik",

60

Research Problems of Semantics

Carnap, as well as other followers of neo-positivism, was


consistent in carrying out that

programme.

few years

after

the philosopliical manifesto of the Erkenntnis group, mentioned

above, Carnap wrote in his fundamental work on the logical

syntax of language:

"Philosophy
is

is

to be replaced by the logic of science

that

to say, by the logical analysis of the concepts and sentences

of the sciences, for the logic of science


logical syntax

of the language of

nothing other than the

is

science"'^.

Let us examine that "turning point in philosophy" more


closely.
I shall

begin with a couple of

necessary

ever

for

remarks, which are how-

trivial

considerations.

future

There are two basic types of idealism objective and subjective.


The former recognizes the existence of objective reahty, which
is ideal by nature (the understanding of the term "ideal" varies
according to the various systems of objective idealism). The
:

latter sees in so-called reality

According to one of

its

only a construction of the mind.

variations, reality

is

tantamount to com-

binations of inner experience (the various versions of

immanent

empiricism, the classical representative of which was Berkeley).

That attitude

leads,

in

its

logical consequences,

According to the other variation, what


is

is

to

solipsism.

given in cognition

a construction of the mind, and the problem of reality

is disre-

garded as transcending experience (Hume's version of agnosticism).

So much for the sake of


Now, there is no doubt
"world" to language, to

recollection.

that

a theory

which

linguistic entities that are

limits

one's

an external

expression of one's inner experience, must be classed as a form

of subjective ideahsm.
tions of that trend

It is

comes

not important which of the two variain question in this case.

The point

R. Carnap, The Logical Syntax of Language, London 1937,

p.

Xm.

61

Semantic Philosophy

of consequence

is

leads to solipsism,

form of subjective idealism,

that this

and that a

man was

also,

found who had enough

courage to draw such a conclusion and to formulate

it in an
was
Ludwig Wittgenstein, one of the
exphcit way. That man

spiritual

set aside

my

undertaking,

few words

marked

particularly

for the time being

Vienna Circle and which accounts

in the

character of

logical empiricism. I

of semantic philosophy,

empiricism
its

which was

at least, with its other aspect

whole trend being called

for the

must now

the purely linguistic issues of semantic philosophy and

deal, in a

the

of neo-positivism.

fathers

Abrogating

empiricism.

strictly

mean

here

speaking the

most general explanation of that

proper understanding of the concept

issue is indispensable for a

of language in semantic philosophy.

Empiricism, in particular in

radical form, appears in a close

its

connection with subjective idealism from the

adopting the form of immanent philosophy,


ence as

The

inner experience.

classical

of empiricism were

interpretation

moment when,

interprets experi-

it

representatives of that

Hume and

Berkeley.

It

is

obvious that contemporary immanent empiricism claims to continue

their

Rehmke,

The

ideas.

immanentists,

who

Schubert-Soldern,

forms of immanentism,
neo-positivism.
icism,

in

trend includes not only

latter

often end in plain solipsism

The

last

particular

to

and

others),

(e.g.,

"pure"

Schuppe,
"shy"

but also the

such as empiriocriticism

and,

named

empiriocrit-

directly

Mach's

refers

doctrine

to

lately,

of elements,

i.e.,

impressions supposed to be philosophically neutral, the subjective

or objective character of which depends on the "co-ordinaorder" (Koordinationsreihe) (Avenarius). The

tion

programme

of the Vienna Circle 5, the subtitle of the periodical Erkenntnis,


the

retrospective

by

held

its

views

founders

Wissenschaftliche

of

the

origin

and outstanding

of neo-positivism
representatives

as
all

Weltauffassung. Der Wiener Kreis, Vienna 1929.

62

Research Problems of Semantics

indicate connections with

these

looked

further

still

Russell criticized
as

that

tences, to

it

its

all

to

Machism^. Neo-positivism

Hume and

Berkeley,

exponents for doing so

may, the

which

the edifice

back,

Be

insufficiently.

of protocol sen-

doctrine

neo-positivist

our knowledge

also

although

reducible, a doctrine basing

is

of knowledge on protocols of certain

elementary

experiences and impressions of the subject as the only content

of cognition,

is

a new, although very radical, form of immanent

empiricism.

From

the psychological point of view,

it

esting to analyse the strange fact that people

might by

who

part represented exact and natural sciences, people


to eliminate

all

for the

inter-

most

who wanted

metaphysics in favour of exact and reliable knowl-

edge, finally ended in

what

is,

in

my

opinion, radical metaphysics

bordering on solipsism.
Empiricism, which refers to the positions of clock hands and

of measuring

recordings
verifiabihty

based on

it,

of

human

must be

instruments,

controllability

and

perception and the knowledge that


attractive

and

of the natural science. The more so


rid

to

is

alluring to representatvies

if it

makes

it

possible to be

of the mysticism of the various irrationalisms that haunted

and which, beginning with phenomand neo-Hegelianism and ending in existentiahsm,


made all philosophy by their concepts of "Wesenschau",
"essence and existence", "das Nichts", etc. repulsive to every
the philosophy of the 1920s

enology

man who

thinks in a sober and precise manner.

And

yet Engels

at one time warned the naturalists that contempt for philosophy


most often leads to the worst philosophy possible. It turns out
that even the boundless confidence in clock hands and scales

of measuring instruments can lead to mysticism and metaphysics


See for instance: Ph. Frank, Modern Science and Its Philosophy,
Cambridge 1950 (in particular "Introduction: Historical Background"):
R. von Mises, Positivism. A Study in Human Understanding, Cambridge 1951;
5

H. Reichenbach, The Rise of


Press, 1956.

Scientific Philosophy, University of California

Semantic Philosophy

when

leads to the behef that the world

it

construction. This
for

tivists,

Little

elements

all

is

their

would have
or

anti-metaphysical

against the

directed

the neo-posi-

criticism

Koordinationsreihe,

and Empiriocriticism,

to

declarations.

added to the

to be

our product, our

is

what happened

precisely

Avenarius's

nin's Materialism
treatise

63

of Mach's

contained in

Le-

in order to obtain a critical

radically subjective,

doctrine of protocol sentences of the 1930s. For

neo-positivist

was rather the


form of the doctrine than its essential content that was changed.
Neo-positivism was born in a new period, and was founded
by new people. Especially were the natural scientists and the
representatives of the exact sciences,

who

it

gave birth to the Vienna

concerned with the precision and

Circle, subjectively

of science, with the elimination of

all

reliability

that which in traditional

philosophy shocked them by indefiniteness and vagueness, by

from questions wrongly formulated, with the

indecision arising

removal beyond the limits of science of

all

correctly called philosophical flapdoodle

For such people, and

in

that they often quite

and pseudo-problems.

such an intellectual situation, the

ditional, philosophical vulgar

forms of ideahsm, with

phraseology, were quite unacceptable. After


that against

which they had

stand contemporary materialism^,

7 I

think that

it is

all, it

was

rebelled. Since they did

they

all

tra-

their

precisely

not under-

chose that version of

a duty of the Marxists to correct certain inexact or

just erroneous opiaions

and neo-positivists in

which were formed with reference to neo-positivism


the past; they have worked to the detriment of the

cause by weakening the force of otherwise correct argumentation, and they

have done undeserved injustice to various people.


It

cannot be doubted that there

the materialist philosophy of

ophy of neo-positivism. But there


that a given thinker represents a

he

is

is

an

Marxism and
is

essential contradiction between

the subjective and idealist philos-

a non sequitur between the statement

wrong philosophy and the accusation

a political obscurantist and a political enemy.

of the

members of

It is

that

a fact that the majority

the Vienna Circle were politically progressive people,

friendly to the workers' cause.

Carnap was known as "the red professor"


communist by conviction.

("der rote Professor"), and Neurath was a

64

Research Problems or Semantics

idealism which offered

them

the appearances of precision and

rehable knowledge. Since he saw salvation in the replacement

of philosophy by the logical syntax of language, that is in fact


by an analysis of the syntax of the language of science, Carnap
wrote: "The step from the morass of subjectivist philosophical
problems on to the firm ground of exact syntactical problems

must be taken" 8. Yet,

in fact, this

ideahsm disguised as language

was again an

analysis, a

new

offer

of subjective

version of ideahsm

prepared by conventionalism (the role of language in

scientific

conventions) and by logistic (in particular the discovery of the


role of language as

an object of study and Russell's doctrine of

atomic and molecular propositions). Thus the combination of

immanent empiricism with a new concept of language and of


its

role in philosophical research gave rise to semantic philosophy.

That marriage was also

new

reflected in the

name of

original

the

trend: logical empiricism.

Thus semantic philosophy


estimates the role of language
cognition,

is

is

not a philosophy that over-

and

its

place in the process of

not just a philosophy of language, but

such a philos-

is

ophy of language which is genetically and organically connected with immanent empiricism. As such it has developed
as a variation of subjective idealism.

As mentioned above, the


and

its

role

and place

neo-positivist concept of language

in the process of cognition

had been

pared by conventionahsm, and also by the development of


guistic research

of a special type, connected with

prelin-

logistic.

In their Principia Mathematica, Russell and Whitehead dealt

not only with the problem of antinomies and the related problem

of levels of language, but also with the analysis of the structure


of language

in

which they distinguished

elementary

(atomic

and molecular) propositions as the category of propositions


of particular import, propositions that are the foundation of

R. Carnap, The Logical Syntax of Language,

p.

332.

65

Semantic Philosophy

all

Ludwig Wittgenstein, Russell's


atomism and, by imparting
already mentioned, a clear form of epistemological
our

knowledge 9,

disciple,

pursued the ideas of logical

to them,

as

solipsism

(we might speak here of linguistic solipsism), bridged the gap

between English

logistic

and continental neo-positivism of which

he was the spiritual father^o.

By coming thus

to Wittgenstein,

we have reached

point of our remarks on semantic philosophy.

the essential

His Tractatus

one of the strangest philosophical books, both in


view of the style and in view of the metaphorical and intuitive

is

certainly

manner of

exposition,

accompanied by claims to

precision.

indeed a paradox that the work which represents the trend

It is

whose intention

to eliminate

is

all

metaphysics

is

in fact the

twin brother of Bergsonian intuitionism and his metaphysical


conceptions. In Wittgenstein's Tractatus

we

find the fundamental

ideas of semantic philosophy: that language

is

the only object

of study; that the task of philosophy reduces to explaining the

meaning of the language of


exceeds that task

is

science;

and that everything which

meaningless metaphysics.

Many

years later,

Carnap, in his fundamental work, confirmed that dependence


of neo-positivists on Wittgenstein and expressed
with his basic ideas except for two that
:

late sentences

about syntax, and that

it is

full

solidarity

not possible to formu-

not possible to formulate

it is

sentences about the logic of science^.

Wittgenstein wrote in his Tractatus:


"5.5561 Empirical reality

The boundary appears again

is

limited

by the

totality

of objects.

in the totaUty of elementary proposi-

tions.

5.6

The

limits

A. N. Whitehead

bridge 1925, pp.


10

and

Compare

XVI

my

& B.

language

mean

the limits of

my

Russell. Piincipia Mathematica, Vol.

world.

1,

Cam-

flF.

the reminiscences of Ph.

Its Philosophy,
1

of

pp. 31

Frank

in

his

Modern Science

ff.

R. Carnap, The Logical Syntax of Language, pp. 282-284.

Research Problems of Semantics

66
5.61 Logic
limits

its

the world: the limits of the world are also

fills

...

What we camiot think, that we cannot think: we cannot


therefore say what we cannot think" 12.
Wittgenstein's position is in no way univocal, I have dehbbegun with

erately

acquires

known that such a statement


a definite meaning only when what is meant by an "object"
well

is

it

explained. But further paragraphs

is
is

statements on empirical reahty as the

his

totaUty of objects. But

my

the hmit of

world, since

tell

us that

we can say only

my

language

that which

think (in paragraph 4 Wittgenstein says that the

we can

thought

is

a meaningful sentence, and in paragraph 3.5 that the thought


is

an apphed sign of a sentence, about which we have thought).

In order to disperse any doubts as to his interpretation of the

enigmatic statement about language and the limits of

my

world,

on sohpsism.

Wittgenstein exposes his view

"5.62 This remark provides a key to the question, to what


extent solipsism

is

a truth.

In fact what solipsism means,

be

said,

but

it

shows

That the world


limits

is

my

5.63 I

5.64

my

am my

it

cannot

itself in

language which

the fact that the

understand)

mean

world.

The world and

5.621

quite correct, only

world, shows

of the language (the

the limits of

is

itself.

life

are

one.

world (The Microcosm)

Here we see that solipsism

...

strictly carried

out coincides

with pure realism. The / in solipsism shrinks to an extensionless


point and there remains the reality co-ordinated with
5.641 There

we can

is

talk of a non-psychological

The / occurs
is

my

in philosophy

/.

through the

fact that the 'world

world'.

The philosophical
'2

it.

therefore really a sense in which in philosophy

is

not the man. not the

L. Wittgenstein, Tractatus Logico-Philosophicus,

human body

London

1933.

Semantic Philosophy

human soul of which psychology treats, but the metaphysical

or the

subject, the limit


Is

67

not a part of the world"i3.

that not a classical illustration of Engel's statement about

who

the fate of those

dare to treat philosophy lightly?

Is there

more self-annihilating philosophy than solipsism?


It is both amusing and piquant to see an author who in the name
of combating metaphysics and vagueness (Wittgenstein says
a worse and

in p. 4.116 that everything that

can be expressed can be express-

ed clearly) introduces in a very obscure manner certain metaphysical

which moreover are not elements but

objects

limits

of the world.

On
courage

the other hand, Wittgenstein cannot be denied moral


:

few people in the history of philosophy have not been

Even Berkeley sought


refuge in objective ideaUsm. For to arrive at solipsism means
to reach philosophical self-annihilation, especially now, in the

frightened

epoch of

by the spectre of

brilliant

positivists did

solipsism.

development in natural science. True, the neo-

not repeat Wittgenstein's thesis (Carnap v/as the

only one to speak of methodological solipsism in his Der logische

Aufbau der Welt), but that part of Wittgenstein's heritage which


they continued to develop was most closely linked with solipsism.

In continuing

Russell's

theory of elementary propositions

Wittgenstein arrived at linguistic solipsism.

of that sohpsism
limits

is

The

essential sense

formulated in the sentence saying that "the

of my language are the limits of my world". The conclusions

drawn by Wittgenstein from that formulation came

to

play

a significant role in the further development of neo-positivism.


First

of

all,

Wittgenstein assigned to philosophy the task

of logically explaining (logically analysing) thought, which


his eyes is

"4.112 The object of philosophy

is

the logical clarification

of thoughts.

Philosophy

13

Ibid.

in

tantamount to a "criticism of language".

is

not a theory but an

activity.

68

Research Problems of Semantics

philosophical

The

work

consists

essentially

result of philosophy is not a

of elucidations.

number of

'philosophical

make propositions clear.


Philosophy should make clear and delimit sharply

propositions', but to

which otherwise

He

are, as

it

were, opaque and blurred"i4.

also says:

"4.0031 All philosophy


all in

the thoughts

Mauthner's

is

'critique

of language' (but not at

sense). Russell's merit

is

to have

shown

that

the apparent logical form of the proposition need not be

its

real form"i5.

Next, Wittgenstein reduced that explanation to the formal


syntactical aspect of language, completely dissociating

from

it

the semantic aspect.

"3.33 In logical syntax the meaning of a sign ought never


to play a role ;

being thereby

must admit of being established without mention


the meaning of a sign; it ought to presup-

it

made of

pose only the description of the expressions" 1 6.

what transcends those hmits was considered by him to be unsense, and consequently a pseudo-probFinally, everything

lem.

"4.003

Most

propositions

and questions, that have been

written about philosophical matters, are not false, but senseless.

We

cannot, therefore, answer questions of this kind at

only state their senselessness.

Most

all,

but

questions and propositions

of the philosophers result from the fact that

we do not under-

stand the logic of our language.

(They are of the same kind as the question whether the


is

more or

And

so

less identical
it is

Good

with the Beautiful.)

not to be wondered at that the deepest problems

are reaUy no problems"i7.

In the conclusion of the Treatise Wittgenstein writes:


14 Ibid.
15 Ibid.
16 Ibid.
17

Ibid.

Semantic Philosophy

"6.522 There
is

it

is

69

indeed the inexpressible. This shows

itself;

the mystical.

The

6.53

right

method of philosophy would be

nothing except what can be said,


science,

i.e.,

this.

To

say

the propositions of natural

something that has nothing to do with philosophy:

i.e.,

and then always, when someone


metaphysical, to demonstrate to

else

him

wished to say something

to certain signs in his propositions. This


satisfying to the other

he

were teaching him philosophy

had given no meaning


method would be un-

that he

would not have the feeUng that we


but it would be the only strictly

correct method"i8.

So much for Wittgenstein's philosophy. The numerous quoabove correspond to the importance of his phi-

tations given

losophy in the development of neo-positivism.

Now

let

us see

and developed by

illustrations

itative

how

his ideas

have been taken over, adapted

neo-positivists. I shall confine myself to author-

given by

way of example, without

delving

into erudite details^^.

In continuing the ideas of Russell and Wittgenstein, neo-

brought to a radical formulation the principle that

positivists

language
plicitly

is

the sole subject matter of philosophy.

And

in ex-

combining that concept of language with immanent

(logical)

empiricism they brought semantic philosophy to per-

fection.

What

are the arguments in favour of such

an interpretation

Most varied both direct and indirect.


They can be drawn from works of all representatives of that
of

neo-positivism?

18 Ibid.
19 Apart from the reminiscences referred to above those of Frank
and von Mises and the work of Reichenbach, the following books pertain

to the issue

now under

discussion:

J.

Jorgensen. "The Development of Log-

Empiricism", in International Encyclopedia of Unified Science, Vol.


No. 9, University of Chicago Press, 1951; V. Kraft, Der Wiener Kreis.

ical
2,

Der Ursprung des Neopositivismus, Vienna 1950;


ination

of Logical Positivism, London 1936.

J.

R. Weinberg,

An Exam-

Research Problems of Semantics

70

trend. Scanty exemplification

is

necessitated only by the consid-

eration explained above.

Even

in the early period of the

in Der

lated,

Vienna

Circle,

Carnap formu-

logische Aufbau der Welt, the theory of constitution

and the theory of methodological

Basing himself on

solipsism.

and

Russell's theory of atomic propositions

his theory of the

world as a logical construction consisting of many aspects, and

on Wittgenstein's radical formulation of


in fact

took the stand of a

Carnap
The theory

Russell's theory,

specific linguistic solipsism.

of constitution which lay at the foundation of Carnap's ideas


at that time,

affirms that

basic ideas.

The

conception

is

latter

joined

with "that which


'reality' to

'that

is

compound

ideas are deducible from

and

linguistic

by immanent empiricism) are

identical

(and in

Carnap postulates "a reduction of


and what has

given".

which

is

this point the logical

given' (das Gegebene),

been postulated, and partly carried out by Avenarius, Mach,


Poincare, Kiilpe, and above

maintains that

his

all

conception

Ziehen and Driesch"20. Carnap


is

neutral

in

the

"materialism

ideaUsm" controversy, that that controversy

versus

linguistic in

purely

is

nature and depends on the choice of fundamental

concepts, that

is,

on the choice of language; but

at the

same

time he urges the following:

"The theory of constitution and subjective idealism agree


that all statements about the objects of cognition can in principle be

ships

...

transformed into statements about structural relation-

The view

to sohpsism

that which

is

given,

is

my

experience,

is

common

and the theory of constitution. The theories of

consti-

tution and transcendental idealism agree in defending the view


that

all

the objects of cognition are constructed (in the idealist

language: 'created in thinking'); that

is,

the constituted objects

are an object of conceptual cognition only as logical forms built


in a definite

manner. In the

last analysis, the

same applies

basic elements of the constitution system"- 1.


20

R. Carnap, Der logische Aufboii der Welt, Berlin 1928,

21

Ibid., p. 249.

p.

3.

to the

Semantic Philosophy

71

That conception was called "methodological solipsism" by


Carnap. In
limits

of

this

my

interpretation,

Wittgenstein's thesis that

language are the limits of

my

'the

world' acquires par-

ticular significance.

Wittgenstein took over from Russell the idea that

all state-

ments are reducible to atomic propositions, of which we build


molecular propositions; higher-level propositions are built only
of those elementary ones. But neither Russell nor Wittgenstein
stated

tions

anything definitive about the nature of those proposi(it

was only under the influence of the

neo-positivists,

especially Ayer, that Russell along with the concept of

"atomic

propositions" introduced the concept of "basic propositions").

Now

on the part of the neo-positivists to make


Wittgenstein's conception more radical was the replacement
the

first

step

of atomic propositions by protocol sentences which, in the neo-

base the whole conception on the sub-

positivist interpretation,

jective experience of the subject22.

the

same

direction,

By making a

further step in

the neo-positivists linked that conception

with conventionalism. They rejected the idea that the protocol


sentences are a distinguished class, their choice

left

to convention.

In such an interpretation of the role of language in cognition,


Wittgenstein's thesis that the task of philosophy

make

sentences, to

is

to explain

"a criticism of language", and to reduce such

a "criticism" to an analysis of logical syntax, must have been

simply salutary to the conception of semantic philosophy. These

came

ideas consequently

to

dominate neo-positivism and to be

repeated in various forms and shades.

In taking

and others

up Wittgenstein's ideas, Schlick and then Carnap


shown at the beginning of the present section)

(as

saw, as the turning point in the development of philosophy,


the

replacement of

all

philosophical problems by the logical

syntax of the language of science.

22

C. G. Hempel,

Analysis, 1935, Vol. 2,

"On
No.

I shall

not illustrate that with

the Logical Positivists' Theory of Truth", in


4.

Research Problems ok Semantics

72

quotations since the most characteristic formulations have

al-

ready been cited above. These ideas have been repeated hundreds

of times in neo-positivist
cept that

what

literature, as

in philosophy

guage of logical syntax,

is

has also the related con-

cannot be formulated in the lan-

mere metaphysics, and consequently

a pseudo-problem, and unsense.

The

latter

divided

view was stressed in particular by Carnap.

He

problems into objective and

The

theoretical

all

former pertain to the objects of the

and are the

exclusive

pertain to the

domain of empirical

form of expressions

proper content of

scientific

traditionally covered

field

logical.

under investigation

disciplines.

The

latter

syntax) and are the

(their

philosophy. The objective problems

by philosophy

either are pseudo-objective

problems, and as such are translatable into the language of syntax,


or are just pseudo-problems, that

is

metaphysical unsense

(Carnap also worked out a complete theory of the

translatability

of sentences from a material into a formal mode,

i.e.,

which makes
to

eliminate

it

sentences

about things, and which consequently

provokes the impression that language, in


of

human

we

it,

fact, is the

only object

study).

"The material mode of speech


In using

a theory

possible to use sentences about sentences in order

is

a transposed

in order to say something about a

mode

word

of speech.

(or a sentence)

say instead something parallel about the object designated

by the word (or the fact described by the sentence, respectively)"^^.

No

wonder, then, that Carnap drew the following conclusion

with respect to philosophy:


''^Translatability

into

the

formal mode of speech constitutes

the touchstone for all philosophical sentences., or

for all sentences

one of the empirical


Everything else

2^

is

sciences"24.

meaningless.

R. Carnap, The Logical Syntax of Language,

24 Ibid., p.

313.

more

generally,

which do not belong to the language of any

p.

309.

73

Semantic Phtlosophy

"In

the

of metaphysics (including the philosophy of

field

value and the normative science) logical analysis leads to a negative

namely that

result,

the alleged theorems in that field are

completely devoid of meaning"^^.

be clearly seen

It will

the thesis that language


the field of philosophy,

how

is

the circle closes:

we began with

the only subject matter of study in

and we end with the

thing which transcends that scope of research

thesis that every-

simply meaning-

is

The negative method has thus confirmed

less.

beyond language can be

that nothing
I

do not engage here

istic theses,

since

it is

obviously

resulting

studied

in the appraisal

the assertion
in

philosophy.

and refutation of

not worth while repeating

trivial

ideal-

statements

from the opposition between materiahsm

and idealism (and arguments pertaining to certain selected

is-

sues will be found in the second part of this book). I deliberately

confine myself to stating certain facts.

The

essential fact bears out

the idealistic character of the philosophy which reduces the object

of cognition to inner experience in the case of empirical sciences


of protocol sentences; the meaningfulness of state-

(the theory

ments

depend on their reducibility to such sentences),


and to the analysis of the language of those statements in the
to

is

case of philosophy. It

is

on the estabhshment of that

fact that

the appraisal of the scientific significance of semantic philosophy

depends.
It

is

obvious that the demonstration of the

acter of a certain
in the eyes

of an

view does not in the

idealist.

On

idealistic char-

least disqualify that

the contrary,

it is

view

at that point that

and the idealist begins. But


which acknowledgment is denied

the conflict between the materialist

when
to

it

comes

idealist

to a situation in

views, such a statement concludes the initial stage

of the conflict. Such precisely


tic

is

the situation in the case of seman-

philosophy, and in particular in the case of neo-positivism

(logical empiricism).
25

That trend,

like

Machism before

it,

pro-

R. Carnap, "Uberwindung der Metaphysik dutch logische Analyse

der Sprache," in Erkenntnis, 1931, Vol.

1,

p. 220.

74

Research Problems of Semantics

claims

neutrality in the conflict between materialism

its

and attempts to demonstrate

ism,

by writing

ity

Now

ideal-

pseudo-problem and meta-

off that conflict as a

physical unsense.

and

alleged scientific superior-

its

these claims are completely groundless.

So-called semantic philosophy, while anathematizing metaphysics

which
itself

to include the

is

whole of

classical philosophical

problems,

sinks into traditional metaphysics, and of a very mediocre

Why make wry faces at Berkeley's subjective idealism


was much more consistent in that respect since he was
not ashamed of the relationship), while copying its ideas via
sort at that.

(Russell

the concept of protocol sentences?

when

physics,

own

one's

Why

thunder against meta-

philosophical attitude

is

based on the

completely metaphysical (in the traditional sense of the word)


thesis that

No

less

language

is

the only object of philosophical analysis?

metaphysical than the thesis about the exclusive existence

of the products of the individual mind

is

exclusive existence of linguistic entities

which are products of

And

the individual mind.


analysis,

such

immanent empiricism

is

the thesis about the

the thesis to which, in the last

combined with a

leads,

specific

philosophy of language and with agnosticism. Linguistic solipsism


is

as metaphysical as all other variations of solipsism.

As

indicated above, the demonstration of the idealistic char-

acter of the basic theses of semantic philosophy does not

that the holders of such theses will be convinced of error.

not think that that could be achieved

at once,

mean
I

do

even should we

bring into play the entire arsenal of materialistic argumentation.

These matters are

much

too complicated and include too

many

elements of various kinds to allow the controversy to be solved


so easily

and simply. But the refutation of the myth about the


character of semantic philosophy and the

anti-metaphysical

bringing to light of

sumptions
case.

is

much

its

openly metaphysical and

easier

and

at the

same time

The testimony of such a thinker

idealistic

as-

decisive in our

as Bertrand Russell will

be for us of exceptional importance in that respect. The same


reasons which urged

me

to choose Russell's

words

as the text for

Semantic Philosophy

this chapter,

75

my

quoting him at the end of the present

Inquiry into

Meaning and Truth Russell gave a de-

prompt

section.

In his

An

of the neo-positivist theory of protocol sentences.

tailed analysis

He

appraised negatively both the empiricist and the linguistic

aspect of that theory,

and wrote

in conclusion, in the chapter

entitled "Basic Propositions":

"When
is

say 'the sun

is

shining', I

do not mean that this


is no connot verbal, and for the

one of a number of sentences between which there

tradiction; I

mean something which

is

sake of which such words as 'sun' and 'shining' were invented.


The purpose of words, though philosophers seem to forget this

simple fact,

fit

my

dinner, I

do not want

into a system with other words, but to bring

ence

what

of food.
I

go into
words to

to deal with matters other than words. If I

is

a restaurant and order

about the pres-

could have managed without words, by taking

want, but this would have been

balist theories

my

less

convenient.

The

ver-

of some modern philosophers forget the homely

practical purposes of everyday words,

a neo-neo-Platonic mysticism.

the beginning

was the Word', not

word means'.

It is

lose themselves in

seem to hear them saying 'in


'in the beginning was what the

remarkable that

physics should have occurred in

and

this reversion to ancient

the attempt

to

meta-

be ultra-empir-

icar26^
I

have nothing to add to

We have so far been

this appraisal.

taking for granted that there exists a lan-

guage which we analyse. But the problem of the origin of that lan-

guage and of

its

relation to reality has been rather left in the

dark. Yet precisely in the system of semantic philosophy that

problem

26 B.

148-149

is

of extreme importance. The lack of a clear answer

Russell.

(italics

An
A.

Inquiry into /Weaning

S.).

and Truth, London 1951, pp.

Research Problems of Semantics

76

'

room

to that question always leaves

for a realistic interpretation

which might be as follows language ought to be analysed because


:

we may learn something about actual realmapped by that language. I have said explicitly that
interpretation might be such had we not dotted the fs and

through that analysis


ity that

the

is

crossed the
sible

in explaining the origin of language

hasten to add that

I therefore

the

?'s

and

its

pos-

connections with reality in the light of semantic philosophy.

?'s

all

the

/'s

have been dotted and

that frees us

from

all

all

manner

crossed by the founders of that philosophy in a

doubts. Moreover, only the introduction

of that issue enables us fully to understand the nature of semantic


philosophy. For whoever proclaims not only that language

is

the only object of philosophical analysis, but also that that language
is

chosen or created by us

in

an arbitrary manner, that

it

is

a result of an arbitrary convention with the change of which the


picture of the only reality that

is

accessible to us

in fact proclaims a radical variation

As we

shall see,

of

is

also changed,

idealistic

philosophy.

such reasoning leads directly to epistemological

sohpsism (stating that every individual can become aware only


of his

own

ideas),

which simply borders on ontological

solip-

sism.

2.

LANGUAGE AS A PRODUCT OF ARBITRARY CONVENTION


One of

is

the fundamental assertions of semantic philosophy

that language

consequently,
arbitrarily

is

a product of arbitrary convention and that,

when choosing

this

or that language,

we may

change our image of the world. This sounds strange

(to say the least), especially

when accompanied by

formulated by so-called radical conventionalism and


that those languages,

formed

in

its

sequel,

affirming

an arbitrary manner, are closed

with respect to one another and mutually untranslatable. The


resulting conclusion

is

not only that

we can change our image

of the world, and not only that there are different images of the

77

Semantic Philosophy

world, but also that there can be different and

of the world between which there

is

images

closed

no connection whatever.

Before proceeding to a philosophical appraisal of such


firmations, let us examine

of that
that
is

what

is

what

form of

strange

concerned here,

naive and

trivial,

is

their origin,

a conscious camouflage of idealism,

and does not bear

having recourse to other arguments,

know about the


As has been

what are the sources


For the claim

ideaUsm.

linguistic
is

af-

it

investigation.
is

Without

by what we

refuted

authors of those views.


said above, the authors of those conceptions

were mostly (especially as far as regards the neo-positivists)


people connected with the exact sciences, people for
theory of deduction was the symbol of perfection.

whom

And

it is

the

very

known what,

well

in the theory of deduction,

build a language" or "to choose a language";

what

meant by the statement that

is

and that the choice of language

is

is

also well

it

meant by "to
is

well

different languages

known

can

and resolution of

known what

is

meant by the

statement that there are "richer" and "poorer" languages,

Was

way not due

to the theory of deduction? In

opinion such was the case precisely.

Moreover, the

temptation

to

consider natural languages

to be arbitrary products of convention

There

sources.

of more or
(the

exist

less arbitrary

"language"

we

convention, such as certain signal codes

of flags

disregard the

we can

speech,

used

by

subtle

this sense,

represents a "language"

Does

ships),

ciphers,

the gesture

"language of flowers",

distinction

etc.

between language and

define language as a system of signs used in

communication. In
vention.

might also have had other

"languages" which in fact are products

language of deaf-mutes, the lovers'


If

etc.

the temptation to treat natural languages and their prob-

lems in a similar

my

exist

decisive for "perspectives of

the world" in the sense of the formulation


certain problems. It

is

human

each of the examples quoted above

which

is

a product of an arbitrary con-

that not increase the temptation to treat the nat-

Research Problems of Semantics

78

languages in a similar way, the more so since natural lan-

uial

much

guages have
It is

and

not

my

in

common

with

ones?

artificial

intention to oiTer here an analysis of the analogies

differences

between natural and

languages. Such

artificial

an analysis would imply an investigation of the relationship

between language on the one hand and thought and


the other. This

of which

is

an immense problem

now,

since

on

aspects

be dealt with in the second part of this book. There

will

however, one point to which

is,

reality

some

in itself,

in

my

opinion

must draw attention here and


a spot-light on the entire

puts

it

issue.

and analogies between


natural languages, one essential point must not be
In

all

artificial

and

it

the comparisons

artificial

and

lost sight of:

languages are always built on the basis of natural ones,

is

only on such a basis that they are possible and com-

prehensible.

This also refers to their arbitrarily conventional

nature: the conventions used in building the various languages


(the languages alike of deductive theories, of codes, ciphers, etc.)

are based

on

existing natural languages

without them. This

is

why, for

all

and would not be possible

the analogies

between the various systems of signs used

and

similarities

human communi-

in

cation, the attempt to extend to natural languages the conclu-

sions

drawn from the

error,

in

well exist

precisely

without

analysis of artificial languages

because
artificial

natural
ones,

languages

whereas the

is

basically

can

perfectly

latter

must be

based on the fonner and as such shine with reflected light only.

We

need not, therefore, study the relation to

reality

of

e.g.,

the

language of deductive theories or the language of the flag code


(although they too refer to the real world) because these languages
are

may

somehow

or other translated into

agree that in those cases

conventions and leave

it

we have

at that.

We may

natural languages.
to

not, however, conclude

that in the case of the natural languages also

from the
a

issue

conclusion,

We

do with products of

we may

abstract

of their relation to thought and reahty. Such

however,

is

suggested by neo-positivists (Witt-

Semantic Philosophy

Carnap,

genstein,

By reasoning
which

my

in

and

others)

and

71)

conventionalists.

radical

such a way one simply commits a logical error,

in

opinion,

the root of the conventionalist con-

lies at

ception of language.

The

starting point of conventionalism

cognition

scientific

is

the assertion that

is

based on a convention, and that building

up a science we in fact produce conventions which are chosen


from the point of view of their suitability. Yet Le Roy only was
the first to develop conventionalism consistently by stating that

theory depends on the choice of language^'^. This choice


there

no

is

is free,

necessity to choose precisely that language, although

the actual choice

made by a

given individual

is

conditioned psy-

chologically28.

That conventionalist
the

ism,

more

so

thesis

was transferred to neo-positiv-

as the neo-positivist conception of language

as the only object of philosophical analysis requires such thesis.

All non-conventionalist solutions of the problem of the choice

of language raise the issue of reality and of the relation "languagereality".

Conventionalist conclusions from the conception of language


as adopted

by the

were in their most radical form


drawn by Rudolf Carnap. In Poland,

neo-positivists

(within the Vienna Circle)

Ajdukiewicz's radical conventionalism took a similar direction


(although independently from the neo-positivists).

Carnap formulated his idea in the form of the principle of


which states a full freedom of the choice of language

tolerance

(both the formation rules and the transformation rules applying


to sentences)

and of

logic.

"In logic, there are no morals. Everyone

up

his

own

All that
27

is

logic,

i.e.,

his

required of him

own form of
is

that, if

is

at liberty to build

language, as he wishes.

he wishes to discuss

it,

E. Le Roy, "Science et Philosophic'", in Revue de Metaphysique

he
el

de Morale, 1899, pp. 529-530. 533.


28

et

E.

Le Roy, "Un positivisme nouveau",

de Morale, 1901, p. 144.

in

Revue de Metaphysique

80

Research Problems of Semantics

must

state his

methods

clearly,

and give syntactical

rules instead

of philosophical arguments"29.

That principle of

arbitrarily

selecting one's language (with

own

the consequences for the creation by the subject of his

all

the

image of the world


to

perspective of the world)

most sweeping conclusion

its

ventionalism. His conception


tivists,

is

was brought

in Ajdukiewicz's radical con-

by neo-posi-

related to that held

but without their theory of protocol sentences, and car-

ries the theses

of semantic philosophy to the extreme.

mean

here the theory of closed and untranslatable languages, chosen

an arbitrary manner on the strength of a convention. In

in

conventionalism, such a theory transforms

radical
into a

monad without windows, which not

only

the

is

subject

in a sense

maker of reality, but also is inaccessible to any argument


beyond the scope of its own language.
As Ajdukiewicz says:
"The fundamental thesis of ordinary conventionalism, represented for instance by Poincare, states that there are problems
the

which cannot be solved by appeal to experience unless one

intro-

duces a certain convention, since only such a convention, to-

makes it possible to solve the


The judgements which combine to make up
are thus not forced on us by empirical data
adoption depends partly on our recognition,

gether with experimental data,

problem

in question.

such a solution
alone, but their

since the said convention

which co-determines the solution of

the problem can be arbitrarily changed by us so that as a result

we

obtain different judgements.

"Tn the present paper

it

is

my

intention to

make

that thesis

of ordinary conventionalism more general and more

radical.

Namely, we want to formulate and to prove the theorem that


not only some, but

combine

to

all the

judgements which we accept and which

make up our image of

the world are not univocally

determined by empirical data, but depend on the choice of the


29

and

29.

R, Carnap, The Logical Syntax of Language,

p.

52

see also pp.

XV

Semantic Philosophy

81

conceptual apparatus by means of which we


those empirical data.

which

apparatus,

ceptual

We

make mappings of

can, however, choose this or that con-

change our whole image of the

will

world"^o,

Ajdukiewicz's view,

published in the Erkenntnis,

certainly

by the neo-positivist
supporters of semantic philosophy. But Ajdukiewicz was not
alone in his opinions which fitted Carnap's principle of tolerance
did not

and,

to influence the opinions held

fail

the theories of C. G.

e.g.,

HempePi.

Additional light on the issue of the choice of language

is

shed by the neo-positivist theory of physicalism^^.

is

how Carnap

This

is

"The

thesis of physicalism maintains that the physical language

describes that theory:

a universal language of science

that

is

to say, that every

language of any sub-domain of science can be equipollently


translated into the physical language.
science

is

From

this

it

follows that

a unitary system within which there are no fundamen-

tally diverse

object-domains, and consequently no gulf, for ex-

ample, between natural and psychological sciences. This


thesis

As I have already mentioned, neo-positivist circles


many people connected with the natural and the exact

30

is

the

of the unity of sciences"^^.

K. Ajdukiewicz, "Das Weltbild und die Begriffsapparatur",

included
sciences.

in

Erkennt-

same author: "Naukowa perspektywa swiata",


in Przeglqd Filozoficzny, Vol. 37, 1934, pt. 4; "Sprache und Sinn", in Erkenntnis, 1934, Vol. 4;
"Die wissenschaftUche Weltperspektive", in Erkenntnis,
1935, Vol. 5; "W sprawie artykulu prof. A. Schaffa o moich pogl^dach
nis,

1934, Vol. 4, p. 259; by the

filozoficznych" [Concerning the Paper


ical

Views], in

Mysl

Filozoficzna,

by

1953,

Prof.

A.

No. 2

SchaflF

On My

Philosoph-

(8).

''1

C. G. Hempel, "Le probleme de la verite", in Theoria, 1937, Vol.

^2

See

e.g.,

R.

Camap,

"Die

physikalische

Sprache

als

3.

Universal-

sprache der Wissenschaft", in Erkenntnis, 1931, Vol. 2; by the same author,

"Psychologic in physikaUscher Sprache", in Erkenntnis, 1932-3, Vol. 3; and

The Logical Syntax of Language ( 82: "The Physical Language"); O. Neurath,


im Physikalismus", in Erkenntnis, 1931. Vol. 2.

"Soziologie
33

R.

Camap,

Tiie

Logical Syntax of Language, p. 320.

82

Research Problems of Semantics

Hence

transform philosophy after the pattern

their tendency to

of the exact sciences, hence, too, their hking for physics and

its

language, a liking that often took a vulgarized form of behaviourism}'^ (cf.

Carnap's view on psychology, or Neurath's views

on sociology as a branch of physics). There is no doubt that physicahsm was a manifestation of the yearnings of naturalists.
But

this

caimot change the obvious and irrefutable fact that the

very idea of physicalism

is

conventionalist

from

its

inception,

was born from the concept of language as an arbitrary


convention, and that it is conceived as a convenient way of making
that

it

science

unified

and not as a concession

to

some form of

epistemological realism.

Conventionalism can clearly be found in


of the founders of physicahsm.

minimum

be confined to a

For

^4

Camap

My

many

statements

proof of that statement

instance, in his article "Psychologic in physikalischer Sprache"

"The

says that

thesis

stating that every theorem

will

be explained and substantiated below,

of psychology can be formulated

language

in the

that all tlie tlieorems of psychology spealc of physical


of physics
namely of a physical behaviour of man and otlier animals. This is a
.

will

of examples.

thesis of the general thesis of pliysicalism that

versal language,

language of physics

partial

is

a uni-

language into which every theorem can be transla-

i.e.,

tlie

events,

ted".

In another of his articles ("Die physikalische Sprache als Universal-

sprache der Wissenschaft'') Carnap generalized his thesis: "Our formulation


of the problem has often been called 'positivist': should any one be willing
to

do

so,

he might also

against such a term,

if

call

it

'materialist'.

one does not lose

No

objections

may

be raised

between early

sight of the difference

materialism and methodological materialism as a purified form of the former".

But somewhat

he explains the meaning of his "methodological

earlier

materialism" in a clearly conventionalist

ons may define the

thesis

as 'methodological materialism'.
that

we have

sibility

here to

do with

non-existence')

and not
of what

"In an analogous manner

The apposition 'methodological' emphasizes

theses

which exclusively

of performing certain linguistic

derivative theses,

spirit:

on the universal character of the language of physics

is

to

some

'given',

a logical pos-

refer to

transformations and

of deducing

'reality'

or 'non-reality' ('existence' or

what

"psychic",

is

what

is

"physical"'.

83

Semantic Philosophy

Moritz Schlick,

who was

much more moderate

supporter

of neo-positivist philosophy than were Carnap or Neurath (see,


e.g.,

arguments against the theory of coherence), had, long

his

before the birth of physicahsm, laid

its

epistemological foun-

dations.

"The

physical' does not

mean any

particular kind of reality,

but a particular kind of denoting reality, namely a system of

concepts in the natural sciences which

necessary for the cog-

is

nition of reality. 'The physical' should not be interpreted wrongly


as

an attribute of one part of

e.g.,

reality,

but not of the other

word denoting a kind of conceptual

rather a

it is

construction, as,

the markers 'geographical' or 'mathematical', which denote

not any distinct properties of real things, but always merely a manner of presenting

them by means of

ideas"35.

The founders of physicahsm took

precisely

by making their theory a sort of linguistic

trick,

that

direction

and transforming

problem.

in into a purely linguistic

Carnap wrote:
"It is easy to

and the

thesis

both

see that

on the unity of

(i.e.,

science

both the physicalist

A.

S.)

thesis,

are theses of the

of the language of science"36.

syntax

And the
syntax may
The

and

principle of tolerance states that language

its

be chosen in an arbitrary manner.

was dotted and

crossed by Neurath, the chief pro-

moter of physicahsm and of other radical neo-positivist

He combined physicahsm with the


imparted to the

theories.

theory of coherence and thereby

latter a purely linguistic

form.

im Physikalismus" Neurath defended


the thesis concerning the unity of science and the unity of the
language of that science. For instance, he interpreted sociology
as social behaviourism, comprising a component part of physics
in a broad sense of the term. And in what was the truth of the
In his paper "Soziologie

35

M.

36

R. Carnap, The Logical Syntax of Language,

Schlick, Allgemeine Erkenntnis/ehre, Berlin 1925, p.


p.

320.

27 i.

84

Research Problems of Semantics

theses of that science to consist? In the coherence of

its

sentences

as between themselves.

"Science as a system of statements


discussion. Statements are to be

is

always an object of

compared with statements, and

not with 'experience', or with 'the world', or with something

more or less subtle


new statement

All that meaningless doubling belongs to

else.

metaphysics and as such must be rejected. Every


is

to be confronted with existing ones, already brought to a state

of harmony between themselves.


correct

if

it

statement will be considered

can be joined to them"^^.

In the light of the principle of tolerance, radical convention-

ahsm

and

physicaUsm, the conventionalist character

caUed semantic philosophy stands beyond


ical analysis is confined to the analysis
is

all

main

As

theses.

doubt. Philosoph-

these are

do not intend
that would above all require

have indicated already,

to discuss those theses here since

so-

of language, and language

chosen on the strength of arbitrary convention

its

of

the solution of the problem of the relation: language-thinkingreahty.

It

serves our purpose here to demonstrate the ideahstic


is in no wise anti-metamust admit that this has been

character of semantic philosophy, which


physical. Every objective reader

done.

word by way of

conclusion.

My

sharp opposition to the

conventionahst thesis on the arbitrary character of natural languages, treated in abstraction from reality and from thinking
that reflects that reality, does not in the least
all

mean

the formulations of the idea of the active role of language

in the process of cognition, to


in the case of

be found in those theses (especially

K. Ajdukiewicz). Comparative

pological studies
in that respect,

(Sapir,

linguistic

reservations.

But these are different matters,

to be discussed again in the second part of this


2).
37

and anthro-

Whorf, and others) leave no doubt

and a moderate formulation of that idea evokes

no philosophical
IV.

that I negate

Neurath^ op.

cit.,

p. 403.

book (Chapter

Semantic PhilosophV

{^5

few more remarks, as a postscript, on the present develop-

ment of semantic philosophy.

One of

the

weak

points to this day in Marxist criticism

is

and transformations of the

the neglect of analysis of changes

opinions under discussion. If such neo-positivists as Carnap or

Ayer

at

one time expounded certain views

e.g.,

the theory

of protocol sentences or the reducibility of philosophy to the


logical syntax of language
fied

opinions either of

such

views are considered as petri-

all neo-positivists

or of some of them.

And

and willingly make


use of that right. As a result it often happens that we criticize
the one-time views of Carnap or Ayer, but do not know their
yet people are free to

modify

their views

present-day opinions.
It is

obvious that a

critic

may

take into consideration and

analyse a certain period or stage in the development of a giv-

en doctrine of theory. Scripta manent, and hence even those

and views which have long since been abandoned retain


historical significance. The logical empiricism of the Vienna

trends
their

Circle is an objective ideological phenomenon to be dealt with


by every one who studies neo-positivism, notwithstanding that,

as far as I

know, none of the former expounders of the theory

of protocol sentences (already


holds

it

to-day, at least in

investigates a

its

phenomenon

much

differentiated at that time)

original form.

The

obligation to point out that he

is

It is

is

who

under

interested only in a definite

stage in the development of the given theory,

theory in general.

researcher

that belongs to the past

and not

in that

also desirable, in such a connection, to

mention what was the further evolution of the theory in question.


This

is

just

what

intend to do now.

In the case of semantic philosophy

should, in a most general

way, describe the changes as a transition from subjectivism to

more

realist views.

86

Research Problems of Semantics

certain duality

and

was inherent

vacillation

in the theory

of physicalism, although Carnap's philosophical position in the


period of Der logische Aufbau der Welt or of his later

was

logische Syntax der Sprache

typical of semantic philosophy,

But

if

ability

we compare

it

work Die
was a position

quite univocal. It

as I have described

it

above.

with Carnap's views in the period of Test-

and Meaning, and even more so the period of Introduction

Semantics or Empiricism, Semantics and Ontology, then, while

to

noticing all the similarities

of semantic philosophy,

and the continuity of

we

shall

not

fail

certain concepts

to see that his position

has undergone a considerable modification.

The same
tivism,

A.

J.

applies to the English

Ayer,

if

we compare

expounder

his early works,

of neo-posi-

such as Lan-

guage, Truth and Logic (1936) and The Foundations of Empirical

Knowledge (1940) with The Problem of Knowledge (1956).


Among the many changes taking place in the views of the
expounders of semantic philosophy, we are here interested above
all

in the changes connected with the concept of language

its

analysis. Externally, this

is

an exclusive recognition of

and

manifested as a transition from

logical

syntax to the recognition

of semantics sensu stricto as well. The causes of that process

were explained in the preceding chapter.


point of view that transition

From

the philosophical

means the introduction of the problem

of the relation of expressions to the objects denoted by them

and therefore to the problem of meaning. In

spite

of lack of con-

and the
the problem of an objec-

sistency in the adoption of the classical definition of truth

point of view of semantics sensu


tive

stricto,

counterpart of language has nevertheless penetrated semantic

Of course, the issue can be made obscure and tangled


one does not take a consistently realist, that is materialistic,

philosophy.
if

stand in epistemology. This


ticists.

is

precisely the case of those

But the most important point

is

seman-

that these issues have

entered the field of vision of semantic philosophy, that semantic

philosophy has withdrawn from the magic

circle

of formal

lin-

Semantic Philosophy

87

guistic operations seen as the only object

of philosophical inves-

tigations.
is

It

obvious that such a step entails further consequences.

have been drawn by the so-called theory

In semantics, they

of models which, as already mentioned, gives a theoretical generalization to the thesis that language

models,

of some reality which

i.e.,

a mapping of some of

is

may

The theory of models has developed

expressions.

theory of deduction and

of considerable

is

one of

philosophical

its

its

find various linguistic

as part of the

instruments, but

an

as

significance

it

is

also

expression

of realist tendencies in semantic philosophy.


It

would be

difficult

continue to develop, the


theses in

one

now how

to say

more

may be accompanied by

field

pansion of such theses in another.


to the changes

now

My

duty

taking place and to

separate mention

is

that philosophy will

so since the rejection of idealist

due here to

the adoption or exis

show

draw

to

attention

their trend.

semiotic, a product of

cross-breeding of neo-positivism and pragmatism.

Semiotic as a general theory of signs has, as demonstrated

by Charles Morrises, a long history leading from the


through Hellenistic philosophy,

Occam

{scientia

Stoics,

sermocinalis),

and Locke (asfi.eioTt.xY]),


to contemporary mathematical logic. The roots of semiotics
in the form given to it by Morris go through neo-positivist influence precisely to that logical tradition, and also to American
Leibniz

(characteristica

pragmatism and
behaviour.
tic,

its

mean

universalis),

analysis of the influence of signs

here above

all

the

on human

works of Ch. Peirce on semio-

and those by James, Dewey and G. Mead.


There can be no doubt that the need of a general theory of

signs arose
38

Ch.

pp. 285

flf.

from the

W.

interest

shown

in syntax

Morris, Signs, Language and Behavior,

and semantics

New York

1946,

Research Problems of Semantics

88

by exponents of mathematical logic and neo-positivism. The


trends of reasoning which started from semantic philosophy
and consequently considered language to be the only, or at least
the principal, object of philosophical analysis, had to engage
in

an all-round analysis of signs in view of

their role in the prob-

lems of language.

Such a general theory of

signs necessitated taking into con-

and the human beings

sideration the relationship between signs

who produce and

perceive them. That aspect of the issue, in

was analysed by pragmatism


and behaviouristic psychology, that is by typically American
principle alien to logical research,

It is, therefore, not astonishing that a new


came to be represented by a man who in his
university studies was influenced at home by pragmatism, and
during his post-graduate research in Europe, by neo-positivist

trends in philosophy.
version of semiotic

Morris

ideas. Charles

which,

semiotic

is

now

the best

known

representative of

by distinguishing between the

in

which the semantic function of

signs

is

absorbed logical syntax and semantics as

As a

result, semiotic aspired to the role

capital

P.

the

syntactic,

semantic and the pragmatic aspect of semiosis (that

is,

the process

revealed) has in fact


its

component

parts.

of Philosophy with a

Morris wrote:

"Semiotic provides a basis for understanding the main forms

of

human

activities

activity

and

the activities
to fulfil

and

their

interrelationship,

relations are reflected in the signs


...

since

all

these

which mediate

In giving such understanding, semiotic promises

one of the tasks which traditionally has been called philo-

sophical. Philosophy has often sinned in confusing in

its

language the various functions which signs perform. But

an old tradition that philosophy should aim

human

own
it

is

to give insight into

and to strive for the


most general and the most systematic knowledge possible. This
tradition appears in a modern form in the identification of philosophy with the theory of signs and the unification of science.
the characteristic forms of

activity

89

Semantic Philosophy

that

with the more general and systematic aspects of pure

is,

and descriptive semiotic"39.

must be mentioned here that the representatives of semiotic

It

proclaim

philosophical neutrahty:

its

"Semiotic

itself neither rests

ticular philosophy.

on nor

science of signs

necessarily

imphes a par-

no more decides between

an 'empirical' and a 'non-empirical' philosophy than


between a
itself it

'naturalistic'

and a

'supernaturalistic'

cannot force one to beheve only

and prescriptions

In

verified

form one's

in the light of science. It will never-

have a profound influence on the course of philosophy,

theless

since

decides

religion.

scientifically

statements, nor to use only scientific discourse, nor to


appraisals

it

deals with topics peculiarly relevant to philosophic sys-

it

tematization

...

In this sense, the philosophy of the future

will

be semiotically oriented. But the nature of this influence will


not always be the same, and will depend upon the role which
given individuals and societies assign to scientific knowledge"40.

This
ogical

is

true in so far as formal, classificatory

considerations (and that

is

and terminol-

the principal subject matter

of semiotic) can be combined with different systems of Weltanschauung. Another point

more

reaUst attitude,

is

that Morris often adopts a

coming

close to materialism, than

colleagues engaged in the neo-positivist analysis

made

Reference might be

much
do

his

of language.

in this connection to the

paragraphs

devoted to the issue of meaning and of the universals in Morris's

article

still

clearer

in

The International Encyclopedia. His attitude

when

comes

it

is

to his analysis of the social aspect

of the process of semiosis.

The last-named

issue

is

of especial

interest.

Yet Morris's

views on the essence of signs and on semiosis will not be presented here, since they are discussed in greater detail in the second
part of this book, where the process of

and the role of signs


39

Ch.

W.

Ch.

W.

human communication

come

to the forefront.

Morris, "Foundations of the Theory of Signs",

of Unified Science, Vol. 1, No. 2, pp. 58-59.


Morris, Signs, Language and Behavior, p. 238.

tional Encyclopedia

in that process

in

Interna-

Chapter Four

GENERAL SEMANTICS

In his splendid exposition of the foundation of semantics in the

"The Semantic Conception of Truth and the Foundations

article

of Semantics" Alfred Tarski wrote:


perhaps worth while saying that semantics as conceived

"It is

in this paper (and in former papers of the author)

is

a sober and

which has no pretensions to being a universal


patent-medicine for all the ills and diseases of mankind, whether
imaginary or real. You will not find in semantics any remedy for
modest

discipline

decayed teeth or illusions of grandeur or class


is

conflicts.

Nor

semantics a device for establishing that everyone except the

speaker and his friends

not

It is

difiicult to

quant remark in the

is

speaking nonsense"!

who

guess

spirit

is

the addressee of that pi-

of the newspaper announcement:

"I repudiate responsibility for the debts

band";

it

is

abroad, but practically

Korzybski,

unknown

"The new methods ehminate or


genic

my

hus-

necessary only to read, for instance, the following

passage from a book by Alfred


literature

and actions of

blockages;

many

much quoted
in

alleviate different

'emotional

in

Poland:

disturbances',

semantoincluding

even some neuroses and psychoses; various learning, reading,


or speech difficulties, etc.; and general maladjustments in professional and/or personal hves. These difficulties result to a large

extent

from the

failure to use 'intelligence' adequately so as to

bring about proper evaluation.

A. Tarski, "The Semantic Conception of Truth

L. Linsky (ed.), Semantics

and

the Philosophv

p. 17.

[90]

...";

reprinted

in

of Language, Urbana 1952.

General Semantics

known

"It is well

as

some

that

91

many psychosomatic symptoms

heart, digestive, respiratory,

and

'sex' disorders,

such

some

chronic joint diseases, arthritis, dental caries, migraines, skin


diseases,

alcohoHsm,

to mention a few, have a semanto-

etc.,

and therefore neuro-semantic and neuro-linguistic oriIn general semantic training we do not go into the medical

genic,
gin.

angle as such.

and

in

We

vided the student

Knowing
I

is

word

use the

work

willing to

at himself seriously"2.

the opinions and the attitude of Polish representa-

of semantics,

tives

eliminate the harmful semantogenic factors,

most cases the corresponding symptoms disappear pro-

can

easily

advisedly

imagine Alfred Tarski's horror

on reading

that "bill of fare" of

problems covered by the respectable term of "semantics".

must be said that Alfred Korzybski also emphasized the difference between former semantics and what he
In fairness,

it

himself called semantics.

"There

is

a fundamental confusion between the notion of the

older 'semantics' as connected with a theory of verbal 'meaning'

and words defined by words, and the present theory of 'general


semantics' where we deal only with wewro-semantic and neurolinguistic

tions

to

living

reactions of Smithi, Smith2, etc., as their reac-

neuro-semantic and neuro-linguistic environments as

environments"^.

One cannot

fail

to agree, also, with the opinion of Anatol

Rapoport in his excellent informative article "What Is Semantics?"4 that there


a justified resistance both in academic
is

A. Korzybski, Science and Sanity, Lancaster (Penn.) 1941,

Ibid., p.

A. Korzybski's book

p. VIl.

X.

general semantics. It

is

is

quite useless as a source of information about

embarrassingly vague and dilettantish, and greatly

ov3rsized at that (nearly 800 pages). I can, however,

recommend

three ex-

from the collection Language, Meaning and Maturity, New


York 1954, which includes selected papers chosen first of all from the periodical
ETC for the period 1943-53. I mean the above mentioned wo!"k by Anatol
Rapoport, "What Is Semantics?" and two works by S. I. Hayakawa: "Secellent articles

92

Research Problems of Semantics

circles

and among the followers of Korzybski against

classing

semantics and Korzybski's so-called general semantics in the

same

intellectual current. In

any

case,

we cannot

tics,

which

eral

semantics which in the interpretation of

is

representatives

seman-

identify

connected with logic and epistemology, with genits

most

radical

a psychotherapeutic technique.

is

In certain matters there are genetic connections between

se-

mantics and general semantics, but they are superficial and confined rather to points of formal influence. In fact,

we have

to

do with trends and opinions that are not only different one from
another but often opposing one another, especially if the attitudes of their exponents are taken into consideration. That

why

the confusion of ideas

Marxist

literature,

on

is

that point, prevaihng so far in

a confusion to which even so rehable an au-

thor as Maurice Cornforth has fallen a victim (in his Science


Versus Idealism),

There

is

is

extremely embarrassing.

no doubt that

this state

of

affairs is

due to the search

for easy triumphs in the criticism of bourgeois ideology; this in

turn results in the tendency to stress whatever in the works under

review
is

is

correct

in error, ridiculous

and

and weak, and

to disregard

intellectually stimulating, since that

what

does not

fit

the "black-and-white" pattern. That way, one can pickle


any opponent, not to mention those whose views are such a cominto

bination of strange and often contradictory statements as

is

the

what is gained from such an


easy triumph over so mishandled an opponent? In my opinion,
nothing whatever. For in such cases the strengthening of our
negative attitude towards the views which we criticize is only
case of general semanticists. But

is not a good adviser on ideological issues,


and as a preventive measure against the influence of alien ideology,
it fails completely at critical moments. On the other hand, the

apparent. Ignorance

losses

we

suffer as a result of such manipulations are obvious

mantics. General Semantics and Related Disciplines" and

by Aristotelian Structure of Language ?"

"What

Is

Meant

General Semantics

93

I
and painful: we lose a certain amount of knowledge acquired
by mankind and, sometimes still more important, valuable stimuli to creative analyses. Books can be read in two ways. First
in search of

wrong and stupid in them, "prey" for critthe manner of reading typical of those who

what

icism this

is

is

do not know how to avail themselves of the achievements of


science. But one can also read them in a quite different manner,
looking for what is new, stimulating and calculated to promote
science, despite all the errors and falsehoods, which may be
quite numerous in given cases. This way of reading books is
characteristic of jJeople
If a scientific

work

who know how

includes even a single

to stimulate scientific inventiveness, then

not be allowed to be drowned

to make use of science.


new and creative thought

it

must not be lost, it must

in the ocean of errors.

That postulate must cause us firmly to


of

many

reject

our practices,

years standing, in the matter of criticism of non-Marxist

views in philosophy. I have referred to that matter above, but


those reflections return with redoubled force

when

v^e discuss

we

general semantics, precisely because in this connection

with particular clarity


icism,

our mistakes and shortcomings in

all

see
crit-

and sometimes cannot overcome an embarrassing susmany cases that criticism has been marked by

picion that in

ordinary ignorance. After


identification, over
tics"

many

all,

only ignorance can explain the

years, of semantics with general

the identification of the views of Stuart Chase (who

semanis

only

a popularizer of Korzybski's ideas) with the opinions of Carnap


or Tarski. Further, only ignorance can explain the failure to
notice the real problems that after all underlie general semantics.
It

is

those

difficult

who

to escape the impression that probably

none of

in our literature have written about Korzybski has

read his book thoroughly. This

is

not said in defence of that

book which, on the contrary, I hold to be, apart from all its
other defects, morbid and marked by monomania. But the criticism of
its

it

in our literature has

essential errors

and has

been at

fault

it

has not pursued

failed to discern the real

problems

94

Research Problems of Semantics


\

it

discusses. It

and Sanity
vastly

is

is

easy to demonstrate that Korzybski's Science

an obscure work of a

different

concepts

other people's ideas: that

it

and

dilettante, eclectically

unceremoniously

a book about which

is

amassing

assimilating
it

has been

and what is new in it.


is in error. Yet at the same time it might be shown how the book
takes up Pavlov's ideas, and that in some of its concepts it comes
close to epistemological reahsm and to the dialectics of the cognitive process. To say that it comes close to Marxism would prove
ignorance, but what has thus far been said about that book in
our literature proves ignorance in no less degree. And what shall
be said about works by some authors belonging to the school
of general semantics, such as A. Rapoport and S. I. Hayakawa,
already quoted above, or Irving J. Lee, Wendell Johnson and
said justly that

what

is

correct in

it is

old,

others?

There

is

no doubt that general semantics

When

the usual scientific standards.


ture

pf Dichtung und Wahrheit,

that

is

it

is

very remote from

studying that strange mix-

should be borne in mind that

a specifically American product, a sect rather than a school.

This is clearly realized by sober observers and even by


more sober among those who themselves are followers and

the
cel-

ebrants of general semantics. In his article already cited, A. Ra-

poport writes on that subject:

"The

accusations

of cultism

leveled

against

Korzybski's

followers are not altogether unfounded. In the United States


there

fs

a large floating population of 'truthseekers'.

Many

of

them lack the capacity of strenuous intellectual effort required


in a fruitful pursuit of knowledge and wisdom; others lack the
power of critical evaluation, which would enable them to tell
the genuine from the false. Still others cannot be comfortable
until they find a

'movements' and

panacea to beheve
cults.

They

in.

These people support

are as likely to 'go for' Cliristian

Science as for technocracy, for psychoanalysis as for theosophy,


for the Great

Books programme as for dianetics. And so insome of them among the adherents of general

evitably one finds

General Semantics

semantics

Whether they were

...

95

actually

helped

by general

semantics or by other factors cannot be determined without


controls.

sufficient

But they went about spreading the

faith,

thus giving a cultist flavour to the 'movement'''^.

General semantics therefore has


a sectarian

movement with

its

own

all

cult.

the

This

of

characteristics
is

author quoted above, an apparently sober and

confirmed by the
realistic critic

of

book
an obscure
work of a dilettante and that, contrary to declarations, it is in
no way a result of empirical research. And then he concludes:
Rapoport

that "sect".

states that Korzybski's

is

"If Korzybski cannot be said to have established

an empiriwhat then has he done? He has pointed a way toward


the establishment of such a science. He was a precursor of an
intellectual revolution which is just now begining and which promises to match that of the Renaissance. If Korzybski is seen in
cal science,

then the question of his originality or erudition

this role,

important.

He

is

not

might have something of a dilettante in him.

He

might have pretended to have more speciahzed knowledge than


he actually had. Great portions of his outlook might be found in

works of more modest and more meticulous workers,

the
is

men

are

all

"that

He

was a man of vision and an apostle. Such


too rare in our age of speciahzation"^.

not important.

way a member of the sect replaces a rationally thinkman! A very instructive phenomenon for an understanding of what is general semantics, and for ascertaining the secret
of its social successes in the 1930s, when the birth of Nazism was
In this

ing

accompanied by growing

interest in the influence

of propaganda

upon public opinion, and the influence of human communication


on the attitudes and behaviour of men. These circumstances
account for the rapid and violent career made in America by Korzybski's

conception.

ideas, since

A. Rapoport, "What

6 Ibid., p.

deliberately

refer

here to

Korzybski's

distinguish clearly between the attitude of the master

17

(italics

Is

A.

Semantics?",
S.).

p. 6.

96

Research Problems of Semantics

and that of at

least

some of his

The

continuators.

latter I

not only

value more, but also appraise them in quite different terms.

Their opinions and activity account for the fact that general
semantics

may

not be treated as mere bluff and shamanism while


:

criticizing in v^ith full force,

real

problems which

I shall

now

it

one has yet to see and examine the

has raised.

try to inform the reader very briefly about the

fundamental concepts of Korzybski.

He

himself defined general semantics as the science of neuro-

semantic and
to

neuro-hnguistic reactions of

"environment"

semantic

above).

(see

human

individuals

Rapoport defined

it

more comprehensible manner as the science of "how people


use words and how words affect those who use them" 7. in this
in a

can clearly be seen that part of semantics which Morris

calls

pragmatics.

For Korzybski, general semiotics


peutic technique covering, as
thing,

is

we have

from stomach ache and dental

above

all

a psychothera-

seen above, almost everycaries to social conflicts.

In his aspirations, Korzybski was guided by Freudian patterns.

The

on which he based

theoretical assumptions

his

ideas

were as follows.

The

starting point of Korzybski's interest

a general theory of
first

books.

From

man and

culture, the subject matter of his

that point of view he approached the role of

symbols in man's social

ogy of

and analysis was

life,

or, strictly, the issue

of the pathol-

signs.

Unfortunately, to recount these matters in abbreviated form

makes them look much more


were in

fact.

rational

Korzybski's conception

is

and serious than they


a strange conglomera-

of various theories drawn from the most diverse disci-

tion

plines.

And

the

main

idea, to

be observed in

tions, is quite striking: all organic

genic. "Listen to me.


1 Ibid., p.
8

and you

and

will

all his

social pathology

be saved!"

4.

A. Korzybski, Manhood of Humanity,

New York

1921.

is

explana-

semanto-

General Semantics

The
the

recipe to be used

obsolete

is

only too simple: one must reject

favour of a non- Aristotelian system

a system which

i.e.,

two-value principle of traditional logic

jects the
tic

system of language in

(two-valued)

Aristotelian

97

blockage, the cause of

all evil,

re-

and the seman-

will disappear

at

once.

Of

somewhat exaggerated.
The problem of the role of symbol in social life was drawn
by Korzybski from many sources, above all from behaviourism
and the Pavlovian theory (Korzybski considered his theory to be
course, his idea, as formulated here,

is

a development of the theory of conditioned responses). Freud-

ianism served him as inspiration for a specific conception of the

pathology of symbols. This was the point where the theory of


a non-Aristotelian

technique set

That conception
is

of

language and the related psychotherapeutic

in.
is

Korzybski's theoretical contribution. But

must be asserted that


all its

are

ideas

in

it is

no way

original. Birth certificates

elements can easily be produced, although the original


often

completely

mystified.

Let us begin with the general concept of a non-Aristotelian

language system. The existing system of language which Korzybski calls Aristotelian (his

more reasonable

followers, e.g., S.

I.

Hayakawa, speak of an Indo-European system of language)


is alleged to impose a two-valued system
of appraisals (e.g.,
"You either are a communist, or you are not"), which is said
to

evoke certain neurotic ailments.

On the contrary,

a non-Aristot-

ehan system, supposedly connected with contemporary science


(in his

arguments Korzybski often

whatever),

is

ideas taken
logic

flings

said to have

an

infinite scale

from Lukasiewicz and

can clearly be traced (Lukasiewicz

common

into

is

any system

many-valued

mentioned in the bib-

manner

that has nothing

with the original concept and without any

fication for their


1

fit

of appraisals. The

his theory of

liography), but these ideas are used in a


in

about such terms as

"quantum", without making them

"colloidal",

new

applications.

justi-

98

Research Problems of Semantics

But

new

this is

First of

the use of the copula

all,

"The word

hibited.

On

only the beginning of the issue, since here

we

learn

about that non-Aristotelian system of language.

details

not the thing

is

the "non-verbal level"

we say "This

is

a table"

we can
we are

for identification

is

is

pro-

denotes", says Korzybski.

it

when
word with

only indicate things, and


said to identify the

the object, which gives rise to serious semantogenic pathology.

Here again we recognize the source from which "the non-verbal


was borrowed credit for that concept belongs -to the neo-

level"

But with Korzybski

positivists.

normal

man

identifies the

and the

suggests,

it

word with

becomes a

No

caricature.

manner he

the thing in the

cure, consisting in repeating: "This

is

not a ta-

ble" or the appropriate mechanical operations of the "structural

invented by

differential"

The second

of shamanism^

Korzybski, savours

principle affirms that "a

map

not

is

which implies that a sign cannot aspire to be

The

tive.

istic",

plete

Aristotelian system of language

is

a'

territory",

fully representa-

said to be "elemental-

to break cognition into elements which pretend to be com-

and

absolute.

And

the non-Aristotelian system

is

"non-

elementahstic", recognizing the necessity for grasping the whole

of the process of cognition. This obviously goes back to

Gestaljz-

psychologie.
Finally,

the

third

of symbols. Not only


of the mapi

is

principle
is

map

affirms

"multi-ordinality"

map

not the map[. In other words, the language in

which we speak about another language


Consequently,

the

not a territory, but also the

we must

is

and avoid ambiguity by placing words

we may know on what

not that object language.

take note of the hierarchy of languages

level

in clear contexts so that

of abstraction

we

difficulty in identifying here the idea suggested

are.

We

find

no

by Russell's the-

ory of types and the related concept of the hierarchy of languages.

article

penetrating criticism of those views

is

given by

"Korzybski's General Semantics", included

and Philosophy,

New York

1949.

in

Max

his

Black in the

book Language

General Semantics

99

Starting from these theoretical assumptions, Korzybski undertakes, in his book,

an analysis of the structure of language,

who

by passing from the language of a person

is

mentally der-

anged to the language of higher mathematics. His purpose

is

to

to

consolidate the

would enable him


of language and to establish

its

relation to

This

discover a psychotherapeutic technique which

sum of

the

new structure
human behaviour.

social health;

hence the

supposed to increase

is

title

of the book

Science

and Sanity.
Korzybski also made a number of concrete proposals intended to modify language by imparting to

it

a "non-Aristotelian"

structure.

First of
eral

nouns

he maintains, indices should be added to gen-

all;

in order to

remove the mystifying conceptuahzation

of classes of objects, and to emphasize the specific and unique


character of the individuals. According to Korzybski,

if

we

say

not "Negro" in general, but concretely "Negro i", "Negro2",


etc.,

we

shall

remove the causes of such a

social

phenomenon

as racial hatred.

Secondly, dates are to be added in order to distinguish the

phenomena and thus to eliminate incorrect


generalizations. Thus, we should not say e.g., "Wilham Shakespeare" in general, but "Wilham Shakespeare January 1600",
various phases of the

"William Shakespeare January 1601",

etc.

Thirdly, all sorts of characteristics should he supplemented

with the

word

"etc.", in order to

remind us that the "map" does

not represent the whole of the "territory", or, in other words,


that

we never

looking

title

semantics:
Finally,

achieve

full

cognition (this explains the strange-

of the principal periodical of the school of general

ETC).
Korzybski recommends the use of quotation marks

to indicate that

we

dissociate ourselves

from the actual meaning

of certain words.

As a remedy, Korzybski recommends exercises with an instrument he has designed and called a "structural differential";

100

Research Problems of Semantics

make

these exercises are supposed to

word

that the

is

words occupy

that

Does

different

brief

this

is

new,

is

correct,

is

it

refers,

first,

and, secondly,

of abstraction.

levels

summary confirm

in Korzybski's theory

what

the patient realize,

not the thing to which

the thesis that whatever

old and well

is

known; and

in error?

do not comment on the therapeutic aspect of the "semantic"

measures. Psychogenic ailments undoubtedly are an extremely

important domain of medicine and

it

may be

that in

some

cases

putting rods into the appropriate holes of the "structural differential"


Is

may

help.

But that

is

a separate

issue.

Korzybski's general semantics, then, just monstrous non-

sense to be

thrown

in the dustbin? Is

it

only a deliberate ideahstic

would appear
from many Marxist publications? I am not at all sure.
For all its oddity and its simply maniacal traits, Korzybski's
mystification dictated

by

class considerations, as

conception includes something which cannot be dimissed


ly.

It

may be

that that

light-

comes from the borrowings which

in

Korzybski's case are innumerable. But precisely that "something"


has

made

it

possible to estabhsh an influential school of general

semantics, which, apart from sectarians, includes people

who

are

voices in science, especially in hnguistics. That "something", too,

induced certain outstanding

Malinowski, P.

scientists,

W. Bridgman,

among them Bronislaw

Bertrand Russell and others, to

endorse Korzybski's book with their approval of his ideas and


with a favourable appraisal of what he had done. Let
point

it

me

also

out to those with a liking for easy triumphs, that from

the philosophical point of view Korzybski

nut to crack precisely for a Marxist

There

is

no doubt

that

all

is

sometimes a hard

critic.

his ideas are

marked by ascribing

an absolute value to the function of language. This in particular


stands out in relief
ski

would

when

it

comes to

social issues

which Korzyb-

also like to reduce to semantogenic perturbances.

instance, he quite seriously tried to treat the

munism and Fascism

in terms

For

problems of Com-

of neuro-semantic

reactions to

definite signalsio.

That notion, which for obvious reasons was

continued by Stuart Chase in

his

at-

was very seriousThe Tyranny of Words^K

tacked with particular force by Marxist


ly

101

'

General Semantics

critics,

But also in Korzybski's works we find understanding of the fact

perform

that language can

That idea

reality.

is

function correctly only

its

it

if reflects

the guiding principle of Korzybski's

cam-

modifying the structure of language, and that

paign in favour of

up by his continuators (e.g., Stuart Chase,


Hayakawa, Wendell Johnson). This is what
Korzybski himself says on the subject:
"As words are not the objects which they represent, structure
and structure alone becomes the only link which connects our
verbal processes with the empirical data. To achieve adjustment and
sanity and the conditions which follow from them, we must
study structural characteristics of the word first, and, then only,
very idea was taken
Irving

J.

Lee, S.

I.

build languages of similar structure, instead of habitually ascribing to the

word

primitive

the

structure of our language...

Moreover, every language, having a structure, by the very nature


of language reflects in

its

own

structure that of the

word

as as-

sumed by those who evolved the language. In other words, we


read unconsciously into the word the structure of the language
we use. The guessing and ascribing a fanciful, mostly primitive
assumed structure to the word is precisely what 'philosophy'
and 'metaphysics' do. The empirical search for word structure
and the building of new languages (theories), of necessary, or
is, on the contrary, what science does. Any

similar, structure,

these structural peculiarities of language

one who

will reflect

upon

cannot

miss

semantic

the

point

that

uses the only correct language-method.


order, while metaphysics

and ultimately a
-

10
11

It

the

scientific

method

develops in the natural

of every description uses the reversed,

pathological,

order" 12.

A. Korzybski, Science and Sanity, pp. XXIII-XXIV.


New York 1938. Stuart Chase repeats that idea in

his later

work,

The Power of Words^ New York 1954.


i2 A. Korzybski, Science and Sanity, pp. 59-60.

Research Problems of Semantics

102

Are these ideas correct or not? Personally


not only correct (their formulation here
tance), but even of great significance

is

think they are

of secondary impor-

for a criticism of idealism

and of semantic philosophy. Consequently, they are something


which does not

fit

well into traditional patterns

The more so

formulations.

certain accidental ideas, but the ideas to

dozens of pages and which he placed

The more

conception.

and hackneyed

since with Korzybski these are not

which he devoted many


root of his

at the very

so, further, since these ideas

have been

taken up by his continuators of the school of general semantics.


Says Wendell Johnson:

"The

crucial point to be considered in a study of language

behaviour

the relationship between language

is

and

tween words and not-words. Except as we understand

we run

ship,

reality, be-

this relation-

the grave risk of straining the delicate connection

between words and facts, of permitting our words to go wild,


and so of creating for ourselves fabrications of fantasy and delusion

'13

Why, even

Stuart Chase, clearly bent

"What

the semantic discipline does

the picture

and create a new

One

on producing a

"best-

same:

seller", says the

is

to

blow ghosts out of

picture as close to reality as one

no longer dogmatic, emotional, bursting with the


rights and wrongs of it, but humble, careful, aware of the very
considerable number of things he does not know. His new map
can

get.

is

may be wrong;
better

his

judgments

by happenings

is

judgments may

But the probability of

err.

greatly improved, for he

in the

is

now swayed more

outside world than by reverberations in

his skuir'i4.

"Good language
things behind the

of the world

we

alone will not save mankind. But seeing the

names

live in.

ty

W. Johnson,

14

S.

People

will help us to

Good

in

language

Quandaries,

Chase, The Tyranny of Words,

understand the structure

will help us to

New York

p.

206.

1946. p.

communi-

113.

General Semantics

cate with

one another about the realities of our environment, where

now we speak

darkly, in alien tongues"i5.

But, ye Gods!, that

is

reasonable! Contrary to

and
to

expecta-

That

radical conventionalists could not understand.

cannot be simply erased; that accurately smashes

made

all

people do understand certain things which neo-posi-

tions, these
tivists

103

conform

And what

evaluations

all

to the "black-and-white" pattern.

done about the

to be

that

Korzybski,

by declaring himself in favour of a mobile, against a

static, inter-

is

fact

comes close to the well-known


some adherents of dialectics? This,

pretation of reality, occasionally

argumentation of at
too,

is

least

not coincidence. In rejecting the "Aristotehan" structure

of language, Korzybski

same time opposed

short of a volume, to revise

"It is impossible,

and to formulate

at the

is

an

(i.e.,

which would be

and our nervous system; but

mentioned, even here, that the 'law of identity'

third' as

acter

to

it is

The

sometimes

is

this

'logic'

S.)oo-

A.

structurally
it

must be

never appli-

'law of excluded middle' or 'excluded


called,

establishes,

'logic',

non-Aristotehan

valued, non-elementalistic semantics


similar to the world

cable to processes.

to the tra-

of the so-called laws of thought and says

ditional formulation

represents only a limiting case

which gives the two-valued chara

as

and

general

what
must

principle,

so, as a general principle,

be unsatisfactory" 16.

But "the worst"

is

still

ahead of

us.

expressis verbis to the intellectual kinship

the Pavlovian theory,

and

tent justified in his claim.

Now

my opinion he is
am even inclined to

in

Korzybski

between

refers

his ideas

and

to a certain ex-

believe that

Ko-

rzybski formulated his ideas under Pavlov's influence. Korzybski himself

speaks of congeniality, and asserts that he became

acquainted with Pavlov's ideas only after he had formulated


his

own non- Aristotelian

system i''.

15

Ibid., p. 361.

A. Korzybski, Science and Sanity,

17 Ibid.,

pp. 315-316.

p. 405.

Research Problems of Semantics

104

"My
to a

structural,

linguistic,

new and important enlargement of


that these independent discoveries

each other

man

the application to

of the Pavlov experimental theory of 'conditioned'


fact

leads

theoretical revision

non-el.,

reinforce

reflexes.

The

and support

a striking instance of the usefulness of theoretical

is

researches"i8.
I

do not know whether Korzybski knew the Pavlovian theory

of the second system of signals, but what he writes on the subject


(e.g.,

on pp. 331-2) resembles that theory even by the

minology he
is

If that

uses.

striking in

is

But even

itself.

if

problem remains

influence, the

Korzybski wrote under Pavlov's

more so

The point

intricate.

concept of semantogenic blockages in


fectly well fits into the theory

ter-

really congeniahty, then the fact

is

that the

human behaviour

per-

of conditioned responses, and the

into the theory of the second system of signals.

The

explanation of neuroses by the mechanism of the conflict between


stimulation and inhibition
is

at stake there, is

comes

precisely

from Pavlov. What

not only the signals which cause conditioned

responses, but also the signals of signals, that


reject

the odd form of

work the

real

is

words. If

problem of semantogenic disturbances' remains.

Moreover, that problem apphes not only to the psychiatric


there

is

we

the pathology of signs in Korzybski's

level

the issue formulated, as mentioned above, by Rapoport

"How

how words
Hayakawa,
on
the
outer
of his
affect those who use them" (S.
book, put it in slogan form: "How men use words and words
in the following

way:

people use words and


I.

use men".).

Thus,

Korzybski's idea

is

concerned with

many

real research issues.

This

is still

followers

in

more

the

true with reference to

school

of general

Their undoubtedly rational idea


matic study of language, that

is

is

some of Korzybski's

semantics.
the

programme of a pragfrom the

the study of language

point of view of the producers of language signs. Rapoport formu-

18

Ibid., p.

326.

General Semantics

manner: grammar

lates the question in the following

investigates

between words; logic studies relations between prop-

relations

ositions; semantics studies the relation

ositions

105

and

between words and prop-

and on that

their referents,

basis establishes their

meaning and truth; and general semantics goes furthest of all


since it also investigates the influence of words and proposi-

on human behaviour:

tions

"For a general semanticist, communication

words in proper order properly

not merely

is

inflected (as for the

grammarian)

or assertions in proper relation to each other (as for the logician)

or assertions in proper relation to referents (as for the seman-

but

ticist),

all

these together, with the chain of 'fact to nervous


i^.

system to language to nervous system to action' "

This clearly reveals the influence of pragmatism, working


principally through semiotic. Certainly, but

that general semantics has developed in the

This

is

no coincidence

the same trend of influence as in the case of Morris's

is

semiotic,

much

it

home of pragmatism.

although here the research programme

different,

is

broader, and going beyond the general theory of signs.

In

its

rational form, general semantics refers

interest revealed

by such

and

cultural

above

factors

of

all

human

to increased

disciplines as psychology, psychiatry,

anthropology,

behaviour.

in

"neuro-linguistic"

Adherents of general semantics

claim that precisely these disciplines provide proof of the fact

human

that

experience consists in selecting

among

the stimuli

coming from the environment, and human behaviour,


izing

experience

in

conformity

with

these functions depend in a definite

definite

way on

in organ-

patterns.

language, on

its

Both
struc-

linguistic habits. Hence the importance of concrete


on language and the programme of such research.
It would be impossible to deny the importance of this research, regardless of our attitude toward this or that justification
of its programme.~And here is a brief specification of problems.

ture,

and on

research

19

A. Rapoport, "What

Is

Semantics?",

p.

14.

Research Problems of

106

SeiCiantics

come the issues connected with human communicaAs


we
already know, the original success of general semantion.
tics was due to its interest in the rapid progress of Nazism in the
First

This

1930s.
in

gave

rise

to

an interpretation of social

conflicts

terms of semantic perturbations. That trend was initiated

by Korzybski, who was followed by Stuart Chase, S. I. Hayakawa, Wendell Johnson, and others. It must be said that as early
as 1949 S. I. Hayakawa, in the preface to the new edition of his
Language
efforts,

in

Thought and Action, confirmed the failure of such

which were correctly treated by Marxists as enemy

Num-

ber One. But there remained other questions, more moderate in


their claims

but very important in practice: the conditions of

success in discussion; blockages

and

facilities in

human communi-

cation; etc. These issues have given rise to a rich literature, based

on the
iour,

and connected with the


thinking and on human behav-

principles of general semantics

influence of language

on

correct

both social and individual.

special field

is

constituted

by problems of education and psychiatry, where Korzybski's


original interest in the pathology of language signs comes to
the fore.

This somewhat practical problem

is

connected with the more

theoretical issues of general semantics,

namely the role of sign

and symbol, the

and thought, and the

relation between language

various questions which

should

call

applied semantics, con-

cerned with vagueness and ambiguity of expressions.


Finally,

general semantics deals with the

influence of lan-

guage on the shaping of culture in the sense of cultural anthropology.


This covers comparative studies in culture as related to the devel-

opment of language, for instance B. Malinowski's The Problem


of Meaning in Primitive Languages and B. L. Whorf's The Relation

of Habitual Thought and Behavior

Language

to

(a study

of the Hopi language).

The
not

aspirations of general semantics go

mean

much

further.

do

here such extravagant formulations as "semantics and

dental surgery", but

more moderate claims

to influence

upon

General Semantics

literature

These

and

are, to

art, to

say the

107

genetic connections with cybernetics, etc.

least, controversial issues.

This

is

why

have

stopped at these problems which are in fact investigated by the


school of general semantics, include real research issues, and
have a literature of their own^o. There are many of such problems,

and

their

importance

is

not secondary. Be that as

it

may,

it

must

be stressed once more that these issues should not be dismissed

Max

hghtly. I agree with

Black,

who

in his

work on Korzybski's

semantics concludes that the theoretical foundations of general


semantics are logically wrong, but adds: "...Thehistory of science

number of examples of how confused

provides a

tems evoked
further that

Do
That

is

and

useful
it

was so when they raised

it

a separate question.

and

technicians,

20 I offer

Its

Many

In

adherents certainly do not admit


fact,

they are above

their theoretical research

here some information which

may

is

social

all

much more remote

be of help to those

who

continue the study of the subject.

like to

value, can

works, differing in the subject matter and varying in

scientific

be fovmd in the yearly volumes of the periodical ETC, which has

been appearing since August 1943. As mentioned above, selected

ETC

be added

real scientific problems.

the problems of general semantics belong to philosophy?

that they are philosophers.

would

theoretical sys-

interesting effects"2i. Let

for 1943-53 have been published in

book form

articles

as Language,

from

Meaning

and Maturity.

series

of monographs has appeared under the auspices of the Inter-

national Non-Aristotelian Library. I should like to mention the following


items, whether included in that series, or not: S. Chase, The Tyranny of Words
and The Power of Words; R. Weil, The Art of Practical Thinking, New York
1940; H. Walpole, Semantics, New York 1941; S. I. Hayakawa, Language
in

Thought and Action,

Affair?,,

New York

W. Johnson, People
Man. A Study

in

in

21

Black, op.

I. J.

Semantic Orientation,

cit.,

Lee, Language Habits in

Talk With People,

Human

New York

1952;

Quandaries; A. Rapoport, Science and the Goals of

New York 1953;


New York 1950.

Philosophy,
Ability,

New York 1949;


and How to

1941,

K.

p. 246.

S.

Keyes

New York
How to

Jr.,

1950,

and Operational

Develop Your Thinking

Research Problems of Semantics

108

from philosophy sensu stricto than the semantic problems connected with logic and epistemology. Nevertheless, general semantics

has certain philosophical implications. There are certainly


loudly bandied

implications,

idealistic

abom

in

Marxist

criti-

cism. But there are also other implications passed over in silence
criticism, since they spoil the schematic pattern

by that

the picture of "pure" ideahsm.

The passing over

since everything that distorts the truth

more so

harmful. The

is

and blur

in silence

is,

harmful, useless and

since that objection can be

raised

not

only against former publications, such as Bykhovsky's article

on Cornforth's book, but also against more recent publications22.


Such criticism is harmful also because, contrary to the intentions of the authors, it undermines the prestige of Marxism. The
notorious article by B. Bykhovsky on semantics (pubhshed in the
Bolshevik in August 1947) was reprinted without any

ETC

in the

Autumn

(in the

comment

of 1948), and the item "Semantic

Philosophy" from the Concise Philosophical Dictionary

(in

ed without

comment

as an

annex to Language, Meaning and

Maturity. In this way, the critical discussions of semantics


to share the fate of

Dictionary

sophical

Rus-

Rozental and Yudin, Moscow, 1951) was also reprint-

sian, ed.

many
which

came

other items from the Concise Philo-

were

reprinted

without

comment

as a separate publication and circulated by the anti-communist

periodical

Preuves.

Thus our

ideological

opponents consider

statements which were intended as "annihilating" criticism of

bourgeois ideahstic ideology, to be the best counter-propaganda,

Marxism and Communism. Can

there

be a sharper condemnation of such forms of ideological

criti-

the best

weapon

against

cism?

22 I

mean

here the

work of

T. A.

EpyTHH "HAeajiHCTHMecKaa cymHocit

ccMaHTHHecKofi 4>hjioco4)hh" [The Idealist Essence of Semantic Philosophy],


in the collection

jective

CoepeMenHbiu cy6heKmueHbm udeoAusM [Contemporary Sub-

Idealism!,

MocKsa

1957,

General Semantics

This concludes the

first

part of

vides information about semantics

my

109
book.

and the

Jt

not only pro-

subject matter of

its

investigations in the various fields, but has also enabled us to

remove a number of misunderstandings and erroneous appraisals

accumulated in Marxist

insufficient

knowledge of the

literature,

mostly as a result of an

subject. It has further given a review

of a wide range of problems, often quite untouched by Marxist


analysis

and

criticism.

Certain issues belonging to semantics,

mainly those related to the theory of communication, signs

and meaning,

will

be taken up in the subsequent chapters.

Part

Two

SELECTED PROBLEMS OF SEMANTICS

Chapter One

THE PHILOSOPHICAL ASPECT


OF THE COMMUNICATION PROCESS

"And the whole earth was of one language, and


And it came to pass, as they journeyed from the
found a plain

And

in the land

of Shinar; and they dwelt there.

Go

they said one to another.

bum them

thoroughly.

of one speech.
east, that they

And

make

us

to, let

brick,

and

they had brick for stone, and slime had

they for mortar.

top

And they said, Go


may reach unto

we be

to, let

scattered abroad

And

the

men

let

upon the

Lord came down

the children of

whose

us build us a city and a tower,

heaven; and

us

make

a name,

us

lest

face of the whole earth.

and the tower, which

to see the city

builded.

And

the Lord said, Behold, the people is one, and they have
one language; and this they begin to do: and now nothing will
be restrained from them, which they have imagined to do.
Go to, let us go down, and there confound their language,
that they may not understand one another's speech.
So the Lord scattered them abroad from thence upon the face

all

of

all

the earth: and they

left

off to build the city".

{Genesis,

"Error

is

never so

difficult

to be destroyed as

when

XI,
it

1-8)

has

its

roots in Language".

(Bentham)

Even

in their pre-scientific reflections

on

power of language. Lucian of Samosata relates


represented Hercules, the symbol of strength,
drawing

men

golden chains.

after him,

people came

reality,

to realize, to a certain extent, the social function

and the

social

that the Gauls


as a patriarch,

their ears fastened to his

tongue with

Those people, says Lucian, follow

their sub-

jugator willingly and joyfully, although they might easily free

From

it

seems that to

the Gauls that physical strength was nothing as

compared with

themselves.

that extraordinary picture

[113]

114

Selected Problems of Semantics

power of the word. And the chains which tied people to Hertongue were but the words which flowed from his lips

the

cules'

to their mindsi.

These problems have many a time been taken up

in hterature,

principally philosophical, both ancient (chiefly Greek)

em

But for a profound, truly

and mod-

and
same time many-sided and broad analysis of the problem
of the social role of language, and in particular the problem of
the philosophical significance of language, we had to wait until
the most recent times. In the 20th century, the problem of language has become the dominant philosophical issue.
Hence an embarrassing situation for one who has resolved
English).

(chiefly

scientific,

at the

to give a

Marxist analysis of the problem. Embarrassing for var-

ious reasons.
First of aU, the wealth of literature. In the latest period alone,

an ocean of ink has been poured out in discussions on the subject.


Next, the immensity of the problem itself. The issue of the
social role

and the

social significance of language (the question

of the role of language in science and philosophy

is

part of the wider problem)

absolutely im-

possible to exhaust

be

selective

Finally,

but

there

so broad that

select

those matters, the frequent analysis


clearly

idealistic

is

but a small

what?

an additional

difiiculty

fact that, while Marxist literature has for

from

it

book. Thus, the necessity to

in a single

it

to
is

is

positions.

up the discussion of these issues

years neglected

made of them has

from the

arising

many

who

Marxist author

and automatically

cussion involving the solutions suggested thus far

often been

takes

also a disis

in

some

The choice of problems, therefore, must be extremely cautious and somewhat restricted, especially at first.
sense a pioneer.

Such being the situation


issues,

think that choice should cover focal

the solution of which

is

the starting point for further.

Described by Stefan Czamowski

in

his

work "Herkules

[Gallic Hercules], included in Vol. Ill of his Dziela [Collected

szawa 1956; reference

is

made

there to Lucian's Heracles.

Galijski"

Works], War-

The Philosophical Aspect of the Communication Process

to begin

and

considerations

detailed

by

analysis.

also seems

It

indicating the problems selected

programme of research rather than by

tive solution, since a full solution

developed research:

and by outlining

of such problems again presupis

the dialectics of

all

necessary to raise, and to solve

is

it

advisable

trying to offer a defini-

poses specialized research. Such precisely


fully

115

in outhne, the general issues in order to

be able to work

fruit-

on detailed problems without the risk of being lost, as to


theory and method, in a jungle of detail; but at the same time,
the understanding of detail is an indispensable condition to
a proper solution of the general problems which otherwise are
fully

overshadowed by

What

generalities

and verbalism.

What

problems, then, are to be classified as focal?

is

to be the starting point of an attempt at a Marxist analysis of

them?

The

great merit of logical semantics consisted in

its

revealing

the philosophical significance of the problems of language.

A num-

ber of mistakes were made: the problem of language, which


is

one of the focal problems of contemporary philosophy, was

made

into the principal,

and even the only

problem was reduced to


tax; analysis

issue ; moreover, that

formal aspects, chiefly those of syn-

its

was confined to

artificial

of the theory of deduction. All

this,

languages, to suit the needs

of course, refers to the

sit-

uation in the 1930s.


Certainly, every discipline

of

its

entitled to

is

analyse the object

research from a particular point of view. This

essary, in

is

even nec-

view of the advancing specialization in science, and

no unfavourable consequences, provided that it means


field of vision. The error of logical
semantics consisted in making a virtue of necessity and, more-

entails

a conscious restriction of the

over, in presenting

its

limited point of view as the absolute. Log-

gical semanticists denied nearly all sociological


ical

issues

of language, precisely

are of decisive importance. This

such a limited approach

is

is

those

and gnoseolog-

which for philosophy

connected with the fact that

quite valueless in the analysis of nat-

Selected Problems of Semantics

11 fi

ural languages. This

is

by no means incidental. The limitations

of logical semantics were a result of the philosophically limited


point of view of

its

founders, both adherents of analytical phi-

losophy, and those of logical empiricism.


those limitations were

got into

made even

Attempts to remove

before 1939, although they

swing only after 1945. The development has oc-

full

curred of the general theory of signs,

i.e.

semiotics, which, although

born of the marriage of neo-positivism with pragmatism, goes

beyond the narrow formal treatment of language by the neopositivists. The tendencies of general semantics to analyse the
social influence and the social power of language go still further.

At the same

time, a critical situation affecting the foundations

of analytical philosophy has been revealed^ and certain formal


analyses which are of

no use

have been subjected to

method of such, have

who

attack

its

its

all

been

Logical semantics, and the

criticized severely

linguists,

of linguistic phenomena.

the achievements and successes of logical semantics,

restricted character is obvious. Obvious, too,

new

by the

purely formal and consequently ahistorical and

asocial analysis

For

also

in the study of the natural language

criticism^.

is

the need of

narrow formalism must be abandoned; the


aspect of the problem must be taken into account ideal-

researches:

social

metaphysics must be

istic

logical

atomism and

rejected

was connected with


and included elements

(this

logical empiricism,

of Platonism on the one hand and those of epistemological solips-

ism on the other); and such philosophical positions must be

adopted as would make possible a profound analysis of the


tion of language to thought

and

rela-

reality (Tarski's semantics,

dopted by the neo-positivists, does not solve the problem,

if

a-

only

because of the philosophical "neutrality" of his conception, declared by Tarski himself,


as to

what

Cf.

'

Cf.

J.

O.

Max

i.e.,

neutrality in the face of the issue

the nature of referents of signs).

is

Urmson, Philosophical

Analysis,

Oxford

1956.

Black's critical essays, extremely important in this respect,

included in his Language and Philosophy.

The Philosophical Aspect of the Communication Process

It is

by no means accidental that the

of the future, that criticism


ticists

in strength. This

is

a matter

is still

growing even among the seman-

is

and that new tendencies are

themselves,

have been

difficulties

for a fairly long time, that their being resolved

felt

117

because at the root of the

gaining

clearly

crisis lie

erroneous

philosophical assumptions typical of the traditional semanticist


trend. It

is

precisely this

theoretical role of

which

Marxist

offers great opportunities to the

philosophy.

By

organically

com-

bining gnoseological and sociological problems, materiahsm in


theoretical

diflficulties

with

interpretations

Marxist philosophy

is

as

it

historicism

in

methodology,

were predestined to removing those

and symptoms of

crisis

which are now observed in

semantics (in the broad sense of the term).


If

we now ask

What

is

the focal theoretical problem which

from that point of view should be investigated


must

be,

beyond

all

doubt

the theory

first ?,

the answer

of communication^.

Semantics certainly includes certain detailed questions which


are in

no way connected with philosophical

issues

certain specific problems of logical syntax.

Far be

for instance
it

from me

when
we leave the purely technical ground, we immediately face the paramount question, a question on the solution of which depends
this or that solution of a number of problems that are usually
to deny the scientific importance of such problems, but

termed semantic. That question

is

what is the

process of

human

most important part of which


is human communication by means of a phonic language) and
what are the conditions of that process?
communication

(the process the

Let us not shirk the issue, but begin by analysing that prob-

lem

first.

Reference

communication,

and

is

made

here to the philosophical aspect of the theory of

which also has other aspects:

technical, the last-named being

now

psychological,

extensively developed.

linguistic,

118

Selected Problems of Semantics

THE ESSENCE OF THE PROBLEM OF COMMUNICATION

1.

In the 1920s, Ogden and Richards, authors

who have

ren-

dered great service to the development of semantics, thus outhned


the

programme of

that disciphne in the study of the theory of

communication:
"In yet another respect

all

these specialists fail to realize

the deficiencies of current Mnguistic theory. Preoccupied as they


are

ethnologist

with recording the details of fast vanishing

languages; philologist with an elaborate technique of phonetic

and

laws

losophy'

of derivation;

principles

all

philosophers

with

'phi-

have overlooked the pressing need for a better

understanding of what actually occurs in discussion. The analysis

of the process of communication

psychology has

now

is

partly psychological,

reached a stage at which this part

and

may

be

succesfuUy undertaken. Until this had happened the science of

Symbolism necessarily remained in abeyance, but there is no


longer any excuse for vague talk about Meaning, and ignorance
of the ways in which words deceive us" 5.
In what however, does the problem of interhuman communication consist?
It

often happens that a philosophical problem begins at the

point where

common

common

sense stops short. Just at that point where

sense seems satisfied, stating plainly that people speak to

one another and in

this

way convey mutually

all

kinds of informa-

and in this sense communicate with one another, the philosopher starts asking questions: but how do they do that? and
why? and what does it all mean? He often questions in a way
involving risk to his reputation, a way which leads laymen to
tion,

impart to the term "philosopher"

must

a derogatory shade.

also be conceded that the poet

may be

So

it

right in ridiculing

philosophical speculation and raising his voice in defence of

common
5

sense.

The honesty of a

C. K. Ogden

1953, p.

8.

&

I.

professional philosopher im-

A. Richards, The Meaning of Meaning, London

The Philosophical Aspect of the Communication Process

pels

me

and profound, though

to quote the witty

poem by

Erich

Weinert, entitled Der

little

119

known,

Philosophenkongress:

Die Philosopher! veranstalteten einen Kongress


Auf Beschluss des Philosophenbundes
Zwecks Eindringung in das Innre des

Noch

Da

unerschlossenen Daseinsgrundes.

trafen sich die delegierten

Mit Dietrichen,

An

Schliisseln

dem

der Tiir mit

Und

bestellten

Aber

die

Weisen

und anderen Eisen

grossen Fragezeichen

da unverdrossen.

Tur wollte nicht weichen;

Sie blieb verschlossen.

Und

schliesslich

kamen

sie iiberein,

Es miisse eine ganz komplizierte Mechanik

sein.

Soeben begannen
tJber: als ob,

Da

trat ein

Aus

sie einen tiefen Disput


an sich und dergleichen,

Mann ohne Doktorhut


dem Fragezeichen.

der TUr mit

Die Philosophen riimpfen ihre Russel:


Woher haben dann Sie den richtigen Schliissel?

Der Bestaunte aber sagte gelassen:


Man braucht nur an die Klinke zu

fassen:

Ich dachte beUeibe noch nie daran,

Dass man die Tiir auch verschliessen kann.


Die Philosophen lachelten ob dieses Manns:
Ja, ja, die heilige

Wollte

man

Ignoranz:

ein so wichtiges

Problem

In so einfaltiger Weise losen,

man doch
Und auserdem

Brauchte

Ware

Do

ja

kein wissenschaftliches System,

auch der Kongress dann nicht notig gewesen.

such things happen? Yes, of course they do

That

will

be

demonstrated before long by way of an example which interests


us.

120

Selected Problems of Semantics

But can such common-sense argumentation

our

suffice for

was Engels who said that what is


termed common sense fares well enough in every-day matters,
but strange things begin to happen to it when it seeks to become
analysis? Certainly not. It

involved in the wider arena of

scientific,

philosophical,

etc.,

investigations.

Human

communication embraces a

sophical problem, strange

'as

far

may

that

from neghgible

seen from the

philo-

common-

sense point of view.


Is there

we say

anything simpler in every-day

to a person

we

are working with,

life

"hand

than the fact when

me the

axe, please"

or "steady that beam, please", that person, on hearing what


say,

behaves as we have asked him

to.

It is

all

we

so simple and

self-evident that should the behaviour expected in reply to our

words not occur, we should see

it

as something extraordinary

and seek the causes of such a state of things. Just the fact what
we are stimulated to reflect on the communication process only
when it becomes disturbed shows how natural that process
seems to us. Nevertheless here lies a philosophical and sociological problem of enormous significance: how does it happen
that people do communicate with one another? It is only against
the background of that problem that we see in a proper fight
and in a proper context the traditional, almost classical issues
of semantics, such as the questions of sign, of symbol, of meaning,
of language and speech, and the related philosophical
gnoseological) issues.

Of

course,

it

is

(strictly:

possible to engage in the

study of logical antinomies, of the theory of types, and of logical


calculus, without paying heed to their wider theoretical aspects.

But only to a certain point. So


to deal only with the

it

is

not a practical possibility

narrow technical aspects of these problems

without crossing the frontier beyond which the philosophical


issue

looms

large.

Neither a logician nor a linguist can avoid

that eventuality, if he sets out to study the theoretical aspect

of language as the subject matter of his investigations.


so sooner or later

and thereby become,

He has

to

do

willy-nilly, a philosopher.-

The Philosophical Aspect of the Communication Process

To understand

that,

take a look at the contro-

suffices to

it

121

versy concerning the promises and conditions of the

communi-

cation process, a controversy between the tv/o principal conceptions in that respect: transcendental

and

guments in

full relief, I shall

Urban

follow

That con-

naturalist.

show the

troversy will be discussed below and, in order to


^^

ar-

in choosing as the

principal opponents Jaspers ^ as the representative of the tran-

scendental conception, and

certain general

Dewey

The problem of communication


is

one of the

certainly

is

it

one of the funda-

to point out that

commu-

essential properties belonging to the de-

of knowledge, of

finition

arguments

mental problems of philosophy. Suffice


nicability

their

remarks are necessary.

I
Bi

as the representative of the

But as background to

conception.

naturalist

scientific cognition: this is so

because

would be impossible without communicability. Moreover, communication seems to be an element


inseparable from all processes connected with cognition: even
speechless thinking in monologue is a specific act of communication (in the extended sense of the term). Consequently, it is no
intersubjective

verifiability

matter for wonder that philosophy cannot disregard that problem, if epistemology

is

to be pursued in any reasonable manner.

Further, the statement that communication


essential
ial.

phenomena of

social hfe

The controversy
Urban

triv-

possibility

of

very lucidly presented by Wilbur Marshall


"Intelligible

Urban

Communi-

declares himself in fa:vour of the transcendental conception,

which he evidently follows Kant.


"^

Chap.
ist

is

life

one of the

would be impossible; this refers

Language and Reality, London 1951, Chap. VI:

cation".
in

is

not only obvious, but

Without human communication, without a

such comrnunication, social

in

is

Cf.
Ill

K. Jaspers, Philosophie, Vol. 2: Existenzerhellimg, Berlin 1932.


"Kommunikation", is devoted to the explanation of the existential-

theory of communication.
8

Cf.

J.

Dewey, Experience and Nature, London 1929, Chap. V.

122

Selected Problems of Semantics

which

in particular to the process of labour,


social
is

It is self-evident

life.

that the

also

one of the focal issues of sociology.

No

wonder, then, that philosophers are

ular attention to
forts

the

must be made

lies at

the root of

problem of communication

now paying particmuch more: ef-

question. I should say

to extract the kernel of the

problem from

of this or that philosophical speculation.

the mystifying shell

Moreover, one should not even allow oneself to be discouraged


a priori in the case of the controversy around "true" communication,

controversy between the

two conceptions mentioned

above.
First of all, however, the concept of

communication must

be explained in greater detail and the object of controversy must


thus be

made more precise.


Urban in dividing

I follow

principal

categories:

acts

acts of

communication into two

communicating a certain behaviour

or emotional state (behavioural communication), and acts com-

municating a certain knowledge or mental state

(intelligible

communication).

Various attempts have been made to give a definition of man,

Marx

to bring out his differentia specifica.


that

man

is

said after

FrankUn

an animal which produces tools; contemporary

semanticists see that differentia in the ability to use signs

symbols. In

fact, in

pects of the

both cases we have to do with different

same process of human

social

labour and the process of using signs,

i.e.,

and
as-

The process of
human communica-

life.

tion are

interconnected genetically and functionally. If that

connection

is

understood, one can well introduce communica-

tion as an element in the definition of

man and human

society.

But what communication?


In a certain sense of the term, communication can be observ-

ed in the animal world as well. The bee, with


flicks

of the antennae, induces

its

hive-mates to

its

"dance" and

fly to rich finds,

and thus "communicates" something to them. The same applies to the ants, which with touches of the antennae warn their

The PHiLosopmcAL Aspect of the Communication Process

123

I
"community"

against

threatening

danger.

Birds'

love

songs,

and the rutting of the deer, are also specific forms of communication. "Communication" of a similar kind with others; expressing fear, exultation,

occurs between

etc.,

men

as well; in extra-

ordinary situations, the very utterance of some sound or even


the very expression of the face or gestures of the

body "com-

municate" to us something concerning the experiences of those


impelled to such actions. But
as

transfers

it

ual

is this

communication? In so

knowledge of emotional

from one

states

far

individ-

to another, or provides information about a certain

sit-

But that is a specific communication,


from the very behaviour of the individuals concerned,
communication working by emotional "contagion". The latter
formulation probably corresponds best to the state which fol-

uation hie et nunc, yes.


resulting

lows those specific

acts

meadows

that

leaving for

of "communication"

abound

of the

bees

in flowers, of a herd in stam-

pede, etc. That communication differs essentially from the typically

human communication which

certain mental states. It

with which

But

we

is it?

is

that

transfers certain

human form

knowledge and

of communication

are concerned in our analysis.

Are we

in fact concerned

above

all

with that form

of communication, with communication in that sense of the term?

Men communicate to
al-

states,

one another their experiences, emotionknowledge and mental states in various ways, and

by various means. Hence the controversy as to what form of


communication is the "true", i.e., the proper form of communication.

As mentioned above, communication working by "contagion",


i.e., transmitting above all the emotional state from one individual to another,

is

also to be observed in

extent the basis of

force in

mob

moments of

men.

It

forms to a large

psychology, and becomes a tremendous

panic, outbreaks of hatred, loosening of the

bonds that hold instinctive actions in check, etc.


But that does not exhaust the issue. Are not music, to a

social

tain extent the visual arts,

and

cer-

to a large extent poetry, partic-

124

Selected Problems of Semantics

of that

ular manifestations

form of communication?

specific

Those connoisseurs of music do well who warn against

grammized"

perception,

that

against

is

its

"pro-

music

"translating"

into a "language" of thinking in terms of notions or images.

They maintain

that music should be perceived as a stream of

emotional states of a specific kind,


ly,

at

all,

mits,

agree with that; consequent-

agree with the assertion that

then

it

reflects

if

only emotional

communicates something to

music

states,

"reflects"

and that

others, then

it

anything

if it

trans-

communicates

just such states only. I mean, of course, good, great music. But

do not agree with the assertion that that is the "true" comcommunication par excellence, although I admit
that it is a different, special form of communication.
Is not, e.g., the escape sought by the visual arts in abstractionism connected with the existence of that form of communication? The abstractionist thesis is that the intellectual content
of visual experience should be rejected and only the "true" transmission of certain emotional states should be left. The promoters of "dadaism" and other similar trends in poetry were also
I

munication,

concerned, they claimed, with a "true

some

direct

states

and experiences.

transmission

Let us sidetrack the appraisal of

common

communication",

of their

others

to

all

those

own

with

emotional

artistic

trends,

them from the point of


view in which we are interested in all those cases we have to do
with human communication, but a communication of a special
type, namely of an emotional, and not an intellectual, nature.
And yet we imply intellectual communication when we speak
of human communication tout court.
Why? Because it is most common in the social lives of men
and plays in their life a special role, is a necessary condition of
all social bonds, and especially of the bonds resulting from producand

try to find out

what

is

to

tive

all

work.

What that
known to

type of communication really consists of


us perfectly well from every-day

life

after

can best

The Philosophical Aspect of the Communication Process

be seen

if

we compare

it

125

with communication by means, for

instance, of music.

Now
presses

the composer experiences an ecstasy of love

it

and

ex-

in the language of music in the form of a Nocturne,

or he experiences patriotic elan due to a national uprising in


his

country and gives expression to his feehngs in the form of

the Revolutionary Etude,

or else he transmits emotionally the

form of a Rain Prelude. After


hstens to those works without knowing

dreariness of a rainy day in the

many

someone

years,

else

knowing
and without any programmed deciphering of their

the accompanying circumstances of their birth, without


their titles,

meaning

in intellectual terms. Nonetheless he does experience the

longing of the Nocturne, the excitement of the Revolutionary


Etude,

and the dreariness of the Rain

and

ever,

this is

no

Prelude',

trifling proviso, that

provided, how-

he belongs to a definite

cultural tradition, in particular as regards music.

who

For a Hindu

has not established contacts with European culture, Chopin's

Hindu music for a European. One point more: since we have to do with an emotional
"contagion" by non-intellectual means here, no one can know
whether he experiences the same feeling as did the composer,
or as do other people listening to the same composition. The
music

is

as non-communicative as

fact that

even the same person

reaction to the

is

old

differs at various periods in his

same piece of music, depending on

his

own emo-

tional "context", forces us to the conclusion that different people

perhaps experience different feelings in reaction to specific pieces


of music.

Some people

"language" of music
not

decide that

parentheses,
arises

is

point

that there

say that

it

should be so, and that the

the best just because


here.
is

some

should

it is flexible.

I shall

hke only to add, in

great misunderstanding which

out of transfering par force, in spite of

all

declarations,

communication to the field of music.


All in all we can say that in the case of "communicating" one's
experiences to others by means of music there is an emotional
the categories of intellectual

"contagion" those
:

who

transmit the communique, and those

who

126

Selected Problems of Semantics

receive

it,

experience definite emotional states;

tlie

similarity

as between these states cannot be ascertained; the perceptive

communicated

largely depends

reaction to v^^hat

is

tional context in

which the comunique

is

who receives it.


What about communication combined
standing, that

on the emo-

placed by the person

with intellectual under-

the case of the communication of certain intel-

is,

lectual states?

No

detailed

analysis

be undertaken here.
of such of

its

of that type of communication

I shall

will

confine myself to taking note only

general traits as

make

it

possible to distinguish

it

from the communication of emotional states.


Suppose Peter is felling trees together with John, and at
a certain moment he says: "Hand me the axe, please". John
hears what Peter says and, comprehending the content, passes

him

the axe.

necessary condition for this act of comprehen-

John should know the language used by Peter, since


otherwise he would be unable to understand him (just as God
correctly concluded in the case of the Tower of Babel). Should
sion

is

that

John, without understanding the language used by Peter, guess,

from the

and Peter's gestures, what it is that


him the axe, that would not alter the esJohn, guided by the situational context, would

situational context

Peter wants, and pass

sence of the case

himself what Peter obviously wanted to communicate to

tell

him. The understanding of Peter by John would be replaced by

But without understanding


no communication.
In what, then, does the act of communication consist?
A person makes a statement, and another person, who hears that

John's guessing Peter's intentions.


the partner there

is

statement, understands

it, i.e.,

{not the same, since that

experiences mental states analogous

depends on the individual context, which

changes) to those of the author of the statement.

Without entering into


cess,

one can

And

that

is all.

details of that very complicated pro-

at once grasp the difference

of communication discussed.

between the two types

The Philosophical Aspect of the Commltnication Process

127

In communication by means of music, images, and to a cer-

by means of poetry as

tain extent

well, the point is to transmit

emotions to others; in the other case, to transmit intellectual


content, to transmit certain mental states. This does not imply
that such a

communication has no emotional

contrary,

purpose

its

tional state:

love,

may

hatred,

the emotional aspect

is

aspect.

On

the

be precisely to evoke a certain emoreadiness for self-sacrifice, etc. But

here always secondary in relation to

and is transmitted through the intermeHence we should agree that the "language"
of music or images is more suitable for a direct transmission
of emotions than is the language of words. As the poet says,
content,

intellectual

diary of the latter.

'^Spricht die Seele, spricht die Seele nicht

mehr".

The intellectual tasks of communication can be performed


only by a language of words, a phonic language (or the written
form of such). All other forms of transmitting

intellectual con-

tent in a civilized society (I exclude here the at least controversial


issue of the

of

human

language of gestures as the hypothetical earliest form

language) must, in the last analysis, be translated into

the language of words ; this apphes both to the language of gestures

and the language of mathematics, or such conventional

guages" as the "language" of flowers, smells,

"lan-

This does not

etc.

apply to the case of a spontaneous language of gestures of deaf-

As mentioned

mutes.

in Part I above, the essential error of all

above

speculations concerning artificial languages consists

in forgetting that those languages are, in the last analysis,

on a natural language, a language of words, and are


able in

all

based

interpret-

it.

In contrast with emotional communication, intellectual com-

being com-

munication presupposes the understanding of what

is

municated. In that type of communication, there

no communi-

is

cation without an understanding of definite intellectual content,

which

is

case

of

not only unnecessary, but even contra-indicated, in the

emotional

"programmatic"

communication
interpretation

(cf.

the

of musical

prohibition

of

perception,

as

128

Selected Problems of Semantics

mentioned above). That

communication

to say

is

in contrast with emotional

communication

intellectual

is

conditioned by

the experience of the communicating parties of analogous mental


states,

and the

ideal sought in this field

munication possible of one's


precision

the most precise

is

own mental

states

com-

(we imply a

and adequacy of communication), with a

full

shifting of

the margin of individuahzation in experiencing analogous mental


states into the sphere of the context
intellectual,

emotional,

provided by the various

associations.

etc.,

The term "context" has been

now be done

with more precision. This will


the problem

with

itself

later. It will

it

provisionally, because

belongs to the issue of meaning, to be dealt


only be outlined here to give a general indi-

mean when

cation of what I

above in
must be defined

several times used

a metaphorical and therefore vague sense; so

say that something

is

understood

"in a context".

The matter is quite clear when a hterary metaphor is resorted


If someone says of someone else: "He is a butterfly flitting from flower to flower", then to understand that statement
requires a change in the univers du dlscours from natural to social

to.

science, since otherwise a

Similar,

though

ings of certain

complete misunderstanding will

less self-evident, is the case

terms of poetry,

of changes in the mean-

according to whether

statements

or science,

or

some

arise.

we speak

specialized

in

disciphnes;

an expression changes according


it is placed. This is due to the
fact that linguistic expressions are extremely ambiguous and admit
of different interpretations. I do not agree with Urban when he
in other words, the sense of

which

to the univers du discours in

says that the communication of emotional states through be-

haviour (so called behavioural communication) does not require


the consideration of any context, at least in the simplest cases;

but

fully agree

with him on the point that intellectual com-

munication presupposes a reference not just to an object, but


to

an object located in a

definite univers

du discours, in other

words, that the content of a communique can be understood

The Philosophical Aspect of the Commumcation Process

129

only in a definite context. "All linguistic meaning

is

referential,

but

is

interpreted

is

it

also systemic" 9. If the univers du discours

not only as a changing system of knowledge to which a given


statement

is

of accumulated experience,

way

general

an individually changing system

referred, but also as

we

achieve what I have in a most

the intellectual

called

and emotional context of

a changing understanding of the same expression by different

Let us be

individual recipients.

satisfied,

for the time being,

with that prehminary explanation.

Thus we come to the following conclusions concerning a more


precise formulation of the

The

to the entire

and to

term "communication":

human manner

specifically

domain of man's

intellectual experience.

of communication

spiritual fife:

refers

both to emotional

Although these two spheres can-

not be separated in an absolute way,

they represent different

and thus they are connected with different


forms of communication (which themselves cannot be separated
in any rigid and absolute way).
Communication in the emotional sphere often takes place

fields

of spiritual

fife

through the intermediary of extra-hnguistic means, and one can


accept the thesis that as far as the transmission of certain emo-

moods (emotional

tional

"contagion")

is

concerned,

the

ex-

means of music and of the visual arts certainly do achieve


something (although it must be borne in mind that the realiza-

pressive

tion of one's emotional state, achieved as a result of extra-linguistic

communication, requires

Intellectual

communication,

linguistic
i.e.,

means).

communication intended

transmit to others certain mental states,

is

to

a linguistic communi-

cation par excellence (since the systems of signs always represent

here

some fragments of a phonic


is

which presupposes not only a


object, but also a

common

cours.
^

language), and the focal problem

an analogous understanding by the communicating

Urban, op.

cit.,

p,

232.

common

parties,

reference to the

reference to the

same

same

univers du dis-

130

Selected Problems of Semantics

The problem of adequate understanding

as

between the

persons participating in the process of communication

is

con-

nected with the controversy between the transcendentahst and

we

the naturahst conception, a controversy

submit

2.

to

are

now

going to

analysis.

critical

THE CONTROVERSY BETWEEN THE TRANSCENDENTAHST


AND THE NATURALIST CONCEPTION
The

transcendentahst

process certainly

of view of
aUty

the

communication

sense and of a scientific reflection on re-

be understood without due consideration of

philosophical background.

empirical

it

its

reaches back to the

cognition placed direct

an insight of the soul into the essence of things,

is

into their ideas,

Historically,

who above

doctrine of Plato
cognition, that

of

one of the strangest concepts from the point

common

cannot

conception

an

insight that

can be neither rendered in words

nor communicated. The continuation of that view can be seen


has also been revived in most

re-

cent times in the various versions of irrationalism, above

all

in neo-Platonic mysticism;

it

and

in Bergson's intuitionismio

in Husserl's

phenomenology ii.

The conception that belittles or even plainly denies the role


of verbal communication has its roots in those metaphysical
ideas.

^'Spricht die

Seele,

spricht die Seele nicht mehr."

And

"Does anyone believe that by means of lancommunication can pass from man to man?"i2

Maeterhnck says:
guage any real

Yet the

direct philosophical roots of the transcendentahst con-

ception of communication must be sought elsewhere: in Kanti-

anism, or rather in neo-Kantianism. The splitting of the world


10

See his Lapensee et

le

mouvant. Chap. VI: "Introduction k la m^taphy-

sique".
11

Cf. Ideen

zu einer

reinen

Phdnomenologie und phdnomenolo gischen

Philosophie.
12

Quoted

after

Urban, op.

cit.,

p. 242,

The Philosophical Aspect of the Communication Process

into

phenomena and noumena

131

(the very terminology indicates a re-

lationship between that conception

and Platonism), and at the


same time the endowing of human mind with inborn forms of
a priori perception by means of which what is given in cognition

is

the construction of the cognitive

philosophical
in

its

of

all

soil

provided

by

mind

such was the

transcendentalism,

especially

Fichtean version, typical of neo-Kantianism, for the growth


speculations concerning a "trans-subjective

T". That

type

of speculation obviously hes at the root of the so-called transcendentahst theory of communication.

According to that theory, the problem of communication


can be reduced to the following question: supposing that communi-

two individuals who are absolutely


and such, according to transcendentalism, is the position of those individuals not immersed in
some mystic ether of "transcendental 'I'" which unites them
cation takes place between

separated from one another

spiritually (please, take

note of that assumption, since

the essential role in argumentation

them

we

ask,

how

is

it

plays

it

pos-

communicate by transmitting from one individual to another certain mental states, by understanding one
sible for

to

and not only by perceiving certain situacommunication?


The transcendentalist conception explains this in two ways:
(1) true communication is direct (the Platonic motif), and (2) at
its root lies a specific
metaphysical community formed by the
"transcendental 'I'" or by a "universal mind", in which the
another's statements

tions hie et nunc, as in the case of behavioural

some way or another, or of which


they are parts (the Thomistic and Kantian motif).
The concept of "true communication" appears most clearly
individual minds participate in

in Jaspers's
is

formulation of "existential communication". Here

from the chapter several dozen pages long, dethe communication process, in Volume 11 of the author's

a quotation

voted to

Philosophy.

"Wenn durch alles Ausserliche hindurch der Mensch


zum anderen Selbst tritt, die Tauschungen fallen,

er selbst

als

das

132

Selected Problems of Semantics

Eigentliche oflfenbar wird, so konnte es Ziel werden, dass Seele

mit Seele schleierlos ohne

alle

Bindung

in der Ausserlichkeit des

Weltdaseins in Eines schlage.

"Jedoch kann

in

der

Welt Existenz mit Existenz sich nicht

sondern nur durch die Medien der Inhalte trefDas Ineinsschlagen der Seelen bedarf der Wirklichkeit des
Handelns und des Ausdrucks. Denn Kommunikation ist nicht
unmittelbar,

fen.

wirklich als die widerstandslose bestehende Helligkeit eines seeligen Seins

ohne

im
ob

Raum und

Zeit,

sondern die Bewegung des

Wohl

es in Augenkann im Transzendieren iiber alles Weltdasein sich erfiillen. Aber auch dann
ist Weite und Klarheit des objektiv gewordenen und nun transzendierten Inhalts das Mass fur die Entschiedenheit des Augenblicks der eigentlichen Kommunikation. Diese gewinnt ihren
Aufschwung durch Teilnahme an Ideen in der Welt, an Aufgaben

Selbstseins

bhcken,

als

Stoff der Wirklichkeit.

ist

die Beriihrung unmittelbar sei ; sie

und Zwecken"i3.
But it is the concept of "transcendental 'I'", introduced in
some form or another into the theory of communication, which
is

here of decisive significance.

That concept was very

who

clearly formulated

in order to explain the essence of

by Karl Vossler,

communication

intro-

duced the notion of "metaphysical language-community" alongthat

side

of "empirical language-community"i4.

According to Vossler,

it

is

always one person

who

is

the

and author of the conversation in which communication


takes place, although his functions and roles may be distributed
over a number of persons. Conversation is something like a drama
which is played within each of the participants in that conversation. This looks as if the monologue were said to be the origin
carrier

of the dialogue. But Vossler goes

show what metaphysical

1^

Jaspers, op.

14

Vossler, Geist

cit.,

much

further.

His explanations

objectives he has in view.

p. 67.

und Kultur

in

der Sprache, Heidelberg 1925.

The Philosophical Aspect of the Communication Process

"In der metaphysischen Ansicht

stellt

sich die Sache so dar,

dass in der menschlichen Personlichkeit als solcher,


Einheit

von

findenden Gesprache

dem

der Zeit auf

sich

begriffen

abspielen.

All das,

gedacht werden, in

dem

von personenhaften Rollen

Geist, in Milliarden
all

samtliche

ist,

die

hienieden

eine
statt-

was im Laufe

Erdball gesprochen wird, muss sonach als ein

riesiges Selbstgesprach

und aus

Vielheit

beliebiger

133

der menschliche
sich entfaltend

diesen Vereinzelungen sich wieder zusammensuchend,

ist,

"Daraus

folgt

nun

freilich,

dass der menschliche Geist als

oder werden miisste; und es


ob der Begriff der Person diese Steigerung ins Absolute aushalt. Dass der Anspruch auf etwas absolut Geistiges
und Einheitliches in ihm enthalten ist, kann nicht mehr zweifelhaft sein; aber die ausserste Formel mit ihrem Verlagen nach
unendlich vielen Rollen in einer Person ist damit noch lange
solcher eine

einzige Person sein

fragt sich,

nicht verwirklicht"i5.

Be
ogist

that as

it

may, these metaphysical speculations of a

philol-

have been used as an argument in the construction of a tran-

scendentahst theory of communication. This

is

what Urban, one

of the exponents of that theory, says in this connection:


transcendentaUsts

certainly

not

Kant

himself

"Few

would

be

disposed to think of the notion of the transcendental Self as

more than a symbol

for this underlying unity.

symbolized

not a myth.

is

itself

It is

But the unity thus

a necessary condition of

mind however small, without


communication are impossible. Whether
for the unity thus symbolized we use the 'myth' of an overindividual self or of an over-individual community is, from the
that universahty, that mutuality of

which knowledge and

its

present point of view, a matter of relative indifference. I do not

much

care in the present context whether one thinks of an

embracing mind in which


their being, or as

15

Ibid., p.

finite

minds

live

all-

and move and have

an over-individual society of minds. Important

13.

10

134

Selected Problems of Semantics

as these issues are in""other contexts, they are not the significant

What
minimum

minimum

thing here.

is

or the

of 'transcendental considerations', necessary

significant in the transcendental

for the understanding

Such an assumption
communication cannot,

And

of
is

intelligible

communication" ^ 6.

made because

it is

human

the process of

claimed, be explained in any other

why one has to believe in some transcendental,


What does it mean? Does it mean anything
at all? That is inessential. One has to believe. And, strange as
it may seem, there are people who in the name of "philosophy"
way.

this is

supraindividual "I".

no exaggeration: the reader is


the answer offered by Urban to the

openly proclaim that. This


invited to study carefully

is

question about the proof of the existence of something like that


supra-individual

"transcendental,

T":

"The

unity implied in intelligible communication

and, therefore, by

definition,

not

supra-empirical

is

verifiable as

supra-empirical

an empirical

fact

"i"^.

by a direct application of the 'empirical criterion'


The arguments adduced by the transcendentalists
of their conception are such a philosophical

titbit

in favour

that

cannot

deny myself the pleasure of quoting Urban once more: "The

argument of the transcendentalist


but
is

so

if

it

may

be of a peculiar kind,

only because the facts upon which the argument

is

based are of a special order. Mind, conceived as wholly

lated separate selves,

it

finds,

to carry the burden of communication

and human
when conceived in
ciology. It is upon

science

Such evidence
so.

The

analysis,

is,

institutions.

iso-

with Dewey, wholly inadequate

and

all its

works, both

But equally inadequate

is

mind

purely naturalistic terms of history and sothese considerations that the evidence rests.
to be sure, largely negative, but not wholly

force of the transcendental argument rests, in the last

on

certain positive considerations,

namely the actual

character of intelhgible communication as

we have

more

involved in such com-

16

especially of understanding

Urban, op.

17 Ibid., p.

cit.,

259.

255-256.

which

is

seen

it

to be,

The Philosophical Aspect of the Communication Process

135

munication. If the character of these facts

is

if

acknowledged

communication as contrasted with merely apparent com-

real

munication

recognized,

is

then the

'transcendental considera-

tions' are necessitated for the understanding of these facts. This


is

evidence in any proper sense of the word" is.

But

let

us

now

leave in peace that curiosity which consti-

the argumentation of the transcendentahsts, and

tutes

us

let

pay attention to the opposing conception, namely that of the


also called the naturahst conception.

behaviourists,

Their thesis

is

that people can communicate,

i.e.,

can mutu-

understand their statements, because they have analogous

ally

and have to do with reahty


communication
is a very prosaic
Hence
that
common
matter, unnecessarily mystified by the transcendentahsts. Its
essence is: someone experiences something and acts on his milieu so that other minds have similar experiences. There is nothing
physical

and

intellectual structure

to

is

all.

mysterious in that process under definite conditions, the various


:

minds have similar experiences. Such


sition

more or

is

of the naturahsts, as expounded by

I.

less the

po-

A. Richards in his

Principles of Literary Criticism^^.

similar formulation

is

offered

by John Dewey whom,

if

only in view of his position in the history of behaviourism,


I

consider the chief representative of the naturalist conception.

Dewey admits

that the transcendentahsts have paid

attention than others to the role of language in

But

at the

same time

fication of the issue,

it is

they

who

since they

to certain supernatural principles.


naturally, if

we

more

society.

are responsible for the mysti-

explain

Yet

it

by having recourse

can be explained quite

it

life and its needs,


which language plays a particular role as

take as the starting point social

in the satisfaction of

the

human

means of communication. To understand the function of

language and the essence of communication one has to study

human behaviour and human co-operation "The


:

18 Ibid.,
p.
19

London

260 (itaUcs
1955, Chap.

A.

heart of language

S.).

XXI: "A Theory of Communication".

136

is

Selected Problems of Semantics

not 'expression' of something antecedent,

sion of antecedent thought. It

much

less

expres-

communication; the establishment

is

of co-operation in an activity in which there are partners, and


in wliich the activity of each

To

nership.

in action;

is

understand

fail to

misunderstand

to

modified and regulated by partis

is

to fail to

to set

come

up action

into agreement

cross pur-

at

poses"20.

And

here

is

further explanation of the communicative func-

tion of language

"Language

is

mode

specifically

two beings, a speaker and a hearer;

of interaction of at least

it

presupposes an organized

group to which these creatures belong, and from


have acquired their habits of speech.

It is therefore

whom

they

a relationship,

not a particularity. This consideration alone condemns traditional

nominahsm. The meaning of


something

we

common

attribute

signs,

moreover, always includes

and an

as between persons

meaning to the speaker as

granted another person

who

is

object.

his intent,

we

When

take for

to share in the execution of the

and also something independent of the persons concerned,


through which the intent is to be realized. Persons and
intent,

in a commonly shared consecommunity of partaking is meaning.


"The invention and use of tools have played a large part in
consolidating meanings, because a tool is a thing used as means
to consequences, instead of being taken directly and physically.

things

must ahke serve as means

quence. This

It

is

anticipatory,

predictive.

'transcendence',

nothing

relational,

intrinsically

absent, or

reference to the

The most convincing evidence


found
In

in the fact that they

many

formulations,

that animals

have no tools

Dewey comes

do not

Without
is

a tool.

'think'

is

..."2i.

close to the Marxist

viewpoint, but the differences between the two positions also

stand in

relief.

These matters

when my own opinion on


20

Dewey, op.

21 Ibid., p.

cit.,

185.

p.

179.

will

be discussed at a

later stage,

controversial issues will be advanced.

The Philosophical Aspect of the Communication Process

Before passing to the analysis of the

of the transcendentalists,

who

let

critical

who

in his

argumentation

us take note of one other author

stands close to the naturahst conception.

Gardiner,

137

mean

here Alan

book The Theory of Speech and Language

formulates Dewey's theses even more sharply.

Gardiner

by analysing the notions of language and

starts

Dewey who,

speech. But, like

while concentrating

liis

attention

to attack the problem of

on the problem of meaning, had

com-

munication, Gardiner also sheds indirect light upon the issue.


Gardiner, like Dewey, interprets the action of speaking in

terms of co-operation between two persons


the

Ustener but

at the

same

time,

the speaker and

and very important for the

solution of the problem, exphcitly introduces as the third ele-

ment that which

is

spoken about. True, the

naturalists

(e.g.

Richards) explain the possibility of communication by the similarity

of minds and by a reahty

common

introducing the category of things


ist

and

to

all,

but Gardiner, by

the requirement of a real-

formulation of the theory of language and speech, thereby

admits a

new

theoretical element

which brings him close to mate-

rialism.

Gardiner begins with a criticism of the definition of speech


as a system of articulated phonic symbols used to express
thoughts, and goes on:

"As a
between

first

approximation

man and man,

let

us define speech as the use,

of articulate sound-signs for the com-

munication of their wishes and their views about things. Note


that I do not attempt to deny the thought-element is speech, but
the emphasis of

my

definition does not

points which I wish to stress are,

firstly,

lie

of speech, and, secondly, the fact that


with things, that

is

on that element. The

the co-operative character


it

is

always concerned

to say with the realities both of the external

world and of man's inner experience"^^.


Thus, a proper formulation of the notions of language, speech,

meaning
22

and

understanding

(Gardiner

draws

an

explicit

Gardiner, The Theory of Speech and Language, Oxford 1951, p. 18.

Selected Problems of Semantics

138

1
(

distinction between the meaning of words and the things denoted


by those words), and consequently a proper formulation of the
process of communication requires, following Gardiner, that four

elements of the situation be taken into account: (1) the speaker,


(2) the listener, (3) the

What

is

thing spoken about, (4) the words spoken.

then asserted by the naturalists concerning the essence

of the communication process?

They maintain that what is at stake is simple and prosaic,


and as such does not require recourse to any supernatural factors. What is involved is simply a mutual transmission of the
content

of certain personal experiences by linguistic means,

which, they hold, can always be reduced to the category of influence.

Such a transmission of the content of experience by

the speaker to the listener

is

possible, because (1) the

communi-

cating organisms have similar structures, and (2) the reality reis common to both parties in the process.
And what do the transcendentalists reply?

ferred to

They say that the

common

entire reasoning,

although

it

a "true" communication, and as regards the latter

a logical error

appeals to

can apply only to an "apparent", but not to

sense,

it

it

includes

assumes the truth of what in fact needs to

be proved.
Their arguments
First

of

all,

are

as

follows.

they question the assumption concerning the

"similarity of organisms". A mere external similarity, which


would perhaps suffice to explain behavioural communication,
communication purely situational in character, is not enough.
Intellectual communication implies understanding the content
of statements, and consequently the ability to think analytically,
to distinguish elements important and unimportant, significant
and non-significant. To ascribe to human minds such a common property goes beyond the recognition of an external
similarity of organism and presupposes the existence in them

of a

common

from the assumption that


separated minds (a thesis imputed by the

factor or force. Starting

there exist absolutely

The Philosophical Aspect of the Communication Process

139

transcendentalists to the naturalists) the naturahsts are unable


to explain intellectual

duce by

stealth, in the

communication and must therefore introform of a tacit assumption, the transcen-

dentahst thesis which they claim to oppose, and which argues


that there

is

some

transcendental, supra-empirical principle which

makes communication

And

possible.

of mere secondary

is

it

importance whether that principle be called the "transcendental

T"

or "similarity of organism".

Secondly,
thesis

transcendentahsts

the

that communication

environment

("common

made

is

reahty").

attack the naturahst

also

by

possible

That

similarity

further claim, not to purely situational similarity alone,

would be

On

sufficient in the case of behavioural

of

they

applies,

thesis

which

communication.

the other hand, intellectual communication requires reference

to a definite universe of discourse, that

to a reality

is

which

is

"construed". Thus, they maintain, communication presupposes

not only a

mind which "construes" a given

universe of discourse,

but also the community of the minds which communicate with

one another. The

is

to be proved

they assert, by introducing the

naturalists,

concept of "reality

common
the

all" simply presuppose what


community of minds, the possibihty of

to

communication.
cannot be said that transcendentalist

It
its

criticism,

even in

extreme form, finds no fulcrum in the expressed opinions of

the

naturahsts themselves,

opinions tendentiously, or

although certainly
is

it

distorts

rational in them. In naturahsts' opinions there are gaps


sistencies

and

is

and incon-

which expose them to attacks by the transcendentahsts.

In particular these gaps consist of:


cial

such

perhaps unable to reach what

historical explanation

(1)

the inability to give a so-

of the problem of the

human

and consequently, of the relationship between the


individual and society; and (2) the inability to take a consistently
individual,

materialistic stand in epistemology, that

of the

common

differences in

is

object of social cognition

its

to explain the issue

and of

the individual

perception by the various individuals.

140

Selected Problems of Semantics

FOUNDATIONS OF A MARXIST FORMULATION


OF THE PROBLEM

3.

No

Marxist, or any

criteria,

man who

thinks in terms of scientific

can accept the transcendentalist position, since

it

is

based on metaphysical speculation incompatible with science.

He

of course, the naturalist position to be

finds,

to his

own;

this applies in particular to those theses

close to consistent materialism.


solidarity with naturalism:

much

nearer

which come

But a Marxist cannot be in full


its points he must criticize,

some of

and he also has to fill many gaps in its argumentation.


Hence the Marxist rejects the notion that there are only two
alternative positions in the issue of the possibility of communication: the transcendentahst and the naturalist. Far less can
he agree that the bankruptcy of one of the two opposing theories
serves to prove the correctness of the other. This is an obvious
paralogism. Both parties to the dispute are right, at least to
a certain extent, in their criticism of their opponents.

No

is right,

however, in treating that criticism as adding to

capital.

Transcendentalism

tion based

is

simply

on mere metaphysical

counterscientific

faith.

Naturahsm
some of

and

inconsistencies; the incorrectness of

and

theses can be demonstrated.

what it
But there must be agreement with many of
they stand for science and
quently, has to treat the

common

two trends

its

reveals gaps
its

opinions

theses because

Marxism, conse-

sense.

differently,

although

declare itself in sohdarity with either of them. That

attempt to

own

specula-

The weakness of naturalism


does not say than in what it does say.

consists rather in

ist

party

its

is

it

cannot

why a Marx-

solve the problem of communication

must be

offered here.

"It is

only now,

when we have examined

the four points,

the four aspects of the original historical relations, that

found that

man

also has 'consciousness'.

we have

But even that

is

not

The Philosophical Aspect of the Communication Process

any a

priori, 'pure', consciousness.

beginning ridden with a curse,

is

The

'spirit' is

141

from the very

with matter which

'infected'

in as vibrations of the air, as sounds, in a word, as speech.

comes
Speech

is

as old as consciousness, speech

is

a practical, real con-

sciousness which exists both for other people and for myself.

And

speech comes to being, like consciousness, only from the

need, the necessity of contact with other people.


tion exists,

it

exists for

For an animal,

its

will

Where a

rela-

in a relation to nothing.

is

relation to others does not exist as a relation.

Thus, consciousness

and

me; an animal

from the very beginning a

is

remain so as long as

men

social product

will exist"23.

These words by Marx on the role of speech in the process


human communication date from 1 844. More or less at that
time, in the spring of 1845, Marx wrote his Theses on Feuerbach,
of

three of

which are of particular importance for us: Theses VI,

VII and VIII pertain to the social character of the

drawn

vidual and to the conclusions to be

manifestation of the spiritual

life

human

of

human

indi-

for the study of the


individuals.

The

theses are as follows:

"Feuerbach resolves the religious essence into

Thesis VI:
the

human

essence.

But the human essence

inherent in each single individual. In

its

reaUty

is

no abstraction

it is

the ensemble

of the social relations.

"Feuerbach,
real essence,
1.

To

religious

does not enter upon a criticism of this

abstract

from the

isolated

The human

hended only as

human
an

merely naturally unites the

itself

individual.

essence, therefore,

'genus', as

and to fix the


and to presuppose

historical process

sentiment as something by

an abstract
2.

who

consequently compelled:

is

can with him be compredumb generality which

internal,

many

individuals."

Thesis VII: "Feuerbach, consequently, does not see that the


'rehgious sentiment'
23

K. Marx,

1958, p. 30-31.

is itself

a social product, and that the ab-

F. Engels, Die deutsche Ideologie, in Werke, Vol.

3,

Berlin

Selected Problems of Semantics

142

whom

stract individual

ticular

form of

Thesis VIII

he analyses belongs in

reality to a par-

society".

"Social

life is

essentially practical.

AU

mysteries

which mislead theory to mysticism find their rational solution


in human practice and in the comprehension of this practice"24.
have opened

my

discussion of the Marxist interpretation

of the problem of communication with quotations from


since they constitute the foundation for the solution

Marx

of that

problem.

As

above, the transcendentalists object that the

indicated

naturalists

commit

the

error called

cir cuius

in

demonstrando:

required to prove the possibhty of communication, they instead

assume such possibility of communication in stating that it is


ensured by sufficient "similarity of human organisms". The
naturahsts are defenceless in the face of this objection, since
the assumption of naturahsm as the starting point for the ex-

planation of social

To

phenomena

involves a fundamental error.

human

begin with the issue of the

individual, the natu-

rahsts, like Feuerbach before them, conceive the

human

indi-

vidual in an abstract manner, only as a specimen of the species

"man". That is also materialism, but materiahsm restricted by


abstraction from the social factor. The only "general characteristic" possible in
istic

such an analysis of

man

is

the general character-

of the species, and hence the naturalists, like Feuerbach,

can conceive the

dumb

human

essence "only as 'genus', as an internal,

generaUty which merely naturally unites the

many

indi-

That weak point in naturalism is correctly attacked


by the transcendentalists. Once again it can be seen that ideaUsm
feeds above all on the limitations and the resulting restrictions

viduals".

of materialism.

For

if

the

human

individual

is

a specimen only of natural species,

isms"
24

the

is

treated naturaUstically,
if

as

"the similarity of organ-

understood in just such a way, then what can we say,

K. Marx, Theses on Feuerbach,

End of

Classical

German

in

Pliilosopliy,

F. Engels,

Moscow

Ludwig Feuerbach ami

1949, p.

62-63.

The Philosophical Aspect of the Communication Process

what

right

143

have we to speak about "the similarity of minds",

"the similarity of consciousness",

The

an assumption?

which

etc.,

transcendentalists

will

not be simply

are right.

With such

an approach nothing reasonable can be said on the matter, and


it really does seem that the naturalists assume what they have
to prove.

This

so because the issue of "the similarity of minds" can-

is

not be solved in purely naturalistic terms. The issue of the

human

individual as an individual belonging to

be solved at

all in

those terms alone.

In criticizing Feuerbach's naturalistic tendencies

ed that he had assumed "an abstract

and drew attention to the

vidual",

whom

vidual

the

in

Do

isolated

Marx objecthuman indi-

fact that "the abstract indi-

he analyses belongs in reality to a particular form

of society". This
cepts"

human

society cannot

is

social

a criticism of

"Robinson Crusoe con-

all

sciences.

the adherents of the naturalistic concept of

cation understand these truths?


listener" relation,

They

communi-

analyse the "speaker-

all

but leave the question of the social link between

Dewey, as quoted above, says that two such


parties belong to a social group from which they have adopted
their speech habits, but he draws no further conclusions from

them

in the dark.

that fact, I

do not imply that Dewey and other adherents of

nat-

uralism are lacking in understanding of the question of social


links, or that

they really maintain, as

transcendentahsts, that
lated".

human

is

imputed to them by the

individuals are "absolutely iso-

But the fact that they do not give voice to such nonsense

does not

mean

that they properly understand the significance

of the social conditioning of the communication process, and


still less,

is

that their opinions reflect that significance. No, there

nothing like that about them, and

it

explains

exposed to attacks by the transcendentalists.

And

why

pass over the point in silence because they consider


dent, a truism?

Such a defence of the

by pointing out that

it

is

probably a

naturalists
still

they are

perhaps they
it

self-evi-

can be answered

greater truism to say

144

Selected Problems of Semantics

two parties speaker


and yet the naturalists do not pass over that
silence (cf. Dewey, and above all Gardiner). If they

that a conversation always involves at least

and

listener

trivial fact in

did see and fuUy understand the importance of the social conditioning of the position of the individual as an individual in

should they see and fuUy understand the importance

society,

of the social conditioning of the communication process, surely


they would not remain

for this

silent,

is

an important

issue

which would help them to dissociate themselves from the standpoint of their opponents

the standpoint of transcendentaUst

mysticism. But in science silence, too,

an eloquent testimony

stances

definite situation

is

under certain circum-

the lack of certain assertions in

impMes approval of assertions to the contrary,

or at least denotes inability to oppose, or the impossibility of

opposing, such assertions to the contrary.

And

yet analysis should begin with the issue of the social

status of the individual: that done, then all secrets of "the similarity

of minds"

mysterious

and "consciousness common to aU", "the

character"

of communication,

etc.,

disappear

at

once.

Marx
in

said that "the

human

each single individual. In

social relations".

essence

its

reahty

is

no abstraction inherent

it is

The crux of the matter

the ensemble of the

lies

in understanding

that point (though Marx's formulation of the thought

somewhat obsolete
lar idea,

in style).

Can

or at least can he adopt

obstacles to his doing so. But

it

is

now

a non-Marxist develop a simiit? I

is

do not see any absolute

a fact that the interpretation

life and of the product of his spiritual life from


and consequently historical, point of view is the undenia-

of man's spiritual
the social,
ble

and immense contribution of Marxism to social research.


To avoid all misunderstandings, it must be added that Marx's

was preceded by
on human opinions

discovery did not take place in a vacuum, but


reflections

on the

influence of the milieu

(French materialists), and on the historical character of


consciousness

(Grerman

idealists).

Marx,

however,

human

not

only

145

The Philosophical Aspect of the Communication Process

gave a synthesis of those views, but also developed them consist-

on a new foundation

ently and,

(the attention paid to the role

human conmade a true scientific discovery which now lies at


of the modern theories of man's spiritual life and its

of the relations of production in the development of


sciousness),

the root

products.

The human

human

as a

cisely

ety

if we

individual

consider

him not only

of study of physics, chemistry, medicine,

object

is

individual,

artistic taste,

spiritual

as the

but pre-

member of human socisame way as are all the manias a

i.e.,

a social product, in the

festations of his

etc.,

hfe:

"religious

disposition",

speech,

consciousness in general.

As a "human

individual"

man

"the whole of social rela-

is

and

tions" in the sense that this origin

spiritual

development can

be understood only in the social and historical context, as a speci-

men

of a "species", but this time not only a natural, but also

a social species. These are historicism and sociologism in the


definite sense

of these terms. Thus, historical materialism has

introduced a sociological, scientific point of view to the study

of man's spiritual hfe in general, and the study of culture in


particular.

All this shows clearly that "consciousness

a social product as long


also, is

speech

men

such a social product from


is

the practical,

for other people".


is

as

...

The

...

real

and

...

and

its

will

...

existing

also

and speech

needs, "in the necessity to

in the origin of language

are interested in something else:

rious factor left in the process of

be

very origin, since "the

consciousness

communicate with other people".


But we are not interested here

We

its

is

But human speech,

origin of both consciousness

to be sought in social Hfe

speech.

exist".

is

and

there any myste-

communication when the

er-

from the naturahstic interpretation of social phenomena are eliminated, and the study is based on the principles
of consistent materiahsm? Can the transcendentahsts reasonably

rors arising

accuse Marxists also of the circuli

in

demonstrandol

146

Selected Problems of Semantics

Of

course, the transcendentalists

Equally inadequate

is

may

say after Urban:

mind when conceived

"...

in purely naturalistic

terms of history and sociology".


pass over the fact that history and sociology have nothing

to

do with naturalism. The

latter

term has probably been used

above to replace the "abhorrent" notion of "materialism". The


transcendentalists can say that communication,

those terms,

is

follow old Joseph Dietzgen,

who

if

we

not a true communication. But

analysed in

can, in reply,

referred the adherents of the

manner of communication to heaven, to the angels.


We here, on the earth, have at disposal only earthly cognition
and an earthly manner of communication. We need not en"angelic"

gage in speculation.

But does the argument which was so

smashed

been

the

thesis

effective in attacking

No, because

the naturalists, remain vaUd?

concerning

its

foundation has

"absolutely

isolated"

beings which the transcendentahsts might with impunity ascribe


to the naturalists in availing themselves of the gaps
in their doctrine.

nothing

is

left

Nothing

of the

is left

critical

of that

thesis,

and

errors

and consequently

argumentation of the transcenden-

tahsts.

The human individual is a social product both in his physical and his psychical evolution, both from the point of view
of his phylogenesis and his ontogenesis. There is nothing mysterious in "the similarity of organisms", as there

is

nothing myste-

rious in "the similarity of minds" or "the similarity of conscious-

by the way, leaves room for individual differences) is most natural and normal, being acquired
by up-bringing in society, by taking over its historical heritage
chiefly through the intermediary of speech. Both these factors
ness".

That

similarity (which,

have the same


is

effect

on aU members or

nothing extraordinary in the fact

tum

for a "similarity of minds". There

and hence there


that they form a substrais

society,

thus nothing extraordi-

nary in the fact that they smash the myths of "absolutely separated"

individuals.

It

becomes

quite superfluous

to

introduce

The Philosophical Aspect of the Communication Process

transcendental factors in

mystical,
cess of

order to

communication. The explanation

not naturalistic.

explain the proquite

natural, but

It is social.

what Marxism contributes to


The Marxist approach makes

This

is

ication.

is

147

problem in a consistently

scientific

the problem of

commun-

possible to solve that

it

manner, dissociating Marxism

both from the metaphysical speculations of transcendentahsm,

and from the vulgar materiahsm of the naturahstic

interpre-

tation.

No

Marxism of

the rejection by

less decisive is

transcendentahst objection

that which concerns

of reahty" to which communication pertains.

ity

the second

"the similarIt will

be

re-

called that the transcendentalists refer to the fact that intellectual

communication always pertains to some universe of


and consider the reahty spoken about in the process

discourse,

of communication to be a mental construction.


base their circulus

in

which had had to be proved

What

On

this

they

demonstrando objection, for the similarity


is,

they argue, adopted as an as-

is not an environment independent of human mind, but an environment


which is a construction of that mind; these constructions are

sumption.

is

involved here, they contend,

said to be similar because their similarity

themselves,
in

or,

strictly

speaking,

is

which the individual minds somehow


All that

is

provided by the minds

by "the transcendental T'


participate.

pure mysticism, pure metaphysical speculation.

The

naturalists are defenceless against such arguments; having

once

made

its

concession to idealism, they cannot effectively fend

oJBF

blows. But for the Marxists, as consistent materialists, that

objection

is

of no consequence.

What

is

at stake here

is

the essential controversy between

materialism and ideahsm in the


ogy.

The controversy

is

field

the object spoken about by persons

another

is,

or

is

of ontology and epistemol-

focused around the issue as to whether

not, because

it

who communicate

exists outside their

dependently of them, the object of a

common

with one

minds and
discourse.

in-

148

Selected Problems of Semantics

Of
sism.

course, in philosophy one can afford the luxury of sohp-

But then not only

in general

all

all

common objects of communication,

the mystic "I" that creates the world as

but

and what remains

objects as such disappear,


its

own

is

private construe-

Here the difference between epistemological and ontological solipsism is blurred. One must realize clearly that we put

'

tion.

an end not only to the object of thought, but to communication


as well, since nothing remains to be communicated. This is understood even by transcendentaUsts

e.g.,

Urban

says that "a min-

imum of reahsm" is necessary for communication, and states;


"No coherent theory of communication can be developed on
subjectivist premises,

and

if

ideaUsm involves subjectivism,

ideal-

ism must be abandoned"25. But Urban declares himself against


subjective idealism

And he

from the standpoint of

immediately continues:

"On

objective ideahsm.

the other hand, no coherent

theory of communication can be developed without the notion

of transcendent mind as well as transcendent objects.

Any form

of reaUsm that denies this must be abandoned"26.

Thus the

objective character of the object of thought has to

be retained (although, according to the transcendentahsts, escape

can be sought in objective ideahsm), in order to avoid aban-

doning

all

reasonable theory of communication.

But the transcendentahsts may urge that they do not mean


a denial of the objective character of the object of thought

witness the fact that they recognize the objective character of

the object of communication, as

point
ple,

is,

we speak about

nite universe

Now

is

done,

e.g.,

by Urban. Their

they might say, that in communicating with other peoobjects; but objects located

by us

in a defi-

of discourse, according to the "language"

to construct that univers du discours,

we

use.

a "transcendental

mind" is necessary.
The argumentation is simply astounding. Yet it sufiices to
ask: "But why?", and the spell is broken. For, in fact, why is
25

Urban, op.

cit.,

p. 264.

26 Ibid.

The Philosophical Aspect of the Communication Process

149

a "transcendental mind" necessary for the construction of a


univers

du discours, but not necessary for the existence of the

object of knowledge?

And what

of communication being objective,

is

the meaning of the object

if in

communication

a subjective product, as a construction?

is

It

it

would be

always
difficult

to answer these questions, without reverting to the old concept

which have an objective existence,


and of apparent phenomena, which are our own construction.
of inaccessible

noumena,

But that concept was annihilated long ago by the criticism of


Kantian phenomenalism.
accepts

an

It

must be

stated clearly: either one

objective existence of the object of communication,

or one considers that object to be a subjective product of

and thereby cancels the

possibility of

mind

communication. Tertium

non datur. The transcendentalists entangle themselves desperately


in their

own

subtleties.

But what about that univers du discours in the

light

of the

materiahstic epistemology?
If

the

we hold

object

outside all

that the object of our cognition, and consequently

communication, exists objectively that is,


minds and independently of them then the mind
of

in the process of cognition

of the word).

mind
some

somehow

reflects

But an isolated object

is

it

(in

a special sense

an abstraction: the

in the process of cognition always reflects the object in

context. In that sense, that context must also exist in the


communication of cognition. The statement that the object

of cognition appears here in a definite univers du discours


ial.

Reference to the univers du discours in the process of com-

munication does not infer any


is

triv-

is

involved here

is

diflEiculties,

particularly since

what

not objects but ambiguous words the meanings

of which become concrete only in a definite context.

And

for

no "transcendental 'I'" is needed;


and to understand the context, which in turn reprecision of formulations on the part of the speaker and

that purpose

suffice

it

to in-

dicate clearly

quires

adequate erudition on the part of the


scendental

T"

will help

me

listener.

No

mystic "tran-

in understanding the univers du dis-

II

150

Selected Problems of Semantics

cours of, for instance,

quantum mechanics,

if I

am

not familiar

with the subject.

One
"Social

compelled to recall Marx's words already quoted:

feels

which mislead

essentially practical. All mysteries

life is

human

theory to mysticism find their rational solution in

practice

and in the comprehension of this practice".


The standpoint of the Marxists in the controversy between
the naturahsts and the transcendentalists thus boils down to
demonstrating

controversy

that

that

that are social par excellence

not a truism;

Is this

ed in science? Not at

That
its

is it

naturalists

to assess

is

fact that not

do not

in that sense natural.

not the standpoint commonly adoptbest proved

entirety as explained above.

proved by the

on phenomena

all.

not a truism

it is

and

focuses

fully

That

it is

by the controversy
not a

in

triviality, is best

only transcendentalists but even

understand the thesis and are unable

properly.

it

Certain standpoints, especially that adopted


scendentalists in the

controversy

now under

by

the tran-

look

discussion,

strange in the light of common-sense. But the opinions of philos-

ophers are not always in agreement with


often
to

common

all their "originality" consists precisely in

common

Should we then

sense.

(as

sense,

and

applying shocks

was suggested

to

me

in

a discussion) consider such opinions as manifestations of imbecili-

and consequently reject them as unimporas was said in that discussion either
ascribe to them such absurd opinions, I do not understand

ty or schizophrenia

tant?

May we

that, if I

say

what these authors mean, or that


absurd then

am

if their

opinions are really so

absolved from discussion in view of the evi-

dently non-scientific character of the position held by

ponents?
other.
is

do not agree that we can say

Contrary to

all

either the

my

op-

one or the

appearances, the holding of such opinions

a manifestation neither of weakness nor of insanity, but

simply part of irrationalist thought, so


geois world.

Even mysticism

is

a social

common

is

in the bour-

phenomenon and

a social

The Philosophical Aspect of the Communication Process

151

which must be noticed, understood and properly assessed,

fact,

one does not want to come out on the losing side in the ideological strife now proceeding in philosophy. Nor may I say,
if

although

may

find

withdraw from the

convenient, "I do not understand", and

it

struggle. After

all, I

do understand the mean-

ing of sentences, even if I reject their content as unacceptable

that

understand the meaning of the sentences to the

(e.g.,

communication

effect

some transcendental

possible owing to

is

meaning of the sentence which


an omnipotent God). The words "I do
not understand" can only mean here that I do not agree with
a given standpoint, that I deny it aU meaning from the scientific point of view, etc. But it would be dangerous to abstain,
as I understand the

"I", just

asserts the existence of

for that reason,

from

discussion, for then the

opponent could

claim successfully that he too "does not understand" what

Perhaps we should not bother about

say.

of being right? But that would

an end to

progress,

of the

So

scientific

let

discussion,

that,

convinced

to all intellectual

a harmful monopolization

standpoint.

us engage in discussion. But

Marxist

the

all

all

mean an end

we

standpoint

on the

if

discussion

issue

of

is

indispensable,

communication

is

neither a truism nor a triviality. Quite the contrary; not only

from the point of view of


it is

historical priority, but also at present

the only consistent formulation of the issue of communication

from

the

social

both

materialistic

and

historical

point

of view.
Certainly,

great

discoveries,

once

assimilated

assume such apparent simplicity as to seem

trivial to

by

science,

those whose

opinions have already been shaped by such discoveries. This


is

the normal course of events.

in the social sciences, usually

The

greatest discoveries, especially

concern simple, every-day matters

which for various reasons have not been noticed or understood


which has been deliberately put off.
by no means detracts from the greatness of the discovery, far less does it preclude one from availing oneself of it
before, or consideration of

That

fact

152

Selected Problems of Semantics

in analysis. This holds especially

when, as

the case of the theory

is

of communication, the discovery in question

is

neither universally

recognized, nor even understood and taken into consideration

by the majority of researchers.

It

has been said that the Marxists treat

human

consciousness

and human speech consciousness for others as products


of social life. That hypothesis has found expression in Engels's
paper On the Role of Labour in Making the Ape Human, and
in the theory of

Marx and

Engels on the role of the division of

labour as a factor in social evolution. Labour

thought lan-

guage: these are the three elements of fundamental significance


in the Marxist conception of the origin of

human

society.

These

from another. Man drew


himself
and the animal world when
a hne of distinction between
he started to produce tools, says Marx. Human labour is inseparably linked with consciousness, that is with thought, which
in turn is genetically hnked inseparably with speech. Consciousness, and consequently speech as well, are products of labour, products of social life, and at the same time indispensable conditions
of a further development of that process, of its higher, more
advanced stages. Human labour is based on co-operation, which
three elements are inseparable one

is

impossible without thinking in terms of ideas, and without

communication. Such

makes

it

is

the dialectics of mutual influence, which

possible to explain the process of communication with-

out recourse to miracles and metaphysics.

Thus, the issue turns out to be "prosaic" and natural, though


evading a naturalistic interpretation by being an essentially so-

phenomenon. But is it so simple? Is it enough to state that


somebody speaks and somebody else listens to him and the
two understand one another?
Here a protest must be registered against a "common-sense"
simphfication of the problem, since that would threaten to eliminate
all deeper analysis and to annihilate the scientific approach.
cial

153

The Philosophical Aspect of the Communication Process

We

say that people talk to one another and that

is

communi-

So far so good. But from the point of view of a

cation.

scien-

problem only begins here. What problem? Not the


as to whether communication is possible.
one,
metaphysical
analysis the

tific

Of

course,

is

it

we

possible, for

witness

it

everywhere.

Not

the

mystic one. as to what transcendental factor makes communication possible. This can be explained without resorting to mira-

and metaphysics. The

cles

this point

is:

how,

problem which begins

scientific

at

what manner, does communication take

in

place?
Social psychologists say that

the

parties

communication

consists in that

concerned mutually interchange their roles with

their opposite

numbers, they mentally place themselves in their

The issue
may be formulated so, it may be formulated otherwise. Be that
as it may, intellectual communication is always connected with
understanding, with the same understanding by the two parties
and thus come to understand

position

their words27.

of certain definite statements.


Intellectual

arably
I

communication based on understanding

connected

with speech.

mean by language and speech

A
will

point I should like to stress here:


fines speech, regardless

of

all

stricter

be given

who

insep-

of what
is

one

how one

de-

later.

regardless of

the enormous

respect between the various authors

definition

is

There

differences in that

often contend one with

another on that issue, every definition refers to signs or symbols comprising

In speaking,
ticular

human

man

speech.

produces certain phonic signs of a par-

kind (other authors speak of articulated sound symbols).

Communication consists in that the person who produces those


phonic signs and the person who hears them imderstand them
in the same way, that is impart to them the same meanings. That
is precisely the definition of communication advanced by Lund27 Cf.

York

G. A. Lundberg, C. C. Schrag

1954, p. 389.

&

O. N. Larsen, Sociology,

New

154

Selected Problems of Semantics

berg

when he

says:

"Communication can be defined

as transmis-

sion of meanings through the intermediary of symbols"28.

Thus we have introduced

into the analysis of

communication

the three fundamental notions which require further investigation: sign (symbol),

meaning, speech (language).

To understand

and to be able to explain sensibly


its
effectiveness,
the social conditions of
we must first analyse
these three notions and their related problems. Thus the preliminary analysis of the process of communication naturally
outlines the programme of further research.
the sense of communication

28 Ibid., p.

360.

Two

Chapter

THE SIGN: ANALYSIS AND TYPOLOGY

ST

"We

next went to the School of Languages, where three Profes-

sors sat in Consultation

The

first

The

all

that of their

and leaving out Verbs and

lables into one,

Reality

upon improving

own

Country.

Projects was to shorten Discourse by cutting PolysylParticiples; because in

things imaginable are but Nouns.

other,

whatsoever:

was a Scheme for entirely abolishing


this was urged as a great Advantage

And

all

Words

in Point of

it is plain, that every Word we speak


some Degree a Diminution of our Lungs by Corrosion: and

Health as well as Brevity. For,


is

in

consequently contributes to the shorting of our Lives.

An

Expedient

was therefore offered, that since Words are only Names for Things,
it would be more convenient for all Men to carry about them such
Things as were necessary to express the particular Business they are
to discourse on.
)

I have often beheld two of those Sages almost sinking under


Weight of their Packs, like Pedlars among us; who when they
met in the Streets would lay down their Loads, open their Sacks,
and hold Conversation for an Hour together; then put up their
Implements, help each other to resume their Burthens, and take
...

the

their Leave".

(Jonathan Swift: Gulliver's Travels)

1.

THE COMMUNICATION PROCESS AS THE STARTING POINT


IN THE ANALYSIS OF THE SIGN

The

assertion

that

the

social

process

of

communication

should be the starting point for an analysis of semantic problems

problems of sign and meaning.


which are fundamental for semantics, the
and sociological aspects are neglected, it is easy to end

finds telling confirmation in the


If in these problems,

social

up

in a blind alley of verbalism

and formalism

(history has ac-

tual examples of such cases).

In order to be able to answer properly the question as to

how

the process of

human communication
[155]

takes place, in par-

il
Selected Problems of Semantics

156
ticular, to

be able to answer the question, all-important fromr

the social point of view, as to what helps and what hampers


effective

human communication, we must

analyse precisely such

and "meaning", and must fully realize the


sense of these terms. But just because both sign and meaning
are elements of the communication process, the starting point
from which to analyse them must be the whole of the social
process of human communication. Any analysis detached from

categories as "sign"

that process

Such

is

would be one-sided, and often

warped.

entirely

the usual fate of the dialectics of the relationship between

the part and the whole.

Of

course,

possible to engage in the typology of signs

is

it

or in philosophical speculation on the essence of meaning


all this

from the

in isolation

cation process.

The

background of the communi-

social

extensive history of the problem can readily

provide appropriate examples.

It

problem

the social origin of the

is

also possible to hold that

obvious and has been

is

tacitly

assumed, and that such an assumption does not introduce new


elements to the analysis of the problem and consequently should

not interfere with

its

course. That, too, can of course be illus-

trated with historical examples.

Yet in both cases we have to do

with standpoints which, by separating the analysis of sign and

meaning from

its

wide for

philosophical speculation.

sterile

The problem

natural

social background,

and

as to the essence

open the door

role of the sign,

sequently the problem of the typology of

its

and con-

various forms and

and when it is considered as


part of the problem as to how men communicate with one another.
Let it be clearly understood that we mean here the human

varieties,

process

can be seen whole only

of communication. As already mentioned, communi-

can be observed not only on the human

cation, in various forms,


level

of development,

reference

if

is

often

made

only to communication

but also on the animal

in

some

among

signals in the process of

specific sense

level.

of course

Hence

not

animals, but also to signs and

communication among animals. What

157

The Sign: Analysis and Typology

ambiguous concepts which may

are concerned with here are

We

obscure the issue and bring about numerous misunderstanding.

Obviously, the communication process takes place in the

animal world only in so far as the process of co-operation

tion

is

in

very origin linked inseparably with co-operation

its

covering both co-operation

the broad sense of the word,

(in

proper and struggle). For

need

and the origin

for,

it

of,

Im Anfang war
(on the

human

level

and not

(scientific,

genetic sense, of
also

who

held that

was the deed).

It is in

action which transforms the world

we speak of

social action), that philosophy

speculative) seeks the solution, at least in the

many problems

communication

the

for

common

find the

communication of the co-operating

die Tat {In the beginning

in

i.e.,

we

in joint action that

is

individuals. This confirms the view of the poet

practice,

is

communica-

involved, a process of social action sui generis. All

of consciousness. This holds

process

and

the

problem

of

the sign.

Bees in the hive co-operate in finding and collecting honey;

can be observed among ants in the anthill,


the deer in a herd, etc. In each of these cases what is involved
similar co-operation

is

a specific process of communication between bees, ants, deer,

etc.

In a sense, a bee which by

to a search for

ing

and

honey gives them

is

etc.,

"dance" stimulates other bees

"signs". All this

is

very interest-

discloses a variety of problems, not only in the matter

of animal psychology,
it

its

calling

for

investigation.

Nevertheless,

obvious that when speaking about communication, signs,


we have in mind something different when we refer to animals

from what we mean when we refer to human beings and human


society. For the sake of clarity, and to avoid ambiguity, it is
correct to reject obscure analogies and the speculations based
on them, and to
cifically

human

restrict

the sphere of our interest to the

process of communication, of using signs,

In referring here to a specifically


tion, I pass
in

animals

human

speetc.

process of communica-

no judgement whatever on the character of that process


I

merely consciously

restrict the field

of

my

analysis.

Selected Problems of Semantics

158

Such a research procedure

is

not only permissible but even ad-

visable in this case.

Men

in various ways,

communicate

and the origin of

the

various concrete manifestations of that process are also diverse,

on the higher levels of the communication process


when the motives of communication are not confined to biologespecially

matters, requirements of production,

ical

need to exchange

Yet

men

Hence

abstract ideas,

need for a
writing,

etc.

always communicate by means of signs in various forms.

the practical

Men

but include the

etc.,

stimulate emotions,

to

theoretical importance of signs

and

definite theory

of

signs.

communicate by means of
pictures,

we have

gestures,

phonic language,

previously agreed upon,

signals

etc.

But

in

do with signs. Gestures, speech sounds,


some form of signs which in
turn, arranged in a system, constitute a form of language.
Just because man always communicates with other men by
means of signs, all social life is permeated with signs and is imall

such casses

writing,

signals

all

to

these are

Even those famous Balnibarbi scholars

possible without them.

of

whom

Gulliver said that to spare themselves the effort of

speaking they carried with them


their conversations,

tures

pointing to

wonder, then, that


ophy, have

had to use

objects,
signs,

now come

all

the objects connected with

signs,

imitative

however primitive:

gestures,

or pictures.

ges-

No

long since an object of interest in philos-

to be considered

by certain philosophical
For instance,

schools to be the principal subject matter of study.

Susanne Langer, the author of an interesting work Philosophy


in a New Key. A Study in the Symbolism of Reason, Rite and Art,
sees even in the

problems of sign and symbol an announcement

of a rejuvenation of philosophy which, in her opinion, suffers


a crisis caused by the exhaustion of the traditional problems
of that discipline.
ally, yet

forms,

it is

One may view such

radical opinions sceptic-

undeniable that the problems of signs do, in various

come more and more

and the

to the forefront in the investigations

undertaken by the various branches of philosophy.

And

rightly so.

.t

i|

The Sign: Analysis and Typology

We

have so far restricted the sphere of our

specifically

human

restrict it still
;

159

process

of communication;

more, to the sphere of

intellectual

interest to the

now we

shall

communication,

thus eliminating those processes which are connected with the

transmission from
intellectual

man

to

man

of emotional states by extra-

means. In the communication process thus formulated,

certain intellectual, cognitive content


to

man by means

gesticulate,

objects

actions

When

man

speak, write,

produce signs characterized by a similarity to the

they

denote

when I

or symbohzing certain

give signals established

in all these cases the specific sign


definite

transmitted from

is

of such or other signs.

mental content, and

use

by a convention,

for

is
it

abstractions

me

connected with a

in order to

one else the same content. In other words,

etc.

or

evoke

we say

in

some-

that in the

same meaning for the


persons communicating, and that the communication process
transmission
consists
in the
of meanings by means of
communication process the sign has

the

signs.

Two

communication as transmission of meanings and communication as


transmission of convictions. These two matters are in the literaaspects

must be distinguished

in this connection:

ture of the subject not only connected one with the another, but
actually confused;

a situation which certainly does not sim-

phfy the intricate problem of communication. The sense of that


distinction will

be demonstrated by an example.

When someone

communicates to me, for example, the thought concerning divine omnipotence, I understand perfectly what he means, but
it

by no means follows that

agree with him. For

it

does not

sufiice to understand the meaning of words or other signs used

in a given case in order to share the conviction of the person

who pronounces the given words or uses the signs in question.


To develop common concordant convictions, people have not
only to understand in the same way the thoughts which are expressed, but also to approve the reasoning behind such thoughts.

In this book,

we

shall

be concerned only with the process of

160

Selected Problems of Semantics

communication in the sense of transmission of meanings

and*!!

the role of signs in that processi.

What

human com-

types of signs appear in the process of

munication, and what

wiU involve a

is

their nature?

specific analysis

The

reply to that question

and typology of

signs as an in-

troduction to an analysis of meaning.

As

indicated above, the starting point of an analysis of the

and consequently of meaning (since sign and meaning


are not two independent "entities", but a whole which in the

sign,

process of cognition
ess of

is

communication,

analysis of the issue,

divided into parts or aspects),


i.e.,

it is

essential to start

is

the proc-

activity.

In a Marxist

from that

point, which

a definite social

also constitutes the rational element in certain endeavours to

give a behaviouiist, pragmatist or operationist analysis

(cf.

Peirce,

G, Mead, or Morris). In approaching the problem thus, we see


in the

communication process

(in particular in the interpretation

of that process as an effective transmission of convictions) an


attempt to place oneself in the position of the partner in that
process. In the literature

on the

subject, there

comparison with a game of chess:


envisage his

own

plan of attack,

the player

is

a well-chosen

must not only

but also the possible plans

i.e., he must estimate his opponent's capacity


and appraise the various moves. All social dialogue
connected with co-operation, and consequently with mutual

of his opponent,

to apprehend

understanding, consists in such placing of oneself in the position of the partner, and in attempts to envisage his situation.

The same

applies to a dialogue sensu stricto, that

is,

exchange

of thoughts and transmission of meanings through the intermediary of signs.


1 The issue of effective communication in the sense of reaching agreement as to opinions, in the sense of conveying convictions, is much broader
and includes, as mentioned above, the problem of good understanding as its
component part. Those who are interested in the issues of propaganda and
in shaping public opinion are above all concerned with that broader sense
of communication. This is an extremely important social problem, some

aspects of which will be discussed below.

The Sign: Analysis and Typology

The learned of Balnibarbi used

to carry

161
sacks containing

a variety of objects in order to spare themselves the necessity


for talking.

But

it

was a misconceived

idea, not only because

need to carry great loads. As already indicated,

entailed the

even in such an extreme and absurd case


to

it

ehminate the use of

signs,

it

would not be possible

such as the gestures of pointing

to the objects in question or imitating certain actions.

The fun-

damental misconception, however, consists in something

else:

those respected scholars could spare their lungs by abstaining

from

work

talking, but they could not

the miracle of eliminat-

ing thinking in terms of language. That was not possible simply

no thinking other than in terms of language,


and "true cognition" and "direct cognition" can at the most be

because there

is

a subject of philosophical speculation.

I shall

not discuss here

whether the system of thinking by means, and through the


termediary, of

all

in-

kinds of signs can be interpreted as a criterion

of being man, a criterion distinguishing the

human world from

the animal world, although everything seems to indicate that

such really

is

one of the possible

from others,
beyond doubt, and

criteria (inseparable

especially the criterion of labour).

But

of the greatest importance here,

that in the stage of phonic

it

is

language and the related system of thinking in terms of ideas,


every other system of signs, that
neris, is

is

every other language sui ge-

dependent on phonic language in the sense that

phonic language and in the

final stage

it

replaces

of communication

is

translated into a phonic language ; thus, used to replace a phonic

language,

it

makes a system of

signs of other signs (of a phonic

language). Failure to acknowledge that fact in the study of the

various systems of "languages"

is

a serious error, pregnant with

untoward theoretical consequences: above

all,

it

falsely

sug-

and autonomy of such "languages". The danger


such an error justifies, among other things, our

gests the equality

of falhng into
thesis

concerning virtue of approaching the issues of semantics

from the point of view of the communication process as a whole.


only on the basis of a social analysis of the problem that the

It is

162

Selected Problems of Semantics

proper hierarchy of the various systems of signs can be recognized,

and the inter-relationships between the various

"languages"

determined.

Such an analysis of signs

reveals,

first,

their

variety and,

and importance in the procof human communication; and, secondly, their homogeneity

in a sense, the hierarchy of their role


ess

in the sense of sharing a

common

property, namely that

all signs,

dehberately produced to serve the needs of the communication


process, are vehicles of meanings, since all of

them are

derivative

with respect to a phonic language as far as their communicative


function

The

concerned.

is

rest

of this chapter will deal with just that thesis and the

related issue of the typology of signs^.

HUSSERL'S TENTATIVE TYPOLOGY OF SIGNS

2.

The
theories

vast

literature

of signs

on the

subject

advancing the various

(and consequently the various suggestions,

often contrasting one with another, as to the typology of signs)

two conclusions.
very volume and divergency of such suggestions
impels rejection of the idea of making a synthesis of them and
implies

First, the

rather forces an attempt to


making use of earlier ideas

give a new, independent solution,


as specific intellectual stimuli.

Secondly, the analysis of such standpoints shows


differences

that the

between them, including those which pertain to ty-

pology, are not only formal or terminological in nature, but

2 When this book had akeady gone to press I came across the book
IIoHHmue u CAoeo [Notion and Word] by JI. O. PesHHKOB. I regret I have
not been in a position to take into consideration here many interesting ideas
of that author. I should only like to emphasize that I am in solidarity with

many

of his opinions and

of the concept of the sign

interpretations, especially as far as the analysis


is

concerned.

163

The Sign: Analysis and Typology

have underlying philosophical aspects. These issues should be


aired so that the
incorrect opinions

As already
process of

problem may be

on the point

better illuminated

and the

refuted.

indicated, the starting point of analysis

human communication

is

the

as a separate entity of a cer-

tain social process. In advancing such a theoretical thesis

not imply that research actually begins from

No,

it.

do

this is

not

a thesis adopted a priori, but a thesis obtained as a result of study

and

investigations. I

among

I referred,

have already taken cognizance of

this v^hen

other things, to Marx's proving that the use

is normal in practice.
Thus we start from the process of human communication
which, though complicated in its course and functions, is a selfevident fact, a fact that men communicate in action, i.e., in
co-operation (since all action is social action), by transmitting
definite meanings through the intermediary of signs; by so doing,
we estabhsh a specific framework and theoretical background

of such a method

for the analysis of signs.

that

In such a context, the contention


a sign only

if

we have

to

cluded in the communication process, seems simply


shall see,
is

do with

a given object, property, or material event

however, that that statement

is

not

in-

is

trivial.

trivial at all,

We
but

of considerable importance for a proper understanding of the

nature of signs and their appropriate typology.

We
of a

say that the freezing of water

fall in

is

(symbol) of mourning,
(signal)
it

moon

that firing

a sign

is

going to deteriorate, that

pass to another category of examples

we

since

a sign (indication, index)

temperature, that the halo around the

(indication, index) that the weather


if

is

a red rocket

black

is

is

a sign

an agreed sign

of some action, that a definite sound means this or that


is a sign (verbal sign) belonging to a given language, etc.

But taken "in

itself" the freezing

of water

is

a natural process,

a natural phenomenon, and no sort of a sign. Likewise a red


flare high in the air, black, certain specific vibrations of aerial
waves,

are

"in themselves"

material

objects,

possessing

their

164

Selected Problems of Semantics

own

properties,

It is

only within the process of human communication that these

constituting actual events,

natural phenomena, objects, events,

namely those elements of a

etc.,

and are not

become something more,

process which

social

signs.

are

called

signs.

There

new

is

is

that

new in this, any more than the requirement


an analysis of the elements of the communication

nothing

process should begin with the social conditioning of that process.

Neither

is it

although
sistent

it

an exclusive achievement and possession of Marxists,


is

Marxism which

most con-

potentially creates the

foundation for such analysis. The issue as to

how

it

is

and events become signs within the process of


himian communication, was noticed and perfectly well understood

that material things

by

Peirce, for example.

He

emphasized forcefully that a thing,

a property of a thing, or an event,


if

they are interpreted,

i.e.,

if

there

all

is

function as a sign only

someone who

in the process

of communication acts as an interpreter of the given thing, event,


etc.,

used as a sign. That opinion

is

now

being supported by,

and forms one of the fundamental ideas


of his semiotic, which by the way can be derived from the ideas
of Peirce. The same view is held by Susan Stebbing {A Modem
for instance, Morris,

Introduction to Logic)

the statement

is

and

others. Thus, the point is not whether

new and who

initiated

it.

The

essential point

what systems does that statement appear, and what are


consequences drawn from it.
is,

in

It is to

be borne in mind that increased interest in the problems

of sign and symbol has opened in contemporary


a

new

the

field

philosophy

not only for study and research, but also for philo-

sophical speculation. Sufiice

it

to indicate,

by way of example,

the basic ideas of Cassirer's theory of symbols to demonstrate


that idealistic speculation can use for
cepts. Cassirer certainly

is

its

purposes

a penetrating thinker,

credit considerable achievements in the analysis

on, symbols

man's

and symbol systems. Yet

allegedly

innate

faculty

all

new

con-

and has to his


of, and research

his assertions concerning

of symbolizing, of "creating"

The Sign: Analysis and Typology


reality

through symbols,

etc.,

cannot be endorsed by anyone

savour of

who

165

idealistic speculation

does not accept their

and

idealistic

philosophical background.

The
a sign

of a consistent observance of the thesis that

first result

to be analysed in the context of the

is

process, that

is,

that

communication

phenomenon within
phenomenon which is inter-

only an object or a

an object or a
preted by someone) can be a sign, involves the undermining
that

process

(i.e.,

or even the abolition of the traditional endeavours to classify


signs. I

invoke here above

all

the views of Husserl, which have

had an enormous influence on the


and meaning.

What

is

striking

is

concerning signs

literature

the great ambiguity of the term

both in ordinary language and in

precision to terms. Another striking factor

is

"sign",

attempts to impart

scientific

marked vagueness

the

and even arbitrariness in the terminological distinctions between


"sign", "index", "symbol", "signal", etc.

No

wonder, then, that

attempts to explain the function of signs are intimately connected with attempts to establish a typology of sign which would

make

possible in turn to establish a hierarchy of signs from

it

the point of view of their extent and content, and thus to bring

some order

What

into terminological matters.

is it

typologies?

that

is

most

striking in the case of such attempted

Probably the division of things and phenomena

which function as signs into those which are natural and appear

human

and only ex post


and into these
which are products of man's conscious social activity and have
been produced by man in order to function as signs. The former
regardless of any purposive

are interpreted

man

by

activity

as signs of something,

are called natural signs, the latter, proper or artificial signs.

In a sense,

we

say that the freezing of water

in temperature, that a halo

weather

is

around the

moon

is

is

a sign of a

fall

a sign that the

going to deteriorate, that wrinkles on a person's face

are a sign that he

is

ageing, etc. In

black crepe on a flag

is

some other sense we say

that

a sign of mourning, that the colours

12

166

Selected Problems of Semantics

of a flag are a sign of nationality, that

monument

to start the attack, that a

on a handkerchief

that a knot
thing, etc. In

still

another sense,

is

is

that a definite

sounds

is

is

is

a sign

a sign of an historic event,

a sign to remind one of some-

we say that the words we pronounce

are phonic signs, that a written sentence

a wink of the eye

a red rocket

firing

is

a written sign, that'

a sign of communication between persons,

number of dots and dashes or

short and long

a sign belonging to the Morse code, that certain ink

marks are mathematical or logical signs, etc. All these are signs
in some sense, but they are different signs with different meanings.
When reference is made below to signs tout court, that will

mean proper
sciously

signs,

i.e.,

produced by

artificial in

man

the sense that they are con-

for the purpose of

communicating

with other men. Although the natural signs (indices, symptoms)

'

fall

under the general category of "sign", they

differ essentially

from all other categories of signs, above all in that they are not
produced or evoked consciously by man for communication
purposes, but exist independently of

in

man

as natural processes

by men as source of information and


such cases function as if they were normal signs, i.e., as if

and are only ex post

utilized

they were consciously evoked or produced for the purpose of

conveying some information to some person (a similar


pretation of the problem
cf.

footnote

3).

By

is

to be

inter-

found in Martinak's work

interpreting "meaning" in the context of

interhuman relations in the process of communication (which


will

be discussed in the next chapter), we

signs a

meaning

signs, a

least for caution's sake,

and

to the natural

sense. This is

why

at

induced by the controversial character

a clear distinction must be

indices (natural signs)


3

ascribe to the proper

in the direct sense of the word,

meaning only in some derived

of the issue

may

and proper

The terminology adopted

in this

made

in analysis

between

(artificial) signs^.

book and

to be used

below requires

additional explanations.

The

division into natui'al signs

and some other

signs,

which are

sort of opposition to the former group, has a long standing.

in

some

There are also

The Sign: Analysis and Typology

167

Husserl in his Logische Untersuchungen effected an essential

of signs into Anzeichen (indices) and Ausdriicke (ex-

division

The former, in his opinion, would have to point to


and replace or represent that something else, and
would express some thought and would be signs sensu
as the two extremes of the wide gamut of signs, we take,

pressions).

something
the latter
stricto. If,

else

on the one hand, the natural signs (indices) of the type of frozen
water or the moon's halo, and on the other the verbal signs,
we reach the model of the typology of signs made by Husserl. He

name

various suggestions as to the

of the latter group, and at least

some

of them require explanations in order to show the motives which have guided
in making my own choice.
The ancients distinguished between

me

the natural sign {signum naturale)

and the conventional sign (signum ad placitum). This preserves a uniform


criterion of division, but the division itself does not bear examination, since

the latter group

would have

which are consciously

to include all those signs

produced by men for communication purposes. Yet not

such signs function

all

on the strength of a convention. This applies above all to the iconic signs,
which function on the strength of similarity to the object they stand for (they
stand for such objects in the sense that a picture makes us think of
the

objects

which

it

represents,

an object, of corresponding
respect:

from natural

of the effects

similarity,

caused by

usually

There

feelings, etc.).

such

a whole scale in that

is

such as in the case of photographs, to the

conventional nature of hieroglyphs or similar written signs (a classical ex-

ample
atic

is

offered by the letters of the

Hebrew alphabet

their

shape

is

representation of those objects the names of which are the

the letters in question: "alef", "beith",

etc.).

iconic sign such as a photograph does not

Yet there

work on

is

a schem-

names of

no doubt

that an

the strength of any con-

vention.

Another terminological suggestion, with regard to which an attitude


has to be adopted, comes from Martinak

(cf.

E. Martinak, Psychologische

Untersuchungen zur Bedeutungslehre, Leipzig 1901).


final

signs,

explicitly stressing in his

of that group of signs which


the fact that
in

men produce

view which

The

is

in

speaks of real and


is

characteristic

an opposition to the natural signs, namely

such signs in a conscious manner with that end

communication.

division adopted

Martinak, yet

is

He

terminology that what

above

in this

book

in principle follows that of

consider his terminology not very fortunate, since the prin-

168

Selected Problems of Semantics

tried to squeeze the full wealth of the

phenomena ambiguously

Two

make

reasons

it

bed of extremes.

referred to as "signs" into that Procrustean

advisable to take precisely that typology

as an object of analysis: (1) Husserl's system demonstrates, so

manner, the consequences of disregard-

to speak, in a classical

ing

principle of interpreting

the

problems of sign and meaning;


a great

many

signs includes

socially
(2) his

and

historically

the

system has influenced

work dealing with the problem of


some typology of signs from this or that point

authors. Every

of view. Martinak, Biihler,

Morris, Carnap, Cassirer, Langer,

own

to quote but a few names, all give their

typologies of signs,

but none of them can vie with Husserl's in the matter of influence
exerted

on

others. It

might be said that Husserl had his pre-

decessor in Peirce: the latter used a different terminology (index,


iconic

sign, symbol),

but the sense of the division was practi-

cally the same"*; yet Peirce

was for many years unknown

as

an

author, and hence his influence could be but small.

Thus the

starting point for a critical analysis is the division

of signs into indices (Anzeidien) and expressions or expressive


signs (Ausdriicke), with the proviso that according to
ciple of division does

Husserl

not seem to be uniform, in spite of the fact that an ap-

propriate interpretation

may

easily

waive that objection aside (the term "real

signs" coincides as to extension with the term "natural signs").

In opposing the proper signs to the natural signs

again start from the

communication process. In that process, these are signs which are being
consciously produced by men for communication purposes. In my opinion
the term "proper signs" is the best term with which to denote them since natural processes function as signs only in a secondary

and a derivative

sense.

Alternatively with the term "proper signs" I use the term "artificial
signs", since I

wish to emphasize that in contrast with the natural signs,

which are natural processes


signs always are, in

such are
point,

artificially

independent of

human

activity,

the

proper

one way or another, products of human activity, and as


brought to existence. This I beUeve to be a very important

which for a better description of the various categories of signs should

also be reflected in terminology.


4 See

above

in Philosophical

all

Ch.

S. Peirce,

"Logic as Semiotic: The Theory of Signs",

Writings of Peirce,

New York

1955.

The Sign: Analysis and Typology

169

only the latter perform the function of expressing thoughts, or,

meaning something. That division undoubtedly

in other words, of

takes into account the specific nature of verbal signs, their specific
is

expressive property, which in the literature of the subject

most often called "transparency

poses to the Ausdriicke

meaning". Husserl op-

to

other signs such as the Anzeichen,

all

blurs all the possible difference

among

the latter (which

may

sing something. Husserl's typology also takes into account


in

that he

certainly

is

be

and denies to them the function of expres-

highly significant)

right

the

of the

specific character

and
in-

same time it blurs


the difference between the indices so interpreted and the signs
of various types, which he lumps together in the same category
But

dices in the sense of natural signs.

(I

mean here above

iconic

signs,

all

such signs as we should

call signals,

symbols,

Husserl's Anzeichen are not indices

Thus,

etc.).

at the

(natural signs) in our classification, but all those signs

which do

not have as attributes the intentional acts (specifically understood

by Husserl, and to be discussed


and,

consequently,

meaning.

in the chapter

on meaning),

Thus, Husserl's typology

should

not be confused with the contention that indices should

fall

outside the analysis of proper signs, since the extension of his


indices (Anzeichen) covers

both the indices and

except the verbal signs. This

is

all

proper signs

the fundamental error of the entire

conception, the error resulting from the separation of the analysis

of signs from the communication process. Whoever sees that

whoever

connection,
in the

understands that every sign

communication process and

loses

its

is

outside the context of that process, that every sign

or an event

somehow

interpreted

included

function as a sign
is

by someone, must

a thing
reject

as

basically erroneous the conception that only certain signs are

expressive,

i.e.,

trary, all signs

express a thought, have a meaning.

On

the con-

have meanings, express thoughts, and are signs

communicompany of language

only in so far as they perform those functions. In the


cation process,

all

signs appear in the

thinking, or even simply as a specific translation of such (in con-

Selected Problems of Semantics

170

formity with an established code). This

is

so because

man

is

unable to think otherwise than by means of verbal signs in some


form, and

all

other forms of signs are derivative,

i.e.,

The Anzeichen theory overlooks

place verbal signs.

they

re-

that fact,

because the signs are detached from the social context of the

communication process and treated as something abstract and


possessing inherent existence. Thus, Husserl not only separates

from the Ausdriicke, but even opposes the latter to the former. And yet in fact these two categories are intimately connected one with the other, and not only are they not in
the Anzeichen

opposition, but they appear in combination;

what Husserl called the Anzeichen


called Ausdriicke.
to

is

"infused" with what he

Consequently, a typology which takes them

be in opposition

is

based on erroneous

initial

assumptions.

All the proper signs (and hence a considerable part of Husserl's

Anzeichen) have meanings. In this sense, they also ex-

press something,

namely the thought which

is

contained in the

meaning of the given sign. And if the word "to express" is interpreted otherwise, namely associated with information concerning emotional, and not intellectual, experiences, then the indices
(natural signs) can also express something
tears express sadness, a blush expresses

in

this

sense

(e.g.,

shame or embarrassment,

etc.)

The bee communicates somehow with other bees through its


the stag which by his cry and by the use of his antlers
urges the herd to flee, also achieves some sort of communication
"dance"

with the herd,


process

for

Man sometimes
instance

on the roadway and


this last is

communication
hand he stops a pedestrian

acts similarly in the

when with
'points to

his

a speeding motorcar.

And

yet

something quite different from the situational communi-

cation of animals. That difference exists simply because, behind

movements and gestures made both by me and by


whom I communicate (let us suppose I am abroad
in a country the local language of which I do not know), there
is a definite content, which we both translate for ourselves (often

the ordinary

the person with

.j

should say that'

The Sign: Analysis and Typology

into

171

the

same gesture of the hand into


and we both realize that we
way. If, for some reason, such transladiffer from one another (e.g., because

different languages)

verbal signs: "Beware, danger!",

understand such in this


tions into verbal signs
different gestures of the

head are used in Europe and in a large


and dissent), then communication

part of Asia to express assent

simply will not take place. This


to signals, symbols,

The same

self-evident

when

it

comes

applies to the so-called natural signs, the specific

nature of which also


gory.

is

etc.

When an

makes us

set

them apart as a separate

avalanche comes down,

it

cate-

accompanied by

is

the noise of falling stones. But that noise becomes the sign of
the fact that an avalanche

is

coming down, and thereby a

to run away, only if there are people


that natural

phenomenon. In

itself,

it

who
is

signal

properly interpret

nothing more than a

vibration of the air due to natural causes. That vibration becomes

a sign (an index)

men understand

when

it

is

preceived by

men and when

of what the noise they hear

is

those

a manifestation.

The simple fact of men's existence does


not suffice for a phenomenon to be interpreted and thus to become
a sign. Those men must have some knowledge concerning that
This

is

very important.

phenomenon, must understand it in order to be able to interpret


correctly. When a Roman saw a man with a brand on his forehead, he understood perfectly well that he had to do with a slave.
We should be rather inclined to interpret the mark on the forehead as a scar left by some accident. If we do not know the meaning of a red spot on the forehead of some Hindu women, we cannot interpret that sign. For a Bushman from Equatorial Africa,
it

who

has no appropriate special

knowledge,

frozen water

is

merely a hard object with certain definite properties that can be


perceived by touch, vision,

necessary

fall

etc.,

but not a sign (an index) of the

of temperature.

Thus, no natural phenomenon

situation

is

different

is

inherently a sign (an in-

anything "in itself". The


when such a phenomenon occurs within

dex) and consequently does not

mean

Selected Problems of Semantics

172

the communication process. But

how

does

it

happen

that a phe-

nomenon becomes included in that process? By our experience,


by our custom. When we become famihar with a given natural
phenomenon and
to perceive

that

is,

it

as

causal or structural regularities,

its

if it

we

created as a sign. In such cases

it is

Nature which

"partner" in

us.

That

specific

our

anthropologization of natural

events blurs the distinction between the index and the

is

the communication process. Nature "communicates"

something to
sign.

begin

were evoked for communication purposes,

artificial

natural phenomenon, without in any degree changing

character, begins to function for us, in the context of our

its

process of communication

know

(it

being assumed that

we have come

to

the regularities governing the given event), in the capacity

is ascribed a meanan adjunct to a natural event, something added


to it in the process of communication, and appears only within
the framework of that process. From the point of view of natural
processes, the function of an index is something secondary and
is always related to a definite cognitive process, and further

of a sign;

ing.

But

begins to express something and

it

all this is

communication process. In that function an index,


every form of sign, is derivative with respect to communi-

to a definite
like

cation by words

an index

it is

so in the sense that the act of understanding

in the last analysis, always based

is,

and by means of verbal

Thus
light,

all

they

somehow

replace verbal signs and,

translation often takes

we always

Gushing

in

terms

signs except the verbal signs shine with reflected

are always translated into a language of

because

on thinking

signs.

in his

when

interpreted,

words (although that

on an abbreviated form). This is so


means of a language of words 5.

think by

work Manual Concepts

says that there exist primitive

peoples which owing to a greatly developed language of gestures also have


issue,

and also the complicated problem

who have

not been learned a special language

a special "manual thinking". That


of thinking of those deaf-mutes
of gestures,

is left

apart, since

it

does not invalidate the general thesis about

thinking in terms of language of words and about the necessity to translate

The Sign: Analysis and Typology

This

is

precisely the point

which

have in mind when

purpose of

signs, as they serve the

all

173

say that

human communication,

"imbued" with a language of words, and thereby with the

are

meaning that

is

Not

specific to that language.

signs express

all

same way; on the contrary, even these general


remarks reveal an essential difference between direct and indi-

thoughts

rect

in

the

substitutive

expression of thoughts, or else a difference

between the communicative functioning of the indices on the


one hand and of the proper signs on the other. But
are expressive in a sense,

and must be so

all

the signs

they are to be signs

if

at all.

Thus Husserl's
untenable.

into Anzeichen

division

not only because

It fails

all

and Ausdriicke

meaning, as demonstrated above, but also because


in a sense indicate something. This

when he

all

signs

admitted by Husserl himself

says that even the words indicate something, that they

are Anzeichen of thoughts. It


it

is

is

signs have in a sense a

is difficult

to agree with that, since

might suggest that thoughts can originate and exist independently

of verbal language,

and the words are only ex post

as their indices, Anzeichen.


cific,

If our standpoint

is

organic unity of thinking and language, then

selected

that of a spe-

we must

reject

such an idea as speculative, and standing in contradiction to

all

what psychology and physiology say about thought processes.


But in suggesting that conception, Husserl at the same time
destroyed the foundations of his own typology. The wrong and
inconsistent division (which applies to both elements distinguished

by

it)

makes

his

typology unacceptable in

3.

view.

DEFINITION OF THE SIGN

All this (the critical, negative analysis


to

my

was made with a view

drawing certain positive conclusions) shows clearly that the

substitutive signs into that language of

common form

of the

words

in the

communication process.

normal and the most

174

Selected Problems of Semantics

controversy with Husserl is not at ail focused around formal


and terminological issues (although these are involved as well).
Of paramount interest is: What do we understand by the sign
and in what context can its nature be understood and a proper
classification

be effected of the different variations of the signs?

Every attempt to

a typology of signs usually begins

offer

with a definition of the sign. Otherwise

would be

it

introduce a hierarchy and a typology of signs. Hence,

difficult to
I

too begin

with such a definition, but with certain reservations which


strict

my

plans.

As has been stated above (without any deeper


tion, since we still lack elements necessary for that
the sign forms a whole which

such as

re-

is

justifica-

purpose),

analysed into parts and aspects

the material and the semantic aspect

by mental abfrom the structure of this book that


the problems of meaning must be discussed after the analysis of
the typology of signs, since without the analysis it would be
straction only. It follows

impossible to undertake a proper study of such problems. Unfortunately,

and

this is the usual difficulty

with

all

attempts to offer

a systematic exposition of the subject, the lack of an analysis

of meaning adds, in turn, something of a


of the

sign.

It

is

essential,

it

is

touch

upon

in so far

indispensable for further considerations.

Moreover,
fine

that

most general way,

the definition of the sign only in a


as

difl&culty to the analysis

therefore,

indicated

above,

in the

shall,

myself to proper,

i.e.,

definition to

artificial,

signs,

be proposed, conand that for reasons

which recommend a separate treatment of

the analysis of indices.

In tackling the problem from the point of view of the com-

mimication process, that

is

a process which

social in nature, I always take as

my

language of which the sign in question

is

par excellence

starting point a definite


is

an element, and

try

and its functions within


beyond doubt that every sign, as an element
of some language (words, gestures, code, etc.), must be a signify-

to understand the nature of that element

the whole.

Hence

it is

The Sign: Analysis and Typology

ing sign, that


presses

is

a sign which, either directly or indirectly, ex-

some thought. But

ing of the sign

is

176

as a

there

is

also this

the explicit function-

means of communication and /or

the purpose

of communication. Understanding of that entails far-reaching

consequences for the interpretation of the sign and for

de-

its

finition.

The use of the

sign in the

communication process (language)

leads to at least a double relativization of the sign:


in that process not just as

an

it

object, a state of things, or

appears

an event

(which under definite circumstances function as signs), but as


a relation.

By saying

that the sign

is

in definite

use
it

it

is

and complicated

is

a relation we abbreviate

which appears as a

sign,

relations with those persons

who

the following statement: the object,

as a sign; with the reahty

etc.,

which

it

denotes or with which

connected by the sign-relation in some other way; with

makes up a linguistic system and only


it become comprehensible, etc.
Thus the sign is related to people who communicate in a definite,
socially conditioned way, and to the object. That double relation
other signs with which

in the context of

(and not, as

is

it

which does

customary, the referring of the sign to the object

and yet

alone) leads to a consequence which seems trivial


great significance for a correct analysis of the sign

function of the sign

is

is

of

the principal

to communicate something to someone,

someone about something. That function is undoubtedly


common to all categories of signs, and consequently serves as

to inform

the foundation of the definition of the sign Every material object,


:

or the property of such object, or a material event, becomes a sign

when

in the

communication process

it

serves, within the

framework

of a language adopted by the persons who communicate, the purpose


of conveying certain thoughts concerning

reality, that is

concern-

ing the external world, or concerning inner experiences (emotional,


aesthetic, volitional, etc.)

of any of the parties

to the

communication

process.

Such a

definition of the sign

only one (though, in

my

is

very general and deals with

opinion, the focal) aspect of the problem.

176

Selected Problems of Semantics

But

it

undoubtedly grasps that property already mentioned

common

and

before

property

the

to

of

variations

the

all

of informing about

something,

communi-

of

cating something. Another merit of that definition


that

it

can serve as the starting point for an

signs

is

the fact

effort to classify signs,

to suggest a typology of signs.

4.

GENERAL FOUNDATIONS OF THE TYPOLOGY OF SIGNS

The task of the typology of


traits

signs

to delineate the specific

is

of each separate variation of signs against their

common

background, and to estabhsh connections between them and,


possibly, a hierarchy of types of signs.

As mentioned above,

the great variety of signs has given rise

not only to a rich terminology serving to denote types of signs,


but also to a considerable arbitrariness in the use of that terminology. There would be nothing
differences

were

conventions. In

connected

many

cases

wrong

it is

in that if terminological

with

different

in fact so,

and then

solely

to understand deeply a convention and

its

its

essential

foundations.

been well known since the time of Plato that there


nexus between the sound of a word and

appropriate
it is

It

has

no natural
meaning, and that we
is

are free to change our terminology whenever necessary.

But

two things to be taken into account in this connection should


restrict
it

any too far-reaching arbitrariness in those matters.

should be borne in mind that terminological differences

veil the

semantic differences revealed

e.g.,

in

different

First,

may
clas-

and that goes beyond the limits of


conventions. Secondly, there should not, without good and sufficient reasons, be any violation of the actual use of words and
sifications

of phenomena

the actual meanings of words, since this gives rise to additional

comphcations and adds to confusion, especially

in the case of

words with a long-established tradition in the ordinary language.

The

fact that various authors base themselves

on

different

foundations in classifying signs leads to typologies based on

The Sign: Analysis and Typology

various principles of division.

Variety

is

177

when

greater

still

comes to terminology. There would be no point

it

in engaging

in a criticism of the various systems; analysis of the

problem

would thereby become clumsy and obscure, and so

is

it

better

to drop all such criticism.

some general

Certainly,

found in the

literature

For

situations.

of classification to

principles

may

of the subject

instance, the division into natural

tional signs, as referred to above,

is

be

be useful in certain

and conven-

of great importance. Mor-

division of signs according to their extension (indexical,

ris's

characterizing

same

and

universal)

may

The

also turn out to be useful.

applies to Carnap's distinction between the sign as an act

(sign-event)

and the sign

as

an inscription

goes back to an old conception of Peirce


genstein

concerning

the

distinction

(sign-design),

and

between the

token and the sign as a type^. But, since

it

is

which

also of Witt-

sign

as

not possible to

agree with Husserl's division into indices and expressions,

it

is

on the
relation the signs and their referents (indices, iconic signs, symbols); with Morris's typology, distinguishing only signals and
symbols and deliberately giving those terms meanings different
from the current ones; with Biihler's, who ascribes specific meanalso not possible to agree with Peirce's typology based

ing to terms and distinguishes signs, indices and symbols; with


the typology of S. Langer (natural signs, artificial signs, symbols),

or with that of S. Stebbing (expressive signs, suggestive signs,


substitutive signs), etc.

may

objection
vision

is

not uniform,

the classification
etc.

Thus,

stimuli

all

but

With

reference to

those concepts, the

all

be validly raised that either the principle of

is

not exhaustive, or that

such classifications
I

may be

it is

written signs, e.g., "cat"

evidently arbitrary,

used as specific mental

would adopt none of them

When we have

di-

or that the extensions overlap, or that

as

my

own.

and "CAT", each of them

is (1)

a distinct, individual inscription (token), and (2) an individual variation of

an inscription of one and the sanae

type.

178

Selected Problems of Semantics

begin with two distinctions

one of them universally

ac-

cepted, the other controversial.


First, the signs are divided, as already

(indices,

symptoms) and proper

mentioned, into natural

(or artificial).

Second, the proper signs are divided into verbal signs (and

and

written substitutes for such)

only in a sense), this

makes with

other signs. In a sense (but

all

an analogy to the

is

reference to the Ausdriicke.

The

similarity consists in

the fact that a clearly distinct character of the verbal signs as op-

posed to

all

other signs

fact that all the rest

is

is

recognized

distinction Husserl

and the

difference

in

the

not squeezed into the Procrustean bed of

The verbal signs and their specific nature will later


be subjected to a more detailed analysis, precisely on account

the Anzeichen.

of that specific nature of theirs. For the time being,

only repeat

what has already been said: because of the special role of the
phonic language and of the verbal signs in the process of human
thinking and communication, these signs occupy the special,
supreme place in the hierarchy of signs.

And now

for the classification graph:


Signs

Natural signs

Artificial, or

proper, signs

indices

Verbal signs

Proper signs
with a derivative expression
Fig.

Thus we have obtained the following result on the one hand,


we have set apart the natural signs (indices) and have opposed
them by the artificial, or proper, signs; on the other hand, among
the latter we have ascribed a special status to the verbal signs
:

as the basis of the process of

again opposed them by

all

human communication and have

other

artificial signs.

In view of the

The Sign: Analysis and Typology

179

reservation that an analysis of the verbal signs will be

made

and importance of the


(proper) signs must be clas-

separately because of the specific nature

problem, the remaining


sified

and analysed.

It is

necessary to reahze

and what

details

how

far such

supposed to cover.

an analysis

is

to reach,

to be borne in

It is

mind

Peirce succeeded in distinguishing over 60

By basing

classes of signs.
it

it is

for example,

that,

artificial

on

the classifications

different criteria,

possible to multiply the classes distinguished in one's ty-

is

pology, a process which in

but in others will boil


present instance I
large,

some

down

am

to

may

cases

mere

be quite interesting,

scholastic casuistry. In the

concerned above

all

with distinguishing

comprehensive classes of signs and in explaining the sense

of the terms relating to them.


It is

sign), is

known

in

that every sign which

The majority of such

signs are at the

in a definite sense of the

indices

are consciously

purposes. This

not an index (a natural

is

a definite sense of the word

is

word

also.

an

artificial

sign.

same time conventional,


The signs which are not

men

produced by

for

communication

achieved either by a natural similarity of certain

what they are to signify, or a


conventional conferring of definite meanings on objects (states
of things, etc.) which lack such a similarity. When we speak of
objects (states of things, etc.) to

the conventional character of the artificial signs


social interpretation of conventionality
ality

special,

appears in the cases of the various categories of such signs,

to a certain extent even in those cases

of something in the capacity of a sign


to the referent, as in the case of
etc.,

we mean a

and such a convention-

is

where the functioning


based on a similarity

maps, hieroglyphs, pictograms,

probably with the exception of simple similes of objects,

such as photographs. Thus the point

is

not that such a convention

be agreed upon simply by the persons communicating hie

et

an agreement, social
and not individual in nature, be concluded at any time in a conscious manner. Artificial signs may be called into existence on the
nunc (although this

is

possible), or that such

180

Selected Problems of Semantics

of a conscious and

strength

at a specified date

owe

(cf.

all

agreement conclud

deliberate

may

the codes), but they

equally well

their origin to the historical practice of the social process

of communication (the classical example here being the phonic


language). In the latter case, of decisive importance

which in turn

action

co-operation that action


;

although

naturally,

is

a definite

is

mainly from the need of many-sided

arises

socially accepted

without

any

and

traces

is

being continued

any

of

deUberate

convention.

Thus

the proper signs are

all

The

ventional as well.
is

artificial,

enabhng

factor

human

connected with their function in the

munication

the
one

action, in

and
their

in principle con-

further

division

process of com-

of direct influence upon human


and the function of standing for certain

function

case,

By the funcmean this that

objects, states of things, or events, in the other.

tion of standing for (substituting) something I

the sign appears instead of

and evokes

some

human mind

in the

object, state of things, or event,

ideas,

images and thoughts which

are usually evoked by that object, state of things, or event (that


substitution also

reflected in

is

human

action, although

it

does

not in the least follow that the appearance of the sign should
always entail the same results as does the appearance of the object,
etc.,

It

for

which the sign

stands).

might be objected that the proper signs always are signs

for something, that they are

every sign

human
is

produced for the purpose

and hence for the

of communication,
influencing

artificial,

behaviour.

It

a sign of something, that

it

and as such performs the function of


true. Nevertheless, there are

consists in directly influencing

(substitutive signs)

of

is

but

AU

that

is

human behaviour
consists

(signals),

in

and

substitution

and the influence of which on human behav-

indirect.

classification,

points "beyond itself",


substitution.

proper signs the function of which

proper signs the function of which


iour

purpose of somehow

might also be contended that

In adopting that difference as the criterion

we may

further subdivide the proper signs with

The Sign: Analysis and Typology

181

a derivative expression into signals and substitutive signs, and


the latter in turn into substitutive signs sensu stricto

Here

is

and symbols.

the graph:
Proper signs
with a denvatii^e expression

Signals

Substitutive Signs
sensu stncto
Fig.

We now

Sumbols

proceed to discuss the various categories appearing

in the classification suggested above.

A. Signals

me

In the definition of the signal, the starting point for


the ordinary sense of the word.

Of

course, that

word

is

is

used in

the literature of the subject in other, sometimes quite arbitrary

meanings also
is

(e.g.,

a signal). But

that the existing

for Morris, every sign

which

have mentioned, too, that

is

not a symbol

adopt the directive

meanings of words are not to be violated

if

not necessary, just because such a violation of linguistic usage

would do more harm than good, although the using of words


in new meanings, consciously designed in view of some needs,
is

not only admissible in science, but actually often resorted


I

am

in agreement with the ordinary usage

usual meaning of the


it

word

We

desist from,

usually

some

is

to evoke, to change, or to

make

action.

do not say (although Morris

that the freezing of water

and with the

"signal" in so far as I understand by

a sign the purpose of which

someone

to.

is

a signal of a

tells

fall

us to do so)

of temperature,

13

182

Selected Problems of Semantics

or that wrinkles

On the
and in

on a person's

other hand,

my

it is

face are a signal of his ageing.

quite in accord with our language intuition,

opinion also with the natural premises of the

classifica-

tion of signs, to say that the firing of a blue rocket was for soldiers

a signal for an attack; that the appearance of the green light


at a street crossing

way

is

a signal for pedestrians to cross the road-

that the wailing of a siren in wartime

is

a signal for civihans

to seek refuge in shelters against an imminent air raid; that the

wailing of a siren of an ambulance car or a fire engine


for all vehicles to leave a free passage

or a whistle blown in a factory

is

is

a signal

that a bell rung in a school

a signal to break or to start

work.

What
have in

is

the point in

common

that

all

makes

a single category, despite

What do

they

combine them

into

such and similar cases?

all

it

possible to

the differences of details?

As already stated, in all such cases we have to do with signs


main objective of which is to evoke, change, or make someone
desist from, some action. Thus, these are typical signs for some-

the

thing, signs clearly

some

intended to evoke (or to change, or to stop)

definite action as the objective

of communication. Thus

they are material phenomena caused especially or utilized


to

in

order

bring about a response, prearranged and agreed upon, whether

socially (in a group),

manifestations
It is

of

or individually, in the form of definite

human

activity.

only the element of convention which calls for an ex-

planation of the latter assertion.

Signals

appear only where

an appropriate group of people have concluded an explicit agreement by virtue of which a given phenomenon functions for

them

as a signal.

Thus,

if

the soldiers in trenches are to understand the ap-

pearance of a blue rocket in the sky as a signal for attack, they

must be so informed

in advance. In such a case,

a case, does the firing of the rocket

mean

and only

in such

for every soldier the

order: "Forward! Attack!" In this case the blue rocket, in con-

formity with an agreed code, replaces the appropriate verbal

The Sign: Analysis and Typology

183

signs. Without precisely such an agreement the blue rocket


would not be a sign (signal) and would be an ordinary physical

phenomenon, lacking the assigned property of conveying informaFor a third party the blue rocket is not a signal at all (since

tion.

know

he does not

the convention agreed upon), and for the

soldiers themselves a red rocket

the convention previously agreed


different.

On

would be no signal, because


upon provided for something

the other hand, an accidental firing of a blue rocket

by a third party, even one unaware of the consequences of the


act, would be understood as a signal, in conformity with the
prior agreement.

The same applies to the lights at street crossings, to the


movements of the hands of a traffic policeman, to sirens sounded
in peacetime and in wartime, etc. In all such cases we have to
do with an explicit agreement pertaining to the meaning of given
physical phenomena which are usually produced by some individuals in order to set up an appropriate activity by others. This
is shown by the fact that we learn the meaning of the appropriate
phenomena (the rules of crossing streets in cities, the meaning
of light signals, etc.) and that without learning them we do not
understand them.

further point

mean an

that whenever

is

we

refer to signals

object (state of things or event) which

is

we always

utilized

or

produced occasionally, especially for the purpose of bringing


about the given action
is

or

to begin; the siren


is

(e.g.,

is

the rocket

is

fired

sounded when the

when

air raid is

over; the lights at a street crossing are changed

the attack

imminent

when

the

pedestrians are to be allowed to cross or told to stop; certain

verbal signs are uttered or written down, yet they do not appear
in their

fundamental semantic function but independently of

that function play the role of a signal, previously agreed upon,


to start or
etc.,

which

abandon some
exists

signal, that is

know

action, etc.); or

we mean an

object,

permanently but functions occasionally as a

when approached and observed by persons who

a given convention

(e.g.,

road signs for vehicles, devices

184

Selected Problems of Semantics

which

with, photocells

or other signals to

at a specified time activate light,

warn of danger,

Thus the signal can be distinguished from other


signs by the following characteristics: (1) its meaning
by

arbitrary, established

group of people;

its

(2)

its

appearance

The

is

fact that

it

occasional

in

con-

replaces a certain

not to be mentioned as a characteristic, since

this holds, as stated above, for all

The

always

is

of a convention valid within a given


purpose is always to evoke (or change

nection with the intended action.


is

artificial

virtue

or stop) a certain action; (3)

verbal statement

sound

etc.)

analysis thus

made shows

proper signs.

that the signal

for the corresponding verbal expressions;

it

is

a substitute

replaces

them

as

The metaphorical statement that every signal is "imbued" with the phonic language
and its meaning, is thus, in the case of signals, explained directly and very simply. The signal has, hke every meaningful group
every code replaces a phonic language.

of verbal signs, a meaning, although

it

has such in a different

manner, in an indirect and derivative way.

What

then

the relation of the "signal" in this sense to

is

the "signal" in e.g. the Pavlovian sense? Let us get

we have

that

but different meanings.

shape,
"signal"

and

it

quite clear

here to do with two words which have the same

refers

reactions,

to

relations

In Pavlov's terminology,

concerning, physiological

by being part of a certain

the

stimuh

situation, treated as

a whole, which brings about definite conditioned responses. Thus


in that case the "signal"

means the same

ulus in a definite sense of the term.


to

its

No

as a physiological stim-

reference

is

made

there

being "imbued" with meaning and to connections with

a phonic language moreover, the very formulation of the problem


;

eliminates the need to resort to such concepts. I shall not stop


for

an appraisal of the correctness and virtue of such an ap-

proach, but simply confine myself to the statement that in Pavlov's

terminology the notion of the "signal"

differs

complete-

ly

from that analysed above. For the time being, that statement

is

sufficient for

our purposes.

The Sign: Analysis and Typology

185

B. Substitutive signs

ive

The second large class of proper signs comprises substitutsigns. By contrast with signals, they are signs of something,
emphasized function of substituting, representing

signs with the

some other objects, states of things or events.


The class of substitutive signs is in turn subdivided

two

whether the object a signs stands for

subclasses according to
is

into

a concrete, material object, or whether the sign,

an abstract notion,

represents

thing

material,

which

is

ships,

material properties,

i.e.,

some-

sometliing

i.e.

connected with the material world, material relation-

object. In the first case

etc.,

but in

itself is

we speak of

not a material

substitutive

and in the second, of symbols.


The problem of substitutive signs sensu

signs

sensu

stricto,

tively simple.

What

for other objects

is

by

stricto is

compara-

involved are material objects which stand

virtue of similarity or convention. Typical

examples of substitutive signs working on the similarity principle


(iconic

signs)

are

all

kinds of images and similes (drawings,

photographs, sculptures,

paintings,

stitutive signs

etc.)

based on convention are

and examples of sub-

all

kinds of written signs

which stand for speech sounds, their groups, words, sentences,


etc.

Of course,

the division

is

not a rigid one and there are transi-

tion stages between the various types,

which

regard here in order to avoid complication.


issue here

is

that of the

mechanism

dehberately dis-

The only

serious

of that "substitution" or

"representation" of one object by another from the point of

view of the mental processes; the problem of meaning

is

in-

volved, but that will be discussed in the next chapter.

on the other hand, when it comes to


symbols. This is due mainly to two causes: first, the issue is very
controversial and approached from different angles in the very
Difficulties

arise,

rich literature of the subject; second, the issue at stake here is

one of a

class (or subclass)

of signs which play an exceptionally

important role in the various

fields

of social

fife.

186

Selected Problems of Semantics

my

In

system, the symbols are a subclass

and

signs

are characterized principally

characteristics:

material objects represent abstract notions;

(1)

the representation

(2)

known

representation

by a

is

based on a convention which must

a given symbol

if

of substitutive

by the following three

be understood;

to

is

based on the representation of an abstract notion

is

a representation which outwardly appeals

sign,

be

conventional

(3)

to

senses

(and semantically works by exemplification, allegory, metaphor,


allusion to mythology, the pars pro toto principle, etc).

In such a formulation, which seems highly significant for


various reasons, the starting point again

meaning of the term "symbol" which

is

is

in

the living intuitive

agreement with the

directive that existing linguistic usage should

not be violated

and that, so to speak, the semantic entia should


not be multiplied beyond what is necessary.
For what, in conformity with the ordinary intuitive usage,
unless necessary,

is

it

that

called a

is

of Christianity, a crescent
of the Mosaic
star

five-point

woman

we

rehgion; that the

a symbol of

is

are certainly in agreement

say that the cross

is

a symbol

of Islam, and a six-point star

Nazism, and an axe with fasces


a

We

"symbol"?

with such an intuitive usage when

hammer and sickle or a red


Communism, a swastika of

of

Fascism; that a figure of

with a band across her eyes and with a pair of scales

and a sword

in her

hands

is

a symbol of justice, that the figure

of Mars symbolizes war and heroism, that of Eros, love, and


a skeleton with a scythe, death; that black symbolizes mourning,
purple

dignity, yellow

that specific colours

on a

envy, white

flag

symbohze

innocence,
(one's) nation

red

love;

and mother-

it is

doubtful

whether, for example, mathematical and logical signs

may be
may be).

land,

etc.

In the light of existing hnguistic usage,

considered symbols (although

And

there

is

it is

no doubt that the

often said that they

firing

of a rocket to start an

at-

tack in a battle, a person's photograph, or the fact that water

has frozen,

etc.,

cannot be said to be symbols.

The Sign: Analysis and Typology

187

Let us analyse those facts and statements from the point


of view of the definition of symbol
It is

beyond doubt that a symbol

have suggested.
is

always a substitutive

beyond doubt that a symbol,

thing. It is also

some-

consists precisely in standing for

sign, that its function

is

like every sign,

something material. Further elements, just those which have

symbols from

to distinguish the

first

the other substitutive signs,

comments, since they are debatable.

require certain

The

all

issue is as to

what

is

represented by a symbol.

low in that respect many authors who are

I fol-

justified in protesting

broad an interpretation of the notion of "symbol",


and against a practical identification of the symbol with the sign
against too

in general.

My

the symbol

is

assertion

is

that the characteristic property of

by a material object (function-

the representation

ing as a sign) of an "ideal object" or, strictly speaking, an abstract


notion.

Let us revert to the examples adduced above: this or that


religion or faith

courage,

justice,

nationahty, etc.
this is

is

why

hterature,

an abstract concept; the same holds for the

Communism

social systems,

love,

or Fascism; for such notions as


innocence,

envy,

they are favoured in


etc.

mourning,

dignity,

consists

all

and
mass movements, propaganda

The most profound sense of the symbols

precisely in that they bring abstract

them abstract ideas in the


form which is easier to be
grasped by the mind and to be preserved in memory. Consequently, as mentioned previously, such material symbols are
very convenient for mass movements, since they substitutively
concepts nearer to

men by

presenting to

shape of material objects, that

is

in a

convey certain ideas which tend otherwise to be

difficult to

grasp

and to understand; as the material symbol becomes independent


of what it represents, as its specific "ahenation" develops, a
mythologization of such a symbol may also occur. A symbol
is

not a purely intellectual product, although

nected with a concept.

It is

it

is

closely con-

equally closely connected with various

188

Selected Problems of Semantics

may

emotional states and therefore

not only bring

concepts closer to men, but also prevent

know

abstract

men from coming

to

The multifarious functions of symbols, and in


myth-making functions, make them an extremely

the truth.

particular their

interesting object of study.

The conventional character of the symbol


a social and historical convention)
trait

is

linked

with the

that

second

of representing

function

notions by material objects, which

(in the sense

is its

is

of

specific

abstract

the proper function of

the symbol in the communication process.

To understand any

the

sjrmbol,

appropriate

convention

must be known. Those who do not know the Old and the New
Testament, Greek and Roman mythology, who do not know
from very childhood) the symboUsm of colours as used
in Europe, who are not versed in our political life and in the
symbohsm of the various national emblems and colours
(usually

will

not understand a single symbol of those exemplified above. By

analogy, a European, even an educated European not acquainted

with Oriental culture, will

fail

to

understand the symbolism

of the Hindu dances, or that connected with Oriental


the specific

symboUsm of

colours, smells, etc. This

is

deities,

so because

no symbol has a natural meaning, on the contrary, eyery symbol


has an
This

is

artificial,

conventional sense which must be learned.

shown by such elementary examples

cultural circle
in the East

it is

it is

black which

is

as that in the

European

the colour of mourning, whereas

white that in our cultural circle


;

it is

purple which

is

the symbolic colour of power and dignity, whereas in China

it

is

yellow; and so on, not to mention the visual symbols of

wisdom, courage,
if

compared

virtue, etc.,

which have nothing

in

common

as between the various cultural circles.

All this explains the third specific trait of the symbol

the

sensory (usually iconic) representation of an abstract idea.

The

fact that the object represented

an abstract notion was indicated as the

by a symbol
first

such

is

trait.

always
It

has

been said that in the case of the symbol we have to do with the

The Sign: Analysis and Typology

an abstract notion by a material and concrete

substitution of

but no explanation has been given as to that material

object,

concreteness of the symbol. This

Like every sign, the symbol


enon.

189

Were

not so,

it

it

stand for anything else.

is

is

to be dealt with later.

a material object or phenom-

could not be perceived and could not

But a symbol

as a rule, not merely

is,

a material object, but also a visual image.

The

painter, the black

and white

artist

or the sculptor

who

wants to use his art to present some abstract concept, such as


heroism, virtue, love or patriotism, must usually resort to a symbol,
in

which he has to choose between the

picture will be

or

it

will

alternatives: either his

an allegory conveying the abstract idea in question,

be an exemplification conveying what

the intermediary of

what

is

is

general through

particular.

The symboHc representation of abstract ideas


resorts to metaphors on which the picture is based
Verbal symbolic images include

in literature).

frequently
(especially

such

formula-

bowl of hfe", "the cup of bitterness", etc.


Mythology is amply drawn on. The serpent of Aesculapius
the symbol of medical art comes from classical mythology,
does Hercules, the symbol of strength, the owl, the symbol

tions as "the

as

as

of wisdom,

And

etc.

the animal symbols in state

emblems

(the

come from ethnic legends.


Representation based on the pars pro toto principle also
frequently occurs. The cross as a particularly important element
hon, the eagle,

in the story
|f.

Of

image

usually

of Jesus has thus become the symbol of Christendom.

course,
is

etc.)

it

also

happens not infrequently that a symbolic

abstract in character,

the abstract notion

sometimes

some

it

and

imaginary

combinations of colours
flags),

interpretations

links with

(e.g.,

(e.g., as

association

symbols of emotions),

as national

symbols in the case

abstract drawings with mythical or magical

(e.g.,

signs (the graphic

its

through

explanation

crops up). This holds for colours

of national

in such a case

represents are purely conventional (although

the

swastika),

symbols of

mathematical and logical

infinity,

negation, etc.)

J90

Selected Problems of Semantics

Gesticulatory, olfactory, etc., symbols are in principle purel]

This applies for instance to the symbolism ofj

conventional.
gestures in

Hindu dances,

spread in the East,

Sound and

the symbolism of

aromas so wide-

etc.

combinations, too, can play the role of

their

symbolic images by virtue of a convention, most commonly


in connection with a certain emotional tone, felt only in a definite cultural circle.

For

with a low tone

ing of bells

and monotonous ringperceived by us a symbol of

instance, a slow
is

i|

mourning; the same holds, by analogy, for the melody and the

rhythm of a

certain type of march.

way of illustration, and not as any exhaustive enumeration. The gamut of symbols is extremely wide
and its possibilities almost unlimited. But in all the cases we
All this

is

only by

have analysed, the factors referred to

symbol

is

(represents)

(which

is

earlier are involved: the

a material object, state of things, or event;

the function

such a representation

of the substitutive sign sensu


is

not

symbol,

difficult to notice that

organically

stricto);

when based on

possible only

conventions (usage, ad hoc agreement,


It is

replaces

it

an abstract notion, and not any other material object


definite

etc.)

such an interpretation of the

connected with a definite conception of

a typology of signs, and with a criterion of division, adopted


for

that

intuition
sions.

purpose,

and

This

is

fact

is

conformity

in

adjusted
is

to

the

ordinary
sense

of

language
expres-

by no means unimportant or secondary,

since the linguistic categories shaped


classification

with

existing

by history

offer a certain

of concepts, based on social appreciation of sim-

and differences in content, which means that they pertain


what such categories express as regards relations in the
objective world. Names can of course be changed arbitrarily,

ilarities

to

but

if this

leads to changes in the classification of

without adequate semantic reasons, that

cannot permit. The point

is

is

phenomena

something which science

that the terminology adopted should

help us to inform others about the real world to which

it

pertains,

The Sign: Analysis and Typology

191

land about the relationships in that world, and not hinder us in

doing so and consequently in communicating with others. Nothing

more

forcefully

which

debunks the conventionalist deviation

shadow over correct underlanguage


and reality than
standing of the relation between
does semantics properly pursued. It is precisely from the position
in the last score or so years has cast a

of semantics that one has to oppose the distortions, so frequent

problem

in the literature of the subject, in the interpretation of the

of symbol, the arbitrary classifications of signs, and the result-

names

ing arbitrary changes in meanings of the terms used as


for the various categories of signs.
tells

us that terminology

is

semantics above

It is

all

which

arbitrary in the sense that there

is

no inherent connection between the phonic sign and the object


denoted by it (although there are different opinions on that
too), but there is no arbitrariness in establishing a correspondence between the verbal signs and objective reality when
issue,

it
it

comes to the

and phenomena, when

classification of things

comes to the
As already mentioned,

of language.

cognitive function

typology and

Husserl's

Cassirer's

theory of symbols, in particular, have spread in the literature of


the subject an interpretation of the term "symbol" which stands
in contradiction to the existing sense of that word,
to the analysis

of that category as performed above. Usually

the class of symbols

is

made

to

include

and sometimes (Morris) symbols are

signs the

all

the verbal signs,

identified with signs

are not signals, or the sense of the term


all

and thereby

is

which

extended so as to cover

meanings of which are not based on a

with the object represented (Kotarbinska), or else

all

similarity

the signs

which are deliberately used as such are considered to be symbols


(Stebbing), etc. These are dangerous practices not only because

they obscure, as

we

shall see later, the highly

character of the verbal signs, but above

all

important

specific

because they pass

over in silence the existence of an important group of substitutive

signs

with marked

common

characteristics.

192

Selected Problems of Semantics

The

significance (especially the social significance) of symbols

understood in a proper,
all

i.e.,

way

restrictive,

is

enormous, above

in view of their role in shaping public opinion

scholars as

and

social

why we must recognize the achievements of such


Cassirer who succeeded in noticing and spot-hghting

myths. That

is

the problem, in spite of having mystified

it.

proper solution of

an open issue. Marxism is the trend potenmost suited to undertake that difficult but promising task.
So far however the matter remains in the air.
The understanding of the nature and function of symbols*!
depends above all on the criterion adopted for the classification of

that problem

is still

tially

signs in general. It does not, of course, settle the question, but

an error there invahdates a proper analysis of the problem


symbols. It is not possible to understand what a symbol is
Husserl

is

Ausdrilcke,

iff*

followed in dividing the signs into Anzeichen and

which practice

virtually

eliminates

of understanding the specific nature of

the

ij

off!

possibihty

i!

'j

occurring in the

all signs

process of

human communication. The same happens

tarbinska

followed in adopting the division (suggested by

is

authors) into iconic signs and symbols. Such a division


possible,

but that

dichotomic

correct, does not lead to

classification,

any interesting

division in those classes of signs

is

if

Ko-

many

certainly

though formally

results

and blurs the

which are observable

in practice

and highly important from the point of view of the communication


process.

This

is

so because the division loses completely the

distinction, for example,

sense specified above)

between the

and the verbal

signal, the
sign.

symbol

(in the

The same objection

may be

raised with respect to the classifications suggested by

Morris,

S.

Stebbing, S. Langer, and to the resulting meanings

of the term "symbol". This criticism once more reminds us


justified is the statement that terminological issues are

as arbitrary as

might appear

how

by no means

at the first glance. It also

reminds

us the importance of a proper starting point for analysis, and

of a proper criterion of the typology of

One more remark

signs.

to conclude these considerations.

The Sign: Analysis and Typology

As already
sifications

of

or categories,

indicated, there

and the

signs,

may

no doubt that some other classome other category

is

distinction of

in

my

opinion,

there will be

may

no

ever, to devote

analysis of

classification which,

most important. Within that

the

is

We

prove advantageous for certain purposes.

have paused here to discuss the principal


further divisions

193

classification,

be effected, based on other

collision

between the two.

criteria,

should

like,

and
how-

a few remarks to a group of important signs the

which may involve certain

of the typology adopted in this book.

in the light

difficulties

mean

on
movements such as gestures, facial expressions, expresmovements of the body, etc. Do not those signs express
I

the signs based

physical
sive

above

all

a uniform group, in view of

human body and


psychic states? Is

connection with the

human

consequently with
it

its

experiences

not a group of signs which

important in view of the fact that

and

particularly

is

always accompanies the

it

phonic language (gestures and facial expressions), and does

it

not

deserve therefore to be separated as a distinct class or subclass?


First
sification

the

of

all

it

must be said

that in the hght of the clas-

we have adopted, based on

communication process,

bodily signs belong to

all

this

the functions of signs in

not a uniform group. These

is

the classes

we have

distinguished:

thus, for instance, tears and laughter, being natural

that

accompany

indices; gestures

speech

may

certain spiritual processes,

and

facial expressions

may be

phenomena
classed as

which accompany human

also be treated as indices of certain emotional states

hand or of some other part


of the body may be a signal; a wink may be a substitutive sign
sensu stricto; an appropriate posture and movement of a part
of the body during a dance may be a symbol; and finally certain
movements of hands or fingers performed in conformity with
a certain code may just be a translation from the phonic language.
The apparent uniformity disappears, and the distribution of the
a conventional

movement of

the

signs in question over the categories

presents

no

difficulty.

Thus there

adopted in our

is

no cause

classification

for alarm.

194

Selected Problems of Semantics

So

we have been

far

discussing the various categories of

signs, excluding verbal signs.


will

As

previously indicated, our analysis

be concluded by the study of the most important problem

in the theory of signs,

namely that of verbal

signs

and

their

specific nature.

THE SPECIFIC NATURE OF VERBAL SIGNS

5.

The

significance of the phonic language,

and consequently

of verbal signs, for the communication process, and thereby for


social

and

life,

its

And

has long been appreciated.

investigation.

deliberate call for such investigation

in the Upanishads,

and the study of language

of Plato's dialogue Cratilus.

From

we

It is

came
is

main

subject

human

attention.

And

human co-operation, and


we must focus our attention

role in

organic link with that process,

precisely

it

its

the

importance of the communi-

realize the social

cation process, in particular

is

found

is

the times of antiquity to this

day, the problem has not ceased to claim


justly so, for if

its

the phonic language

verbal signs have for long been an object of study and

on the phonic language.

possible to engage in the various speculations as to which

first:

the phonic language or the language of gestures;

possible to discuss the role

categories of signs;

it

is

and

significance of the various

from others on the

possible to differ

point as to whether there are various systems of "languages",


etc.

But

it is

not reasonably possible to deny that in

civihzations the phonic language

was and

is,

all

the

known

not only the principal

means of human communication, but also the means without


in science, culture and technology would be

which progress
impossible.

All this sounds like a truism

and a

triviality that

dispensed with in semantic analysis. But


so,

especially in semantic

analysis.

For

it

it

is

might be

only seemingly

follows from

that

truism, or apparent truism, that the position of the phonic language

The Sign: Analysis and Typology

in social processes is so exceptional that

unless

we

with

cannot be explained

recognize the specific nature of the phonic language

and the verbal signs as compared with


i.e.,

it

195

all

other "languages". If

it is

all

other systems of signs,

true that all other systems

of signs are "imbued" with meaning taken over from the verbal

and that they shine with reflected light, then the division into
the (phonic) language and "languages", drawing attention to the
signs,

inequality of these categories,

contemporary
raise

semiotics,

is

also justified.

semiologies,

etc.

And yet

the various

which

after

important issues concerning the geiieral theory of signs

all

do not cancel outright, precisely that specific nature


of the phonic language and the verbal signs. (This might be
explained by the attempt to bring out what is common to all
blur, if they

and thus to formulate a general theory of signs.)


The theoretically erroneous tendency to disregard the specific

signs

nature of the verbal signs can also be observed in the termi-

nology that implies a definite classification and characterization


of the various classes of signs. True, that terminology is not yet
well established, but

subject

verbal sign.

For

all

on the

especially in the latest literature

the term "symbol"

is

most often used

to denote the

the definitional conventions, this fact obscures

the difference between the verbal

and other

signs

and simply

adds to the confusion. In some cases the term used is just "the
sign", which is certainly correct, but by being too general does
if

we bear

conceals

concep-

not help to give precision to the problem (especially


in

mind

tions

that in these cases the terminology

which certainly do not favour any explanation of the

specific

nature of the verbal signs).

Let us begin with the statement that ahhough the phonic


language consists of signs in the sense defined above, so that a
verbal sign (and this is a trivial statement) always falls under
the general definition of the sign, yet the verbal sign

neither

and nothing more, nor a symbol in the ordinary


word and in the sense suggested ad hoc by the various
In the latter case, we have to do not only with an arbitrary

just a sign

sense of the
theories.

is

196

Selected Problems of Semantics

change in the terminology adopted, but also (since otherwise

would be

and devoid of any

scienti-

meaning) with a different definition of the symbol, a

defini-

the operation
fic

childishly naive

tion which again obscures the specific nature of the verbal signs.

Thus, the verbal sign complies perfectly well with the general
definition of the sign as

adopted above, which definition, by

recognizing the communication process as the foundation of

our analysis, sees in the communicative function of the sign


its

and consequently relates the sign both


which it communicates something and to

principal property,

to the object about

the language in which that something

is

communicated.

no doubt that the phonic language is a definite,


specific system of signs. But here the issue is opened with the
question: a system of what signs?
There

is

Can any

the subject? Certainly, yes.


all

the

tives

way

on
Such information concerns above

positive information be

in

found in the

literature

which a verbal sign means something. Representa-

of the most diverse trends seem to reveal fairly

agreement of opinion in that matter;

mean

common

here the issue of

"transparency to meaning" of the verbal signs.


If

we

take authors

who

independently of one another have

written about the specific nature of the verbal signs

Rubinshtein,

Urban and Ossowski^

we

Delacroix,

find that each of

them

uses the expression "transparency to meaning" precisely with


reference to the verbal signs,

property of those signs

is

and for each of those authors

that

associated with their specific nature.

In resorting to that metaphor, those authors have in mind that

when we
proper

perceive verbal signs, then, by contrast with aU other

signs,

we do not

perceive their material shape as some-

H. Delacroix, Le language et la pensee, Paris 1924; C. JI. Py6HHOcHoebi oOufeu ncuxoAoauu [The Principles of General PsycholoChap. XI: "PeMb" [Speech], MocKsa 1946; W. M. Urban, Language
7

uuTeflH,
gy],

and Reality, London 1951 S. Ossowski, "Analiza poj?cia znaku" [An Analysis
1926,
of the Concept of Sign], reprinted from Przeglqd Filozoficzny,
Nos. 1 & 2.
;

The Sign: Analysis and Typology

197

thing autonomous, but just the contrary: that shape

to such

is

an extent confused with meaning that except for the cases of disturbed perception we do not reahze the existence of the material
aspect for the verbal sign.

an

sential; this is

As

to

who has

the priority right to

"transparency to meaning"

the formulation

issue that

that

not es-

is

of interest rather for an historian

is

of the problem. But the fact that the otherwise diverging opinions are in agreement

the implication that

tion in the literature

When we

on that point has

we

implication

its

at least

face a conception which has a firm posi-

on the

speak of

subject.

signs,

substitutive

signals,

and

signs

we always speak of material things or events which


purpose of human communication, because precisely

symbols,
serve the

by virtue of the

fact that the

communicating

parties understand

them in one and the same way, each of such signs informs the
communicating

parties.

Each of these

signs

is

in a relation to

the object (understood in the broadest sense of the term, so


as to cover things, their properties
events, psychic processes, etc.)

and to a
a sign,

definite

i.e.,

and

between them,

relations

about which

it

language within which

conveys information,
functions

it

only as

means something. But when we speak about the

proper signs with the exception of the verbal signs, then

it

is

always a fact that the relation between the material and the
semantic aspect of the sign admits of a certain "autonomy" of

meaning;

meaning

this
is

signifies

that

except

for the

iconic

signs

always shaped independently of a given sign

sense of the material sign- vehicle), as

it

(in the

were outside that

sign,

and consequently may be combined with another material shape of


the sign (e.g., we might change the convention on the road signs
without impairing their meaning, that is, what they communicate
to us). This

is

connected with the fact that

all

those signs function

only within a phonic language, and with thinking in terms of ideas,

which

must
signs

is

specific to that language. This is

have "ready-made" meanings for

which are not verbal

signs.

This

is

why we can
all

the

so because

and

categories

we

of

are simply

14

198

Selected Problems of Semantics

unable to think otherwise than by means of verbal


all

and

signs,

sign-making (except for the use of a phonic language, incul-

cated in us by our being brought up in society, or, in other words,


except for thinking in terms of language signs)

is

a secondary

process, a result of the various conventions (in the broad, historical sense of the term),

and

it is

and as such

in this sense that

And what

it

about verbal

Here the situation

is

is

always preceded by thought,

always

signs,

is

"imbued" with thought.

and language

quite different

first

of

as their system?
all,

because they

have nothing "behind them", they are not based on meanings


of some other language. This

form a

so because thought and language

whole. There

single, indivisible, organic

that exists separately


is

is

and language that

only thinking-and-language. There

separately,

and

that

sign

concept-and-verbal-sign.

exists

Of

is

no thinking

exists separately, there

no concept

is

separately,

that exists
is

only

who

think

there

course, there are people

that not only can one think without resorting to verbal signs,
to language, but that

There have been

from Plato up

it

many

is

just such thinking

believers in

which

is

"true".

"true", "direct" cognition,

to the phenomenologists, intuitionists, adherents

of the idea of a mystic union between the cognitive subject

and the object of cognition,


But these

etc.

It is

they

the Marxist position, these

by

sadly

repeat

mehr".

are irrationahstic speculations, denied by such disci-

plines as psychology, the physiology of

so

who

Seele, spricht die Seele nicht

after Schiller: "Spricht die

Stalin, in his

From

opinions were ridiculed and rightly

Marxism and

It is just this specific

the brain, etc.

the

Problems of Linguistics.

unity of thinking-and-language which

meaning" of the verbal signs.


They are meaning, although they are not only meaning. A passage from Die deutsche Ideologic, quoted previously, refers
to consciousness, the real form of which, according to Marx,
is speech. Thus the verbal sign is not mere meaning. It is also
a sound, the material phenomenon consisting in the vibration
of air waves, without which there would be no sign and no comgives rise to the "transparency to

The Sign: Analysis and Typology

munication

why

this is

ing",

if

we

reject

199

"pure" and "direct" cognition.

although

it

expresses an idea that

is

valuable for the under-

standing of the specific nature of the verbal signs,


since

leaves unsolved the real

it

And

the metaphorical formulation "transparency to mean-

is insufficient,

problem of the relationship

between the phonic aspect (the sound picture) and the meaning
(the conceptual content)
It is

esting

of the verbal sign.

in this connection that de Saussure has

made an

inter-

comparison between the verbal sign and the sheet of paper

one side of which

is

the sound

image and the other, the con-

ceptual content (de Saussure used the terms signifiant and signifie,

being of the opinion that a language sign

is

a psychic entity

composed of a sound image and a concept). He argues further


that one side of a sheet of paper cannot be eliminated without

destroying the other, and in the

same way

in a verbal sign the

sound cannot be separated from the concept (meaning). This


is a metaphor, too, but it somehow sheds light on the problem
from a

To

different angle.

explain the specific nature of the verbal sign,

we must

engage in greater detail in the analysis of the concept "the verbal


sign", since

it

appears in two different meanings, combined with

different theoretical conceptions. In

one formulation "the verbal

means the sound (in the sense of acoustic vibration or


a sound image) with which a definite meaning is somehow con-

sign"

nected. In the other, "the verbal sign"

is

the specific whole con-

sisting

of sound-and-meaning, characteristic of really existing

entities

of the phonic language.

There

is

no doubt that verbal

signs consist not only of meanings

but also of the phonic aspect of meanings, that "curse" which

form of vibrations of the air persecutes the spirit. There


no doubt and this is not challenged probably even
by the most radical of those who contend that meaning and
sound in a verbal sign form an organic unity that that unity
is not absolute, but relative, which means that it can be broken
in the
is

also

in certain specified cases.

This refers not only to such

trivial

200

Selected Problems of Semantics

cases as the perception of alien, incomprehensible speech merely


as a sequence of sounds. It also covers

more complicated and more


by Head and others,

interesting cases, such as those investigated

consisting in aphasia

when a person

preserves the

memory

of the

phonic form of words but ceases to understand their meanings,


or loses the faculty of speech while understanding what

Both

to him.

and

said

is

in the case of perception of incomprehensible speech

in the cases of aphasia

we have

to

do with a

certain

anomaly

from the point of view of the process of human communication.


That process presupposes the community of speech of the communicating persons as well as a normal state of mind in those

And it is here,

persons.

in view of the relative character of the unity

of sound and meaning in the verbal sign, that the fundamental

problem

arises:

how

is

its

links with the sign (this

although

it

is,

unfortunately, unavoidable

breaks the planned exposition of problems), and to

engage in the examination of certain issues pertaining to the nature


of those links in the case of the verbal signs.

Two

competing attitudes are to be found in

One of them

is

that of the associationists

sound and meaning


their

exist independently

combining in the verbal sign

between a

definite

sound and a

is

this connection.

who

maintain that

of one another and that

based on the association

definite

ready meaning. Such

was the opinion of Delacroix, who held that everything ultimately


boils down to an association in human memory between sound
and meaning, an association which is arbitrary in nature 8. The
same underlying associationist assumption was also observable
in the case of Russell at the time

when

ducted a discussion on "meaning" 9.

the periodical

Mind

similar position

con-

was

oc-

cupied by Sapir in his study of language, when in classifying


verbal signs as symbols he defined them as sounds automatically

Delacroix, op.

Mind, 1920, No. 116,

cit.,

!'

that unity brought about? I have in this

connection to anticipate the results of the analysis of meaning

and

'

p. 365.
p. 398.

201

The Sign: Analysis and Typology

associated with

meaningly.

variation

of the associationist

theory assumes the form of the conception of specific association,

who

defended by Martinak

in that respect followed Hofler.

theory consists in asserting that association

takes place through the intermediary of other judgements,


is

not a mechanical repetition of the relation,

perceived (judiziose Association).


tionist

Be

that as

it

That

indirect, that

is

as

it

and

originally

may, the associa-

conception reduces the problem to the mnemonic asso-

sound with a ready meaning, which consequently must have been somehow ontogenetically shaped outside
the language and independently of it. The verbal sign is here
treated in the same way as any other sign with relation to which
ciation of a ready

meaning

is,

as

we have

seen,

"autonomous",

i.e.,

shaped outside

precisely the Achilles' heel of the

and independently of it. This is


whole concept, which not only runs contrary to our sense of
the specific nature of the verbal signs and even the simplest analysis
(it

of them, but moreover presupposes the object of controversy

assumes that the verbal signs do not

by the nature of

their links

differ

from other

signs

with meaning). Consequently, the

strongest argument against the associationist conception, which


at

one time was described by K.

Biililer

endeavour to

as

"simply

naive"

its

adherents do not even

justify their assumptions,

even though those as-

(geradezu naiv), hes in the fact that

sumptions run contrary to our intuition and to the results of


analysis of linguistic entities.

The other

attitude

can be characterized by the interpretation

of the relative unity of sound and meaning in the verbal signs


as a connection sui generis, different

from that which

is

specific

that

the meaning
and marked precisely by this
is not "autonomous", that it can neither be shaped
nor appear outside of that unity which is language-and-thinking,
word-and-idea. The only argument, and a very naive one, is that
to other signs

of a verbal sign

we

usually learn foreign languages

10

E. Sapir, Language,

New York

by looking for appropriate

1921, p. 10,

202

Selected Problems of Semantics

sounds to be associated with ready meanings; but that argument


can very easily be refuted. It is indeed a fact, but always on the
basis of some known language by means of which we think and
from which we translate into that foreign language. We have

learned a foreign language only

when we
is, when

start to think in

when we

cease to translate,

terms of that foreign language, that

the links between the sounds of that language with

meanings cease to be for us something "external", based


on ad hoc associations, and become organic and direct, and the
verbal signs become "transparent" to meaning. Thus, there
is a difference between the study of a foreign language, and the

their

knowledge of a language already well learned. Our analysis


pertains, of course, to the latter case only.

What

is

meant by the connection between sound and meaning

in the verbal sign being sui

generis that

is

another question.'

One is fully justified in demanding an explanation of these matters.


Whoever advances a thesis in science, must prove it. Yet I must
make it clear that even if one is unable to give a satisfactory
answer to questions pertaing to a theoretical
if

such an answer

is

controversial (which

is

thesis,

or even

most often the

case,

and the humanities), this fact


the thesis in question. I mention that

especially in the social sciences

does not of
here,

itself refute

because the thesis concerning a sui generis

connection

between sound and meaning in the verbal sign, although supported by formidable arguments,

still

lacks a consistent explana-

and mechanism of that connection. The only


such an explanation which, in my opinion, deserves

tion of the nature

attempt to offer
attention,

namely Pavlov's hypothesis of a second system of

signals, will

There

is

be discussed below.

one point more to be mentioned here and now:

the arbitrary nature of the connection between sound and meaning.

This thesis was submitted by de Saussureii

Whitney.
11

F.

It

who

referred

to

must be emphasized, however, that while de Saussure

de Saussure,

Corns de linguistique generale, Paris 1949,

p.

99.

w
The Sign: Analysis and Typology

the

stressed

signifie (in

found already in Plato's

to be

of the connection between the


which he repeated the old idea,

nature

arbitrary

and the

signifiant

Cratilus, that the

are not connected in any natural

community

given language
free,

but

way with

same time

denote), he insisted at the

203

sounds of words

words

the objects these

that with respect to the

the choice of the signifiant

is

not

conditioned socially.

is

obvious that Marxists, in defending the thesis of the


organic unity of sound and meaning in the verbal signs, must
It

is

sharply

the

criticize

conception

between sound and meaning

of the

(in the sense

arbitrary

of

its

connection

being established

by virtue of a convention) which leads to those conventionalist


oddities which are to be found in at least part of the neo-posi-

must

tivist literature. It

adherents

among

only

also be

added that that conception had

formahstically-minded

and understand the

failed to notice

of language and of verbal signs.

historical

logicians

and

who

social character

on the other hand, always

It has,

been opposed, for fundamental reasons, not only by Marxists,


but by linguists as well, regardless of what trend they represent.

For a

linguist

ample,

S.

L.

Rubinshtein
as

apphed

historical
its

such argumentation as that advanced by, for exRubinshtein

criticizes

to language signs
analysis,

is

quite natural

from the point of view of their

cognitioni2.

is

genetic,

and demonstrates that the verbal sign has

"social" hfe independent of us,

conventions, and

and comprehensible.

the conception of an arbitrary convention

its

history independent of our

connected with the objective nature of our

similar argumentation

linguist Zvegintsevi3. In particular, he

is

resorted to by the Soviet

emphasizes that his protest

against the conception of the arbitrary nature of the link between

sound and meaning


by him for the

12
13

is

thesis

by no means to be interpreted as support

which claims that there

is

some natural

PyGHHinTeHH, op. cit., pp. 406-407.


B. A. SBcrHHueB, FIpodneMa SHaKoeocmu asbiKa [The Problem of the

Sign-nature of Language], MocicBa 1956.

Selected Problems of Semantics

204

connection between the verbal sign and the object it designates.


It is very important to bear these two issues in mind.

verbal signs

a link sui generis

is

conception of such signs:


sign,

and meaning in the


combined with a different

that the link between sound

The view

is

not the sound alone wliich

it is

with some autonomous "meaning" as

its

is

the

"partner", but

the indivisible whole, consisting of sound and meaning, under-

stood as a signifying material object (acoustic vibration). This


is

the only consistent interpretation of the verbal sign as a sui

generis

hnk of

the unity of sound and meaning, as a sui generis

sign characterized by "transparency" to meaning. That "trans-

parency" can appear


terial,

if,

less

if,

we

cease to perceive the ma-

of the sign as something independent,

physical shape

with which a no

and only

independent meaning

is

combined

in this

or that way. "Transparency" to meaning, so characteristic of

when we

verbal signs, appears precisely

cease completely to per-

ceive the material shape of the sign (except for cases of disturbance

in the

of

its

normal process of communication) and are conscious only


semantic aspect.

Thus, the verbal sign

is

not any symbol, although the use

of that term with reference to verbal signs

is

now, as referred

to above, almost universal in the literature of the subjecti4.

The

verbal signs should not be confused with that specific and very
useful
I

subclass of substitutive

have called symbols

in

signs

conformity

already

with

analysed,

intuition, because the verbal sign reveals features of which


1^

which

current linguistic

symbols

Let us take note of the characteristic protest by de Saussure who in


(in a certain specific interpretation of that issue)

defending the arbitrary

connection between sound and meaning in verbal signs comes out against
calling the latter symbols, because

symbols characterize some natural link

between the sign and the object. The statement

is

very controversial, but

the intention to introduce terminological distinctions between the various


categories of signs

Urban

also

is

is

most laudable.

against including the verbal signs in a general category

of symbols. His argumentation

is: if

the verbal signs were just symbols,

would be the sense of talking about a symbolic use of language?

what

The Sign: Analysis and Typology

206

term are deprived, and

in the proper sense of the

is

in turn

deprived of certain properties that are characteristic of symbols

And

understood in conformity with current usage.


to the term "symbol"

we

fashion, than

an ad hoc meaning, treated

either restrict the class

we impart

if

an arbitrary
of symbols to verbal
in

term

signs and, violating the linguistic usage, cease to apply the

"symbols" to those signs which so far have been called

we conventionally extend the


to cover with

erate the specific nature


I

want to avoid.
The verbal sign

For

all

or

sense of the term "symbol" so as

and thus

for example, all non-iconic signs

it,

so,

of the verbal signs

which

is

oblit-

just

what

not a signal, nor yet a signal of signals.

is

the respect for those suggestions (since thus far they are

nothing else but very general hypotheses and suggestions for

which are contained in the Pavlovian theory of the

research)

second system of signals (in which theory the verbal sign


a signal of signals),
tions

it is

is

just

important not to ignore the contradic-

and the dangers of vulgarization that are inherent in that


and fetishiza-

theory. In this field, too, as in any other, worship

No

tion of opinions have detrimental consequences.


that Pavlov
it

is

was a

brilliant scientist.

a far cry to canonizing

no one

certainly

is

precisely in the

is

bound

views and hypotheses, and

all his

entitled to identify Pavlov's opinions with

dialectical materialism (as


it

one denies

But from that statement

some authors have

name of

done).

Anyhow,

dialectical materialism that

one

is

to protest against the theory of the verbal sign as a signal

of signals

at

least in that

form

in

which that theory

is

known

at present.

The contradiction
from

his

many

consists in that Pavlov,

as can be seen

statements published in Pavlovian

Wednesdays,

realized perfectly well the specific nature of the verbal sign

and

phonic language, and yet the traditional formulation of the


concept of the second system of signals and of the verbal sign
as a signal

of

the

of signals

verbal

sign.

is

Hence

denial

the

of

danger

that

of

specific

nature

vulgarization

206
of

Selected Problems of Semantics

problem

the

(similar

the

to

vulgarization

characterist

of behaviourism).

The primary reservation must be made against the very ten


"signal". As v^e remember, a signal in the process of human
communication

is

verbal statement

the process of

a conventional sign which replaces a certain

and

is

comprehensible only as such. Thus, in

human communication,

the signal

a semantic product, has a meaning and


tic

qualitatively

after all

par excellence
linguis-

not be said of the stimuli that

These phenomena are


and therefore the terminology, which
to be understood in some metaphorical sense, is not

produce conditioned

the

may

meaning, which of course

is

"imbued" with

is

is

reflexes in animals.

different,

most fortunate, since

it

contains the danger of serious misun-

derstandings. Yet one has to admit that, after


pline the researcher

is

all,

in every disci-

adopt an arbitrary terminology, on

free to

the condition that the sense of the terms involved

is

sufficiently

precise.

But no such argument as that can be used

meaning

in a conventional

animal and the

human

same

reflexes is the

in

is

man and

of conditioned reflexes?

field

attitude

It

also

is

not

is

Now

man

and of the

my

both to the

the point

socially.

we encounter problems

But the mechanism

business to analyse and

But in so

and consequently

always conditioned

implies that

in the animal.

of communication processes,

towards them.

nected with

the term "signal"

world. The mechanism of non-conditioned

criticize physiological theories.

on the

if

identically referred

far as they encroach


I

is

his

have to adopt some


that everything con-

conditioned reflexes

This general formulation

of

human communication,

role of processes that entail consciousness

not think, etc.

And

and are

man

can-

the conditioned reflexes that develop in

man

inseparably connected with language, without which

as a result of certain external influences are not usually formed


outside the sphere of his consciousness, but within that sphere.

This

is

why we have

qualitatively

to

from the

do here with a phenomenon that

reflexes in the animal.

Whoever

differs

fails

to

207

The Sign: Analysis and Typology

notice that,

and seeks to reduce the processes taking place in


to mere physiological stimuli and reac-

human organism

the

tions,

as

is

the case with the behaviourists, simply vulgarizes

the problem, since

phenomena

is

all

naturalism in the interpretation of social

a vulgarization.

Matters become

more comphcated when we reach the


The verbal
be a signal of what signals? Those referred
still

higher level, namely that of the "signal of signals".


sign

would have to

to in the experiments with dogs? If so, then this

is

a misunder-

standing, because except for simple physiological stimuH

the stimuU produced by food), such "signals" just


in

man's social

life.

Of some

other signals?

(e.g.,

do not appear

Then which? Undif-

ferentiated terminology prevents us from grasping the differences,

and experiments could be informative only if they were treated


not from the naturahstic, but from the social point of view,
that is not by eliminating those processes which entail con-

by treating such as organic elements of human


behaviour, including human reflexes. But experiments of that
kind have not been made. The hypothesis of a second system of
signals suggests that they should be made, but so far it has not
sciousness, but

beyond a suggestion.
Thus, at best we do not know what kind of signals is meant.
In a worse situation, we do know, and then we may say that in
principle there are no such "signals" in human social processes.
What then is that signal of signals? The same signal, only on
got

a higher level, which


ing

human

is

a reduction of social processes involv-

consciousness to a two-level system of physiological

stimuU and reactions? That would be an extreme vulgarization


of the problem in the spirit of mechanistic materialism, with which

an adherent of

dialectical

materiahsm could on no condition

For as psychology may not be reduced to physiology,


in the same way the issues of epistemology and semantics may
not be reduced to pure physiology. And if what is meant is something else, a "signal" in some other meaning, then what is that
agree.

meaning to be? This we

just

do not know, and

it

is

better to

208

Selected Problems of Semantics

we have to do with an incomplete hypothesis. But


we adopt the most favourable interpretation, if we virtually
conclude that we do not fully grasp what is being said in that
agree that

even

if

hypothesis, there remains a terminology which

standings and with a possible vulgarization. This

now

say the

to

is,

ambiguous and which threatens us with serious misunder-

least,

to conclude that the verbal sign

is

is

why we have

neither a signal,

nor

a signal of signals, and that before these matters are clarified


experimentally and theoretically,

we must

abandon

rather

that

terminology.

Thus the verbal sign

known

of the other

a strictly defined meaning


that has

its

ovm

at all

names have

a sign sui generis, a sign

it is

specific nature.

But just because of


the verbal sign

not a symbol, not a signal, not any

is

categories of signs, if their

may

its specific

nature, as a result of which

not be identified with any other sign,

some of those

take over the functions of at least

it

signs.

can

The

is not a signal, since it has different features and


from them, but it can function as a signal. The verbal
sign in the same sense is not identical with a symbol, but it may
assume its role. Examples can be multiplied, if a more detailed

verbal signs
properties

classification

of signs

special significance

is

and

made. This
role

is

one more proof of the

which the verbal sign has in the

communication process.

The second
is

first (i.e.,

of sound and meaning in the verbal


in

the

process

From
is

in fact

with the sui generis unity

sign), is its function

and

role

of abstraction.

the genetic point of view, the verbal sign, like every

other sign,

There

which

characteristic of the verbal sign,

linked organically with the

is

evidently a product

nothing strange in

this.

of the abstraction process.

All cognition and

always work on the selection principle. This

is

all

perception

connected with

the specific requirements of action, which without such selection

would not be

possible.

The same phenomenon

in connection with every kind

of sign which,

is

in

observable
functioning

The Sign: Analysis and Typology

within the framework of

209

human communication,

the general regularities governing

subject to

is

cognition. Every sign

all

is

2i product of the abstraction process and at the same time an impor-

tant instrument of that process. There


line

which determines a qualitative

verbal signs and


sign

may

other kinds of sign.

all

formulate

simphfy,

form, be pars pro

is

toto,

etc.,

here, however, a border-

between the

differentiation

The point

something

is

an

in

that every

abbreviated

and may thus be an important

instrument of the abstraction process, but in

all

these cases

it

always remains connected with definite sensory data, with a definite

mental image;

this refers also to

what are known

as generic

representations. Thinking in terms of ideas requires a different

instrument, which
specific

property,

is

the verbal sign, precisely because of

"transparency to meaning", which makes

its
it

possible to rise to the highest levels of abstraction, inaccessible


to other types of sign,

and

to be separated

form concrete sensory

data to an extent exceeding the possibilities or other kinds of sign.

As

the psychologists

in our

kind
here

mind with

of connection
is

tell us,

the verbal sign, too,

representations. But this

with

is

is

associated

a totally different

sensory image what

is

involved

either the imaginative associations of objects connected

with mental processes and accompanying those processes, or


the associations of the images of written signs, words spoken,
their sounds, etc.

The semantic content of the

verbal sign

is,

however, independent of such associations, which only accompany


that content but are not a condition of
sides they differ as

its

between the individuals

existence (and be-

who

understand a

given sign in the same way). Because of the unity of sound and

meaning, and because of

its

"transparency to meaning", the

verbal sign has special properties of abstraction. Every

word

generahzes, wrote Lenin. Similarly Sapir in his Language says


that cognition

becomes communicable only

if it

is

not

strictly

subsumed under some class of things or


is the work of the word. The same is
maintained by Susanne Langer and others.
individual but can be
events,

which subsumption

210

Selected Problems of Semantics

Of course,

the role of the verbal sign in the abstraction process

would require

special researches

graph. In this place


is

quite sufficient for

And

finally,

and a comprehensive mono-

only draw attention to the problem, which

my

purpose.

the third characteristic of the verbal sign:

its

from the point of view of a precise communication between men. The issue here is so obvious that no lengthy

special properties

analysis

needed.

is

One can communicate, and men


by means

existence of (primitive) tribes that prefer


by

do communicate,

in fact

of different "languages". Anthropologists testify to the

means of

manual "conversation"
As has already been

gestures to verbal conversation.

mentioned, Gushing in this connection speaks of a special

style

of "manual thinking", different from our thinking in terms of


language. All this
at the

is

neither extraordinary nor impossible. But

same time one thing

is

certain: neither the philosophical

system of Hegel, nor Einstein's theory of


a

grammar of

the

simplest

kind for

relativity,

this

or

that

nor even
language

could be formulated by means of such a "language". There


a special power, inherent in the phonic language,

is

which makes

and its rise to highand higher levels of abstraction, which makes it possible to
discover and formulate increasingly wide and profound regularities in the universe, and thereby enables man to become the
master of the world. One may complain about the ambiguity and
possible a further development of thinking
er

imprecision

of ordinary language, but even such a complaint

can only be made by means of that ordinary language and within


its

framework. This is why the phonic language

ularly convenient

and

flexible

process, but also an instrument that


to

not only a partic-

is

exceptionally easy subject

and endowed with almost boundless possiof perfecting itself. And this is one of the most important

improvement

bilities

is

instrument of the communication

aspects of the specific nature of verbal signs.

The Sign: Analysis and Typology

The

undertaken in

21

chapter was concentrated


and the typology of signs.
Yet, especially in the part dealing with verbal signs, the problem
of meaning could not have been eliminated completely. This is
chiefly

analysis

on the

this

issues of the definition

and normal. An analysis of the sign which would


abstract from meaning, and thus from the characteristic function
of the sign, would of necessity be incomplete, and hence it may

quite natural

be admitted only when

justified

by

essential reasons.

reasons emerges from the plan adopted for

my

One of such

work.

It is

now

necessary to complete that analysis with a discussion of the prob-

lems of meaning.

To

that I shall proceed in the next chapter.

Chapter Three

THE MEANINGS OF "MEANING'


"

'When

use a word',

means

ful tone,

'it

nor

'The question

less'.

mean

diiTerent things',

'which

is

to

Humpty Dumpty

what

just

is',

choose

said in a rather scorn-

mean

neither more
you can make words
said Humpty Dumpty,

to

said Alice, 'whether

'The question

be master

it

that's

is',

all' ".

U
(Lewis CarroU: Through the Looking-Glass}

"Men
use,

content themselves with the same words as other people

as if the very

sound necessarily carried the same meaning".


(John Locke)

SAY to someone ignorant of French: "Donnez-moi

peau,

vous

s'il

plait".

understand you."

my
He

His only reaction

repeat the

is

same thing

mon

cha-

do not
English: "Give me

to reply: "I

in

and the same man smiles and passes


I had said.
Two persons want to cross the street when the
hat, please",

me my

hat.

understood the meaning of what

One of

flashes red.

traffic light

the pedestrians stops immediately, while the

other, apparently not familiar with traffic regulations, goes on.

His companion stops him and explains:

"You

see that red light?

no one may cross the road now; when the green


hght comes, then you may cross". At the next crossing the learner
It

means

that

stops of his

own

volition

when he

sees a red

light.

He now

understands the meaning of that sign.

And

so

it

always

sign-situations:

is

when we have

to

do with

signs

and with

such a situation occurs only when the communi-

cating persons understand in the

same way the meaning of the

sign in question.

The problem of meaning emerges


to use a different

and simpler formulation,


[212]

in

sign-situation

in the process

or,

of human

i(

The Meanings of "Meaning"

communication, since that process,

if

we

213

disregard the issue

of telepathy and other forms of alleged "direct" communication,

by means of

consists in trasmitting thoughts, emotions, etc.,

signs;

it is

a process of producing sign-situations.

When

or operate beacon lights

place road signs

speak, write,
crossings,

at street

draw maps or plans, fix "poison" labels on bottles, sew epaulettes on uniforms, hoist signal flags, etc., then in every case I am
using certain signs for communication purposes (even a mental

monologue

is,

as

we know,

each of these cases

in each of these cases the


It

a dialogue in a masked form), and in

produce a sign-situation. Consequently,

problem of meaning emerges.

has already been said that sign and meaning form a unity

is broken into parts or aspects only by abstraction. There


no sign without meaning (since even an authentic sign which
for any reason is not comprehended, e.g., because the language

which

is

in question

is

not understood,

is

nothing more than a material

object or event, a blot of ink, sound, etc.);

and meanings "by

themselves", without a sign-vehicle, exist only in the minds of


incurable

metaphysicians.

properly performed, that

Hence, an analysis of the


is

an

sign,

if

analysis of the sign-situation,

an analysis of communication by means of

signs,

must take

into account the sign as a whole, as a unity the material sign-

vehicle (sound, picture, conventional drawing, etc.)

and meaning.

we have
we

In discussing the definition and the typology of signs,

unity for granted; although occasionally

been taking that

have referred to the semantic aspect of the


referring to

it

as to

best method, yet

it

signs,

we have been

something given and evident. This


is

unavoidable in

all

is

not the

those cases where

we

have to do with a mutual inter-relationship or with a close connection between the aspects or elements of the phenomenon under
is high time now to fill that gap, for otherwise
both the analysis of the sign and consequently the analysis of the

investigation. It

communication process would


problem of meaning

is

suffer.

The more so

since

the

extremely important (in view of the impor-

tance and frequency of sign-situations in social

life,

if

not for

15

214

Selected Problems of Semantics

and moreover it is neither unambiguous no


on the contrary, it is exceptionalKi

other reasons),

universally well understood;

complicated and

difficult.

I shall also set

now

the limits of

those important issues which


I

my

interests

and

indicati

intend to analyse, and those whicl

shall disregard.

'

Ogden and

C. K.

I.

A. Richards once devoted to the problenj

of meaning a monograph entitled The Meaning of Meaning.


set

out to investigate in what senses the term "meaning"

is

Illj

actually

used in the literature on the subject (chiefly philosophical). Theii!

Chapter VIII gives samples of the various meanings of "meaning"'

and shows that the actual usage is marked by a chaos "not dreamedi|
of in our philosophy", although

precisely philosophers whoij

it is

are principally responsible for the situation existing in that respect.

To put an end

to that not entirely edifying logomachy, the

authors resolved to bring out those senses of the term "meaning"!

which are

in actual use. In

strate that there are as

Chapter IX, therefore, they demon-

many

as

6 gi'oups of meanings of that

term,

some of the groups embracing

ings.

In

all,

additional, secondary

mean-

they proved the existence of 23 different meanings

of "meaning", some of them very remote one from another. The


questions arise: Is their

list

of that term appear?

it

Is

exhaustive?

Do

not other meanings

not possible to conceive of yet other

meanings?
These questions are certainly important and interesting from

some points of view.


list

and
to

We

are not, however, interested in a complete

of the possible uses of the term "meaning" in ordinary language


in scientific usage.

Moreover,

draw up such an absolute

term "meaning"

manner

(in

languages).

list

(in

it

would hardly be

possible

view of different uses of the

in the various ethnic languages) or in

any

rigid

view of the incessant developmental processes within

We

of that term

are interested in a certain special group of uses


that

group which

refers

to

the interpretation

of the "meaning" in relation to the function of the sign in the

communication

process,

the

function

which

makes

possible

The Meanings of "Meaning"

215

a passage from the sphere of subjective thinking about something


to the sphere of inter-subjective transmission of such thoughts

so that they are understood by the parties concerned. If

we

say,

most generally and without any claim to precision, that meaning


is simply that by which an ordinary material object, a property
of such an object, or an event becomes a sign,
ing

i.e., that meanan element of the sign-situation or of the communication

is

process, then

we

as

shall see later

ambiguity from that term, yet


restriction

is

we

we

restrict

important for two reasons

do not eliminate
it

first, it

all

considerably. That

absolves us from

laborious semantic analyses exceeding the scope of the issues


in

which we are

more

interested,

strictly the

Thus,

we

are

and secondly,

it

enables us to outline

scope of our positive research.


still

concerned with meanings and not with

a definite meaning of "meaning". But the essential point


in seeking a

maximum

precision of such meanings

is

that

we should

put an end to empty verbalism and empty logomachy, and be


able to repeat after

G. C. Lewis:

"A

Ogden and Richards

verbal discussion

of significance, but

it is

their

may be

motto taken from

significant or devoid

at least well to realize that

it is

verbal".

The ambiguity of the term "meaning", even in that restricted


scope, is explained, first, by the objective many-sidedness of the
sign-situation which includes that function of the sign which we
call

meaning, and secondly, by the diversity of aspects which we,

as analysts of the sign-situation, extract

from that sign-situation

by asking various questions, referring to

real processes, questions

conditioned by the concentration, for practical or theoretical


reasons, of our attention

on

this or that aspect

of those processes.

This co-operation of objective and subjective factors gives

rise

to the ambiguity in question.

1.

My

task, as I

ambiguity

arises,

ON THE SIGN-SITUATION

understand

it,

is

not only to state that such an

and to establish those meanings which actually

216

Selected Problems of Semantics

appear when we use the term "meaning" under certain circumstances, but also to

adopt an attitude towards the various con-

ceptions connected with those different meanings of the term


involved.

And

that requires something

to subject a given

to light

its

to one's

own

more than the abihty

term to semantic analysis and subtly to bring

different senses.

One must

give a precise formulation

views and take a stand not only in the controversies

that divide philosophical schools, but also in the specific issue

involved in the process of

human communication. That

is

so

because general divisions and general standpoints in philosophical

do not automatically produce solutions of

controversies

and intricate problems.


The problem of meaning

is

special

most im-

certainly one of the

portant and philosophically most baffling issues of our times.


Incessant declarations concerning

to

may

importance

its

seem boring, but one more repetition

is

already

nevertheless worth while,

avoid being suspected of neglecting or underestimating the

matter. But then that important issue

ways

either

cessful

may be

used in different

as a subject matter of scientific

and correct or not, but that

is

analysis (suc-

a further question), or as

a springboard for vertiginous metaphysical speculations.


the second alternative
best proved

is

That

not merely a theoretical possibility

is

by Husserl's theory of meaning which, strangely

enough, finds adherents even


istically-minded. This

is

why

among

thinkers

special stress

who

must be

are positivlaid

on the

proper starting point of analysis, since that will determine success

of

Our

failure.

starting point has already

process of

human communication,

in analysing the
in

been defined:

it

is

the real

the real sign-situation. But

problem of meaning one would hke,

especially

view of the considerable number of attempts and solutions

that are clearly metaphysical in character,

more

to

First

what has been said about that


of

all,

to

add sometliing

starting point.

should like to follow Tadeusz Kotarbinski

and to begin by attacking that hnguistic hypostasis which

is

The Meanings of "Meaning"

'

responsible

problem.
[

many

for

"There

is

a sin of metaphysical distortion of the

no

such

thing

meaning"

as

case that formula directed against hypostases


ficial

pedantry (as

is

217

is

in

this

not just a super-

proved by Husserl's theory), but a very useful

recollection.

"Meaning"

which

not to any entity (whether material or ideal) called

refers

is

a typical term used as abbreviation,

men who communicate

meaning, but to

with one another by

using certain objects or events to transmit to one another what


they think about the world around them. This

when one plunges into the


meaning so as to avoid being

worth remem-

is

bering

whirlpool of the problems

of

led astray

by the delusive

name

but nevertheless attractive metaphysical thesis that every


has

its

One

counterpart in an object, an entity, to which

an urge here to dot one's

feels

/'s

refers.

it

and cross one's

?'s

more

as regards the ontological aspect of the issues involved, the

meaning and notion entail some ontoland a fragmentary treatment of those problems

so since the problems of


ogical attitude,

prevents us from discussing


First of

all,

it

all

the issues concerned.

must be explained that when

say "to exist"

or "to be" I understand these words in a materialistic sense.

According to that interpretation,

all

that "exists" or "is" has

a material nature and consequently exists irrespective of any

an external stimulus of our sensory exThus there exist things (material objects) which in

cognizing mind, and


periences.

a broad sense of the

is

word

(covering also such things as fields

of energy) are manifestations of what


of "matter"

(it

is

is

Engels in his Dialectics of Nature). This


of the word "exists", and this
is

known

in

is

name

given the abstract

in this spirit that the

problem
is

is

treated

by

the direct meaning

the proper interpretation of what

mathematical logic as the existential quantifier.


is reduced to two

Thus, the direct meaning of the word "exists"


statements:

(1)

whatever

exists

has

objective

existence,

independent of any cognizing mind; (2) that existence


istence in the material sense, such as
in

is

is

i.e.,

ex-

the attribute of things

the broad understanding of the word.

This interpretation

Selected Problems of Semantics

218

of the direct meaning of the word "exists"

characteristic

is

form of materiahsm, and therefore the corresponding

every

theses of pansomatistic reism are unquestionably materialistic!

But things

make

it

exist

not in an isolated form, but in relations which

possible to speak of a material unity of the world.

also say that there exist connections

that there exist traits

and

relations

and properties of things

We

between things,
that which

(i.e.,

the elements of a given class or set of things and which

marks

all

in the

human mind

finds reflection in

what are termed abstract

notions), that there exist processes or events (certain fragments

of the material world, things,

somehow

change), that there exist

attitudes and actions of those fragments of the material world

which we

call

men,

etc.

In

such cases, when we say "exists"

all

or "is", then the condition of objective existence (in the sense


that

we have

do not with an arbitrary product of the cogniz in some approximation of

to

ing mind, but with cognition

something what

and of

all

mind

in general, although the act of cognition itself

has a objective tinge)


preceding case.

the
etc.,

is

satisfied,

Relations,

processes,

pertain to things, that

we have
but the word

world. Consequently
material world,

an

but in a different manner from

properties,

attitudes,

are not things (as are men, houses, chairs, stones,

ahhough they always

in

of the cognizing mind

independently

occurs

indirect sense (the reists

stitutive abbreviations: instead

to

"exists"

speak

must now be understood

in this

connection of sub-

of saying that things are marked


to things,

simply that there exist properties, relations,


is

to the material

do with some fragment of the

by something, that something happens


in so far as reism

is,

etc.),

etc.,

states,

we

etc.).

not identified with nominalism, the

say

Thus
reistic

standpoint on this issue also coincides with the general materialistic


standpoint.
If

we say

that notions, meanings, etc., exist, then

we simply

speak of their existence in the indirect sense. To ascribe to them


direct existence

would be

(objective) idealism of the first water,

since objective idealism maintains that

some

ideal entities exist

The Meanings of "Meaning"

219

objectively either as the only entities or alongside material enti-

Such a philosophical standpoint is, of course, unacceptable


to a materialist. Hence the importance of a precise distinction

ties.

between the various senses of the word "exists", a word the


ambiguity of which cannot be eliminated in ordinary (natural)

language hence also the importance of the struggle with hyposta;

ses,

It is

the

and

especially in the case of discussions so subtle

as are those pertaining to

also

meaning, notion,

slippery

etc.

worth while now to be precise about the matter of


of the communication process (see

philosophical aspects

125127 above). In particular, I refer here to the reply to the


question, "What does it mean that different men experience
the same states of mind?" We have agreed that one of the conditions of the development of "the same states of mind" is the
reflection of one and the same object by minds that possess the
same structure. What is and what can be that "object", reflected
by the mind and giving rise to "the same experiences"? We
pp.

absolutely reject the existence of certain ideal objects

a con-

ception dear to the hearts of such thinkers as Bolzano, Frege,

Brentano,

Husserl

and

of view, only one solution


relation

of cognition

subjects

is

in the

is

From

others.
is

the

the

which in the

possible: the object

common

point

materialistic

counterpart

of various

the material world, which manifests itself concretely


.

form of things

broad sense of the word). All con-

(in the

nections, relations, properties, attitudes, actions, etc., are objects

of cognition not in the sense of

some independent material or

ideal objects,

but as objectively occurring relations, properties,

attitudes, etc.,

of some fragments of the material world

things.

Thus, they are an objective object of cognition only in so far


as they are attributes of those fragments of the material

which always are objects in the

word

(this is

and the

an analogous
this

and

direct sense of the

issue to "existence"

indirect sense of the

Having made

essential

word

in the direct

as discussed above).

brief explanation

of the words "object" and "exists",

world

of

my

understanding

revert to the

main course

220

Selected Problems of Semantics

of

my

what

analysis.

is

Two

problems must be formulated precisely

a sign-situation where that which

we

call

meaning

ap-

and what do we mean by meaning?


Many references have already been made to the communication process as the foundation of an analysis of sign and mean-*
pears;

we have

ing. Consequently,

referred to the sign-situation, because

whenever we have to do with communication, we likewise have

do with a sign-situation: for we communicate only and exby means of signs. That thesis is correct enough, yet
it conveys but little. To add something on the subject we have
to analyse the sign-situation. For that purpose I should like
to present three conceptions of Ogden and Richards, of Johnson, and of Gardiner. The choice is not accidental in so far as
all these conceptions resort to schemes similar and yet different
and somehow complementary, and as such form steps leading
to what in my opinion is a correct interpretation of the problem.
Ogden and Richards are authors of a now classical book
on the theory of meaning! and undoubtedly have in that reto

clusively

spect considerable achievements to their credit. This fact explains

why

their

diagram of the sign-situation

is

of the fact that the idea

it

classical, in spite

formulated earlier

(cf.

Russell).

Since

considered almost
expresses

am

had been

not interested in

chronological priority, but in a typical standpoint on the issue

we

are analysing, the

work of Ogden and Richards can

well be

taken as the representative of that standpoint.

Ogden and Richards stand

for a causal theory of meaning,

since they claim that between the various elements of the relation of

meaning there

is

a causal nexus.

We

interested not in that aspect of the problem, but

are,

above

however,
all

in the

formulation of the sign-situation (and, consequently, meaning)


as a definite relation.

Ogden and Richards


ments of the sign-situation
1

1953,

C. K.

Ogden

&

I.

distinguish
:

the

following three

ele-

the symbol (which in their terminol-

A. Richards, The Meaning of Meaning, London

;j

The Meanings of "Meaning"

ogy

is

the

same

as the sign), the object (referent),

vening thought (reference). The somewhat


(referent, reference)

definite

and the

artificial

inter-

terminology

explained by their unwilhngness to use

is

existing terms, such as "object", in

with

221

semantic

view of their being burdened

tradition.

THOUGHT OR REFERENCE

SYMBOL

REFERENT

Stands for
(an imputed relation)

TRUE

The

situation

and to the

is

clear.

The authors

refer to

communication

sign-situation, the latter being in their opinion reducible

to the three elements

shown

in the

diagram the symbol symbolizes


:

something and evokes an appropriate thought which refers to


the referent (object). The relation between the symbol and the
referent

is

an

indirect one; a direct relation (along the base of

the triangle) occurs only in those exceptional cases

when we have

between the symbol and the referent (i.e.,


in our terminology, when we have to do with an iconic sign).
It is worth while drawing attention to the vigour with which
to

do with a

the

similarity

authors reject the

suggestion

that

the

sign-situation

is

characterized only by a relation between the symbol and the

Such a standpoint (they quote Baldwin's Thoughts


and Things as an example) they treat as solipsist, and they stress
the necessity of "explicitly recognizing the world beyond us"2
thought.

2 Ibid., p. 20.

222

Selected Problems of Semantics

Yet they remain silent when


comes to men who communicate with one another.
These appear in Johnson's diagram. Johnson's book^ is

in our analysis of the sign-situation.


it

not a work of any signal significance, but

American work concerning the


it

an

as

illustration

and Richards but

because

its

interprets

it

is

a rather typically

social sciences.

have chosen

author adopts the scheme of Ogden


in

an

entirely different way. Here,

too, there are three elements of the sign-situation (Johnson refers


explicitly to the

and

their

communication

arrangement also

manner. Here, then,

is

is

process), but they are different

and that

different,

in a characteristic

Johnson's "triangle of reference":

The Symbol or Reference


B.

A
The Speaker, or SLjmbol User
Fig.

Not only

the

erent) are clearly

The Hearer and Referent

scheme but the terminology

(reference,

ref-

borrowed from Ogden and Richards. John-

son gives the terms an entirely new sense ("reference"

is

used

not as "thought" but as "frame of reference", and "referent"

who is spoken to"),


One may legitimately object

not as "object" but as "the person

yet this

only results in confusion.

to such

terminology as "the symbol or reference" and "the hearer and


referent", but
tal

my

point concerns something else: in the fundamen-

change of the entire sign-situation, in which the relation

"symbol-referent" through thought


"speaker-hearer", then that relation

York

E.

S.

1956.

Johnson.

Theory

is
is

replaced by the relation


that of

and Practice of

the

men who com-

Social

Studies,

New

223

The Meanings of "Meaning"

municate one with another through the intermediary of the sign.


The author somehow hnks the symbol with thought (ahhough
it is

not clear in what that link

silence the referent,

i.e.,

is

the object which

Gardiner's work"* offers a

still

and so avoids a

consequently,

(and,

is

different

picture.

and consistently

represents a consistently realistic

of communication

and passes over


spoken about.

in

to consist),

of the

fetishization of the sign-relation

Gardiner

social theory

sign-situation),

and the meaning-

relation.

Gardiner

is

an eminent
which,

statements

teristic

purpose, are by no means

"As a

linguist.

his charac-

my

exceptional in his work:

approximation,

first

Here are two of

although chosen especially for

let

us define speech as the use,

between man and man, of articulate sound-signs for the communication of their wishes and their views about things. Note
that I do not deny the thought-element in speech, but the emphasis of

which

my

definition does not

wish to

stress

are,

lie

firstly,

on that element. The points


the co-operative character

of speech, and, secondly, the fact that

with things, that

is

it

is

always concerned

to say with the realities both of the external

world and of man's inner experience" 5.

and the things spoken about


factors of normal speech. To these must be

"Thus the speaker, the


are three essential

added the

actual

listener,

words themselves" 6.

Gardiner did not draw any diagrams of the sign-situation,


but one can easily reconstruct such a diagram on the strength
of his statements. But then one will have to resort to some geoother than a triangle, because, according to
Gardiner, the description of the sign-situation requires at least

metrical figure

four elements. First of

all,

there are the

municate one with another, then there

which they communicate, and

is

two men who comthe object concerning

finally the sign, finked

with the

A. Gardiner, The Theory of Speech and Language, Oxford 1951.

Ibid., p.

Ibid., p. 28.

18.

i
224

Selected Problems of Semantics

thought, by means of which communication takes place. If thee


sign

and the thought are treated

elements in question rises to

five.

number of the
The geometrical shape is ir-

separately, the

(it might a trapezoid, or two triangles with a common


The important point is that he grasped the essential
elements and relations occurring in the sign-situation. It is for

relevant here
base, etc.).

that reason that I consider Gardiner's standpoint to be a correct


one.
First of all (and this

is,

as

we

shall later see, extremely im-

portant for the analysis of meaning), an end


cific fetishization of the sign

is

put here to a spe-

and of the sign-situation, so

in the hterature of the subject. In other

common

words: the suggestion

means a relation between signs


or a relation between the sign and the object, between the sign
and thought on the one hand and the object on the other, etc. (such
a suggestion comes from lexical and logical operations and from
partial analyses of "what the word 'x' means", "what 'the red
flag' means", etc., that is, from all those analyses which suggest
that signs have some sort of independent existence). The fact
that the sign-situation develops as a relation between the men
who communicate with one another and "produce" signs for that
purpose, is well understood and explicitly indicated. In disis

rejected that the sign-situation

have

in Capital coined the

term

cussing the struggle against the fetishization of the sign,

borrowed something from Marx who

"fetishism of commodities" in connection with a

very
ing",

problem which

resembles ours. We are seeking the sense of "meanand Marx wanted to explain the sense of "value". In the

much

course of his analysis, he realized that people

who

investigated

on the market succumbed to the


notion that commodities exchanged themselves so to speak by
themselves, so that the relations of economic value were relations
the exchange of commodities

between commodities. To Marx goes the credit of proving that


in fact they are relations

that

is

between producers of commodities,

men, so that they are social relations, social labour

"embodied"

in

is

commodities, and becomes the foundation and

The Meanings of "Meaning"

225

measure of exchange relations and of what we call value. The


discovery of the "fetishism of commodities" was indeed a revolution in the interpretation of economic relations.

phenomenon
situation:

is

now

An

analogous

observed in the case of meaning and the sign-

here, too,

a specific "sign fetishism" prevails and to

a considerable extent hampers the understanding and solution


of the problem. Considerable scientific credit must be ascribed
to those

who

abolish or at least oppose that fetishism by drawing

attention to the apparently trivial fact that the sign-situation


relation

is

between men who "produce" signs (or use them). This credit

holds good even

if it is

tions of theirs. I

mention

neither the first

nor the only one to advance such an opinion.

not possible to agree with other asserthis

because on that point Gardiner was

Leaving the Marxist conception apart for the time being,

it

must

be stated that the understanding of that point, essential for the


theory of communication,

is

to be

found

in principle in all the

pragmatists and behaviourists, and also in adherents of other


trends

Morris,
is

who have been influenced by pragmatism (for instance


who is connected with neo-positivism, and Urban, who

connected with Cassirer; both are Americans and both have

been influenced by pragmatism). The exceptional position and


role of Gardiner consists in his having succeeded in combining
his struggle against "sign fetishism" with a consistently reahst

materiahstic) interpretation of the rela-

(one should rather say

tion between the sign

and the

Gardiner, of course,

is

objective reahty.

concerned with language and speech,

and consequently he speaks of verbal

signs.

But

this fact

does

not hmit the significance and the general vahdity of what he says.
Linguists,

he maintains, have never forgotten the role of words

(verbal signs) in the

communication process.

On

the contrary,

the importance attached to the role of those signs veiled all their

other elements, that

is

the

men who communicate with one another,

and the objects to which the communication process pertains


and without which the signs would lose all sense. This is why
Gardiner wrote:

Selected Problems of Semantics

.226

The statement

"...

am

about everything

short,
is

my

about

writing,

ever to

realistic

make wide

Why,

can speak about

house,

appeal,

my

books,

in the world.

else
it

must

this

pen with which

my

family, and, in

If linguistic theory

clearly be placed

The rudest

than at present.

basis

he can talk about


touch.

speech serves to express thoughl

that

simply ignores the fact that

villager

the various things which he

all

then, should that truth be hidden

upon a more

knows

that

can see or

from the

theorist

of language?" 7.
It is

not

difficult to

observe that in the course of the analysis

presented here and especially as a result of a proper selection

we have obtained a positive answer to the question as


to be meant by the "sign-situation". The criticism of
fetishism" inherent in the scheme suggested by Ogden

of

texts,

to

what

"sign

is

and Richards was contained

both in the

implicitly

text of

John-

son and in that of Gardiner. The criticism of idealism in the


interpretation of the sign-situation

words of Gardiner who

contained explicitly in the

is

same time notes the

at the

thought element in the communication


case of the

word he combines

ble whole. All in

may

Gardiner, one
least
in

sign

say that

into

an

indivisi-

be in solidarity with

the sign-situation

occurs when at

two men communicate one with another by means of signs

order to transmit one to another their thoughts, expressions of

feelings, will, etc., connected with


to

and thought

in declaring oneself to

all,

role of the

process, although in the

some

object (universe of discourse)

which their communication pertains. In other words, whenever

and the sign-situation occur, the sigiT must refer to some


object (directly or indirectly), and there must be at least two
partners in the process of communication by means of that sign:
he who uses the sign in order to convey his thoughts, and he
the sign

who

perceives

and

interprets

it

(and consequently, understands

Thus, the sign-situation turns out to be something extremely


common, as common as is the process of communicating by
it).

7 Tbid.,

p.

22

(italics

A. S.)

The Meanings of "Meaning"

227

means of signs, the only process of human communication,


that is, which we know in practice.

The problem of meaning, too, is rooted in the sign-situation,


meaning is, as we have seen, inseparably connected with
the sign. A sign without meaning is an inherently contradictory
notion, for only what we call meaning turns material objects
and events into signs; meaning without a sign is a product of
idealistic speculation in the same way as is motion without matsince

ter that

moves.

Thus we are

in possession of a start-line for our attack

on the

problem of meaning, for the assessment of other people's standpoints,

and

for the formulation of a view of our own. Let us

begin with a

of the possible interpretations of meaning, so

list

drawn as to include both the opinions which are in fact represented, and also such important combinations as can be obtained
by manipulating the elements of the
^

sign-situation:

(1)

meaning

is

the object, of which the sign

(2)

meaning

is

a property of objects;

(3)

meaning

is

an

is

a relation:

ideal object,

is

the name;

or an inherent property of

thought
(4)

meaning
(a)

between

signs,

(b) between the sign and the object;


(c)

between the sign and the thought about the object


in question,

(d)

between the sign and human action,

(e)

between the

men who communicate

by means of

Of

course, like

of schematism, but
in

all
it

with one another

signs.
classification,

this

one commits the

sin

enables us to take into account and put

order the most important opinions held in that matter, and at

the

same time, through the exposition and

people's views,

own

it

standpoint.

criticism

of other

leads systematically to the presentation of our

228

Selected Problems of Semantics

2.

Nothing

MEANING AS A REAL OR IDEAL OBJECT


simpler than to take such elements of the sign-

is

situation as the object to

thought

which communication

pertains, or the,

which is inseparably
to divorce them
for

about that object

with the sign that stands

it

connected
j

from

that

and thereby to impart to them some absolute


character. The more so since the concentration of our attention
and interest on such elements is not always due to an erroneous \
theoretical standpoint, but is often dictated by practical needs.
Consequently, although the identification of meaning with the
object denoted is from the theoretical point of view simply in
sign-situation

:j

i|

contradiction

with

its

identification

about

with the thought

two solutions are of the same type and may

that object, the

as

such be treated, in a sense, jointly.

A. "Meaning" as the object denoted

The differentation between "meaning" and "denotation",


which played such an important role in Frege's analysis 8, was
and then again brought to light by Russell 9. Thus,
the issue is linked with the names of these two logicians and
philosophers (from the formal point of view, the issue was also
forgotten,

raised

by Husserl

in Logische Untersuchungen, Vol. II, Pt.

p. 47,

but that differentiation acquires, in the light of his theory of


intentional acts, a special character

and would require separate

analysis).

Frege postulates two expressions: "the morning star" and


"the evening star", and raises the problem of their meanings.

His conclusion
8

Kritik,

G. Frege,

is

"Vom

1892; there

Philosophical
9 B.

that in a sense the meanings of these expres-

is

Sinn und Bedeutung", in Zeitschrift fiir Philosophische

an English-language version

Writings of Gottlob

Russell,

Frege,

in Translations

"On Denoting", Mind, 1905; quoted


London 1956.

Logic and Ktwwledge. Essays 1901-1950,

from

the

Oxford 1952.
after

B,

Russell,

t!

The Meanings of "Meaning"

229

sions are identical, but in another sense they are different.

we ask "What

When

meant by 'the morning star'?", "What is meant


by 'the evening star'?", and bearing in mind the object which
these expressions (of which they are names) denote, then the
meanings of the two expressions are identical, for the object
which they denote is one and the same. But when we consider
the content of these expressions, the way in which they denote
is

their designata,

then their meanings appear

be demonstrated by the fact that people

agreement

concerning these

expressions

different.

may

This can

reach a linguistic

and formulate

their

definitions without realizing that they refer to one and the same
object.

There

is

nothing strange in that, since we have to do with

and consequently with

questions

different

different

meanings

of "meaning". Frege reserved distinct terms for each: "Bedeutung"


for the

former (translated as "denotation" by Russell and as

by Black), and "Sinn" for the latter (translated


as "meaning" by Russell and as "sense" by Black). In the former
case, when we ask "What does ... mean?", we ask in the sense
of "What is ...?", and we are concerned with an object, the
"reference"

designatum of the name. In the

latter case,

content of the name. Therefore Frege

"A proper name


expresses

its

of a sign

we

is

the

correct in stating:

(word, sign, sign combination, expression)

sense, stands for or designates

express

we ask about

its

sense and designate

its

reference.
its

By means

reference"io.

do not raise here the issue of denotating sentences, which


is the main subject matter of the analysis carried out by Frege
and above all by Russell (it is the foundation of Russell's theory
I

of descriptions).

We

are here interested principally in the distinc-

meaning and denotation. That distinction is equivalent to the statement that in a certain sense "meaning" is identical
with the object of which the given expression is the name (I understand here the term "expression" in a broad sense, covering
tion between

both verbal signs and the combinations of such).

Frege, op.

cit.,

p. 61.

16

230

Selected Problems of Semantics

This

is

an indubitable

fact

which leads to the following co

elusion: in view of the clear difference existing between these

meanings, and in view of the specific character of the question

"What does

...

mean?" one should,

in order to avoid misunder-

was done by Frege. Kotarsame course. Consequently, we note this


specific meaning of "meaning" in the case of names, but must
insist that we have here to do with a special case (that of names)
and that such ambiguity can to a large extent be eliminated
by a consistent use of distinct terms, "meaning" and "denotation".
Frege's idea goes back to the tradition of the denotation
and the connotation of ideas as formulated in the conception
of J. S. Mill, although the two ideas are not identical. But then
standings, use different terms. This
binskiii followed the

the adherents of the opinion listed in the classification above

under

(2)

go back directly to Mill's conception. They understand

"meaning" as an inherent property of the object to which the


sign refers.

an ideal

The "object"

is

understood here in a

object, a notion. In that case, the

specific

sum of

that constitute the essence of the notion understood as


jective

entity,

i.e.,

its

connotation,

is

way, as

the properties

an ob-

equivalent to meaning.

Thus meaning is inherent in the object, and it might equally be


said that meaning is the object, since it is equivalent to its essence.
Of course, I take no responsibility either for these "objects"
or for "essences".

have merely related a certain point of view

in order to realize those

meanings of "meaning" which may be

encountered in practice. The other extreme, within the type of


the solutions of the problem of meaning which
discussing,

B.

is

The conception of

intentional

In our classification,

are

now

meanings

Husserl's theory

is

listed

which groups theories interpreting meaning as an


or as an inherent property of thought. Should
11

we

represented by Husserl's theory.

T. Kotarbiriski, Elementy

...

[Elements...],

under

we take

Lwow

(3),

ideal object,

1929.

Husserl's

The Meanings of "Meaning"

Statements at face value,


class, since

tion.

must therefore

anon.
I

to include his theory

Husserl vigorously disclaims such a concep-

in this
I

we would not have

231

justify

my

classification,

explanation

must now begin with another

devote to Husserl's conception so

but of that

much

as to

why

attention, out of all

compared with my treatment of other theories.


are two reasons: the influence of Husserl's
theory upon contemporary philosophy in general, and its influence upon Pohsh philosophy in particular. The latter point
seems to me decisive. The fact that such positivistically-minded
thinkers as Ajdukiewicz^^ and Czezowskii^ have been influenced
precisely by Husserl's theory of meaning is, to say the least,
proportion

For

if

this there

striking,

and

my

in

is

opinion a serious philosophical misunder-

standing. Regardless of
positivist trends,

doubt in direct
serl's

all

our likes or dislikes concerning the

must be conceded that they are beyond all


contrast with the initial assumptions of Husit

phenomenology.

Husserl's

philosophy

is

undoubtedly

same time it proves that


an aUiance between precision of formulations and an idealism
that borders on mysticism is not impossible, and that no undue
a product of a brilliant mind, but at the

importance should be attached to the postulate of formal correctness and precision of analysis. I am not one of those adherents of the

Lwow-Warsaw

school

who

are so

"radical" in

their opinions that, following the neo-positivists in that respect,

they are ready to consider as nonsense any exposition which

does not comply with their criteria of what

is

scientific.

There-

do not understand" Husserl.

do under-

stand the sense of the words in his books, although

often find

fore I

do not say that

that difficuh.

But

"I

do not understand how people who advocat-

ed the principles of a rigorously


cal theses

12

sions],
1-^

can

fall

K. Ajdukiewicz,

Lwow

scientific character in

philosophi-

in with the mysticism of Husserl's concep-

znaczeniii wyrazen

[On the Meaning of Expres-

1931.

T. Czezowsici, Logika [Logic],

Warszawa

1949.

'

232

Selected Problems of Semantics

tions of intentional act, etc. Is that just a misunderstanding?

Or

is

it

an "indulgent" interpretation of what Husserl

interpretation disregarding context? This, however,

most important

What

point.

is

important

says,

an

not the

is

the fact that Pohsh

is

ij

philosophy, the traditions of which are so ahen to phenomenol;

This fact indicates

ogy, did prove susceptible to Husserl's ideas.


that the problem should be given

seem necessary
It is

more

attention than

would

'

at first glance.

a truism to say that in order to understand any idea

belonging to a philosophical system one has to interpret that


idea in the context of that system. This remark

is

addressed to

the adherents of the intentional conception of meaning.


interpretation of Husserl's theory of

meaning and of

The

intentional

act does involve his entire system including his epoche, eidos,

Wesenschau,

the

theory

of universals,

phenomenological

the

would not be possible to understand


what an "object" means in his system, what is an intentional
method,

Otherwise

etc.

it

act that contains as object, etc. Unfortunately, there

word about

is

tional conception.

would be easy

On

to be misled into beheving that Husserl's theory

represents a variation of the psychological theory, which

cerned with the dispositions of

How

not a

works of Pohsh adherents of the intenthe contrary, when reading those works it

that in the

sharply

men

Husserl would protest against

misunderstanding of his basic ideas!

is

con-

using a certain language.

such a flagrant

not

It is

my

intention to

expound here the extremely comphcated theses of Husserl's


philosophical system and

all

the resulting imphcations for the

problem of meaning (and it must be borne in mind that all his


more important theses have such implications). I should only

form of quotations,
theses without which

as far as possible)

like to present (in the


at least those of his
is

in my

opinion

it

not possible to understand the conception of intentional act

and, consequently, Husserl's theory of meaning.


I
is

shall begin

by stating the well-known

an ideahst who, in continuing Platonic

fact

that Husserl

traditions, not only

The Meanings of "Meaning"

of ideal

recognizes the existence

233

but also

entities,

considers

meaning to be such an entity.


Let us be fair: Husserls does not admit he is a Platonist,
and even disclaims that expressis verbis:
"Die Bedeutungen bilden, so konnen wir auch sagen, eine

im Sinne von

Klasse von Begriffen

darum

Sie sind

in der 'Welt', so in

existieren;

Greiste

'allgemeinen Gegenstdnden\

nicht Gegenstande,

einem

wenn

die,

oupdcvLoi;

xo-koc,

nicht irgendwo

oder im gottlichen

denn solche metaphysische Hypostasierung

ware absurd"i4.

Yet the old Marxian principle that men,

like social classes,

should be judged not by what they think of themselves, but

by what they actually do, makes us sceptical.

we read on:
Wer sich daran gewohnt

And

our scepticism

grows as
"...

hat, unter Sein nur 'reales' Sein,

unter Gegenstanden reale Gegenstande zu verstehen,

dem

wird

Rede von allgemeinen Gegenstanden und ihrem Sein

die

grundverkehrt erscheinen:
finden,

als

dagegen wird hier keinen Anstoss

wer diese Reden zunachst einfach

als

Anzeigen

fiir

Geltung gewisser Urteile nimmt, namlich solcher, in denen

die

iiber

Zahlen, Satze, geometrische Gebilde u. dgl. geurteilt wird, und

nun

sich

fragt,

ob nicht hier wie sonst

als

Korrelat der Urteils-

geltung dem, woriiber da geurteilt wird, evidenterweise der Titel

'wahrhaft

seiender

In der Tat:

Gegenstand'

logisch betrachtet,

die

Korper sieben Gegenstande, ebenso wie


Satz

vom

Krafteparallelogram

werden

zugesprochen
sind

ein

miisse.

sieben regelmassigen

die sieben

Weisen; der

Gegestand so gut wie die

Stadt Paris"i5.

Has not Husserl made a mistake,

then, in assessing his

opinions and their genealogy? But that

What

important

is

is

that

14

And

that fact

is

beyond

entities,

and meanings

dispute:

E. Husserl, Logische Untersuchungen, Vol.

15 Ibid.

own

a secondary issue.

Husserl represents the standpoint

of objective idealism as concerning ideal


in particular.

is

2, Pt. 1,

HaUe

1913, p. 101.

Selected Problems of Semantics

234
"Naturlich

ist

auf eine Stufe zu


Widersinnigen

...

es nicht unsere Absicht, das Sein des Idealen

mit

stellen

Die

dem

idealen

Gedachtsein des Fiktiven oder

Gegenstdnde

...

wahr-

existieren

Es hat evidenterweise nicht bloss einen guten Sinn, von


solchen Gegenstanden (z. B. von der Zahl 2, von der Quahtat
haft.

vom dem

Rote,

Satz des Widerspruches u. dgl.) zu sprechen und

mit Pradikaten behaftet vorzustellen, sondern wir erfas-

sie als

sen auch einsichtig gewisse kategorische Wahrheiten,

die

solche ideale Gegenstande beziiglich sind. Gelten diese


heiten, so
setzt

muss

all

das sein, was ihre Geltung objektiv voraus-

Gilt uns alles,

...

was

ist,

mit Recht

seiend vermoge der Evidenz, mit der wir es


erfassen,

auf

Wahr-

dann karm keine Rede davon

berechtigung des idealen

Seins

seiend und
im Denken als

als so

als

sein,

seiend

dass wir die Eigen-

In der Tat

verwerfen diirfen.

kann keine Interpretationskunst der Welt die idealen Gegenstande


aus unserem Sprechen und Denken ehminieren"i6.
This needs no comment. And if there is any difference between
his standpoint and Platonism, then it is such as the difference
between a green devil and a yellow one. Anyhow, his views on
ideal entities in general are closely connected with his view of meaning as an ideal entity, since for Husserl meaning
ideal entity.
his

Whoever

fails

is

an objective

to see and understand that, blocks

road to understanding the theory of intentional acts and the

intentional

conception

of meaning.
|

"Wir haben bisher vorzugsweise von Bedeutungen gesprochen, die, wie der normalerweise relative Sinn des Wortes Bedeutung es schon besagt, Bedeutungen von Ausdriicken

An

sich besteht aber kein notwendiger

den idealen Einheiten, die faktisch

sind.

Zusammenhang zwischen

als

Bedeutungen fungieren,

und den Zeichen, an welche sie gebunden sind, d. h. mittels


welcher sie sich im menschlichen Seelenleben realisieren. Wir
konnen also auch nicht behaupten, dass alle idealen Einheiten
dieser

16

Art ausdriickliche

Ibid., pp.

124-126.

Bedeutungen

sind.

Jeder

Fall

einer

The Meanings of "Meaning"

neuen Begriffsbildung belehrt uns,


realisiert, die

vorher noch nie

235

wie sich eine


war.

realisiert

Wie

Bedeutung
Zahlen

die

dem von der Arithmetik vorausgesetzten idealen Sinne


nicht mit dem Akte des Zahlens entstehen und vergelien, und wie
in

daher die unendliche Zahlenreihe einen


idealen

einer

Gesetzlichkeit

objektiv

scharf umgrenzten

generellen Gegenstanden darstellt, den

festen,

Inbegriff

von
von

niemand vermehren und

vermindern kann; so verhalt es sich auch mit den idealen,

rein-

den Begriffen, Satzen, Wahrheiten, kurz


den logischen Bedeutungen. Sie bilden einen ideal geschlossenen
logischen Einheiten,

von generellen Gegenstanden, denen das Gedachtund Ausgedriicktwerden zufallig ist. Es gibt also unzahlige
Inbegriff

Bedeutungen, die im gewohnlichen relativen Sinne des Wortes

Bedeutungen sind, wahrend sie niemals zum Auskommen und vermogen der Schranken menschlicher Erkenntniskrafte niemals zum Ausdruck kommen konnen''^'^.
bloss mogliche

druck

This,

on the

again,

issue of

calls

meaning

to the principle

for
is

no comment.

Husserl's

adopted in the present work,

almost exclusively to quotations. This


does a quotation

standpoint

presented plainly and clearly. Contrary

present

its

author's

is

opinion

we have to do
which look somewhat strange to a modern
authoritative way, but also

I confine

in

myself

not only

deliberate:

the

most

here with opinions


author, so that he

might easily be suspected of having distorted them or presented

them

inexactly.

Husserl's opinions as quoted above are nothing extraordinary


in the context of his system:

this

is

proved by his interpreta-

tion of sentences belonging to so-called pure logic.

When

speaking

of the vacillation of meaning of words, Husserl makes a distinc-

meaning
and changeable, and meanings as ideal entities which
are unchanging. Thus meaning always is one and the same, but
its expressions can be various and changeable. Now there is a

tion as between individual psychic processes in which


is

vacillating

17 Ibid.,

pp. 104-105.

236

Selected Problems of Semantics

from unchanging meaning of

direct passage

and

its

so-called pure logici

tasks.

"In der Tat hat es die reine Logik,

wo immer

fen, Urteilen, Schliissen handelt, ausschliessUch

sie

von

Begrif-

mit diesen idealen

Einheiten, die wir hier Bedeutungen nennen, zu tun..."i8.

This

is

Husserl,

quite comprehensible

who

in that matter

meaning as an

ideal entity

is

is

when

it

is

realized that

for

faithful to Bolzano's tradition,

identical with a proposition (as

opposed to "sentence" and "judgement"). For him, there

of judgement, experienced actually

a difference between acts

and changeable; and

is

ideal content or proposition as the unchang-

<l

;j

'

ing partner of different statements. This distinction between the


,

two kinds of judgements

analogous to Peirce's distinction

is

between token (experience of a given sign-situation) and type

The

(a certain type of experience of sign-situations).

between Husserl and Peirce

is

ositions the status of ideal entities

which are absolutely

with meanings and are the domain of "pure"

"Die

Ideahtat

Bedeutung

des

zeigt sich in

difference

that the former ascribes to prop-

Verhaltnisses

zwischen

identical

logic.

Ausdruck und

Beziehung auf beide Glieder sofort daran,

dass wir, nach der Bedeutung irgendeines Ausdrucks


quadratischer Rest) fragend,

(z.

B.

unter Ausdruck selbstverstandhch

nicht dieses hie et nunc geausserte

Lautgebilde

meinen, den

und identisch nimmer wiederkehrenden Schall. Wir


meinen den Ausdruck in specie. Der Ausdruck quadratischer
Rest ist identisch derselbe, wer immer ihn aussern mag. Und
wieder dasselbe gilt fiir die Rede von der Bedeutung, die also selbstfliichtigen

verstandhch nicht das bedeutungsverleihende Erlebnis meint"i9.

The matter

is

quite clear,

and the mechanism of building

a system based on objective idealism can be seen in detail. Husserl's starting


is

point for constructing meaning as an ideal entity

the fact that

when we have

18 Ibid.,

pp. 91-92.

19 Ibid.,

pp. 42-43.

to

do with any statement,

e.g.,

ij

'

The Meanings of "Meaning"

"the square

237

a quadrangle having equal sides and equal angles",

is

we are concerned with a certain thought which repeats


whenever that statement
sense

is

itself

actually reproduced, provided that

is

its

understood, but regardless of the individual differences

between the psychological acts which accompany the various

For Husserl,

of that statement.

occurrences

proof that meaning must

exist as

an

this

sufficient

is

ideal entity, regardless of

the concrete experiences of judgements having that meaning.

"Mein

Urteilsakt

vergehend. Nicht

ist

Inhalt, dass die drei

ist

ein fliichtiges Erlebnis, entstehend

und

aber das, was die Aussage aussagt, dieser

Hohen

schneiden, ein Entstehendes

eines Dreieckes sich in einem

und Vergehendes. So

Punkt

oft ich,

oder

wer auch immer, diese selbe Aussage gleichsinnig aussert, so oft


wird von neuem geurteilt. Die Urteilsakte sind von Fall zu Fall
Aber, was

sie

urteilen,

iiberall dasselbe.

Es

ist

verschieden.

das

ist

Identisches,

So verhalt
falsch

es

ist

die

es sich bei alien

sie

im strengen Wortverstande

selbe geometrische

Aussagen,

oder gar absurd sein

erkennen wir

ein

und

eine

auch jeweils

wir legen sie nicht willkiirlich

...

was die Aussage besagt,

mag

auch, was

Wahrheit.
sie

sagen,

Als Identisches der Intention

Akten der Reflexion;


den Aussagen ein, sondern finden

in evidenten

sie darin"20.

That conception of meaning as an objective


to objective idealism.

As we

entity

intimately connected with subjective idealism. But

it is

amounts

shall see later on, in Husserl's case


let

us not

be misled by that apparent opposition. The ideal objects and


absolute ideas in systems of objective idealism, absolute idealism,
etc.,

are nothing but individual consciousness, artificially shifted

into the supra-individual sphere

and thus transformed

into

thing absolute. In this interpretation, objective idealism

someis

but

a transformation of subjective ideahsm. In Husserl's case, the

mechanism of that transformation

20 Ibid., p. 44.

(as

demonstrated

above)

238

is

Selected Problems of Semantics

obvious. Tlie individual acts of understanding certain expres-

what

sions are taken as the starting point:

common

is

to these

individual experiences (after the rejection of their variable ele-

ments connected with the

with the situation of a given experience,

and in this way


meaning in the

is

individualities of the persons concerned

constructed

what

etc.)

is

i|

extrapolated,

said to be a proposition or

is

That

ideal sense of the word.

it is

a construe-

and what construction it is, can be shown (contrary to Husserl's


assertions that what is involved here is an act of direct Wesenschau supported by the testimony of self-evidence) by what Hus-

tion

Hence Husserl's

ideal object or as

says himself, as here quoted extensively.

serl

theory can be headlined

"Meaning

an

as

an inherent property of thought", since meaning interpreted


as an ideal entity is but an absolutized and inherent property

of thought connected with certain experiences in understanding


expressions.

This

is

'

confirmed by the intimate links between

that conception and the theory of intentional acts.

Now

that

we know, probably

clearly enough,

understands meaning as an ideal entity,


tional acts

let

how

Husserl

us pass to inten-

and to the intentional conception of meaning, which

has found reflection in Polish philosophical literature.

Whoever would understand the conception of intentional


acts, as conceived by Husserl, and apply it to the problem of
meaning, must bear in mind not only meanings as ideal objects,
but also the entire theory concerning the essence of things and the
direct seeing of things.

and

eidetic seeing,

True, the conception of epoche, eidos,

fundamental to the phenomenological method,

developed in the period that followed Logische Untersuchungen,

wherein the theory of meaning and the theory of intentional acts

were formulated; yet that work already includes ideas that came
to be the foundations of those theories. This refers above

the idea of a direct seeing of

meaning

as

an ideal

all

entity,

to

and

the realization of the consequences resulting therefrom for the


intentional theory of meaning.

'

The Meanings of "Meaning"

"Was 'Bedeutung'

ist,

das kann uns so unmittelbar gegeben

ist,

was Farbe und Ton

wie uns gegeben

sein,

nicht weiter definieren, es

We

ist.

Es

lasst sich

ein deskriptives Letztes"2i.

ist

have to do not only with a direct perception of meaning,

but also with a criterion which makes

same meaning

the

231)

it

possible to single out

in all individual statements,

ion of self-evidence.

That

namely the

criter-

together with the theory

criterion,

of intentional acts and of intentional objects, was taken over

by Husserl from
"...

sich

die
soil;

durch

erst

um eine blosse

Erklarungsergiebigkeit

ihre

sondern wir nehmen es

heit in

Franz Brentano.

master,

his

Es handelt sich dabei auch nicht

als eine

Anspruch und folgen

Hypothese,

rechtfertigen

unmittelbar fassliche Wahr-

hierin der letzten Autoritat in alien

Erkenntnisfragen, der Evidenz"22.

Only when we have understood that meaning,


view,

only

is

an ideal

entity,

in Husserl's

"a unity in plurality", a unity of species,

when we have understood

in direct perception (Schau),

that

meaning

is

accessible to us

can we understand the intentional

conception of meaning.

Let us begin by recalhng the elementary facts concerning


the theory of intentional act.

As already

indicated, that theory

was taken over by Husserl from Franz Brentano who was

in

turn a transmitter of scholastic tradition, where, in the last analysis,

the conception of intentional acts

is

rooted. But, without

reaching so far back, one has, in order to understand Husserl,


to
to

become acquainted not only with Brentano's Psychology,


which Husserl

refers expressis verbis,

but also with Brentano's

correspondence with Marty, Krause, and others, on the subject


of intentional objects23.

Even a nodding acquaintance with the theory of intentional


shows that it would be an unfortunate error to identify the

acts

intentional
21

act,

as

Ibid., p.

183.

22 Ibid., p.

100.

23

Cf. the

understood by Husserl

posthumous

collection

(and

consequently

Wahrheit und Evidenz, Leipzig 1930.

240

Selected Problems of Semantics

his intentional conception of meaning), with the thesis that

we

express certain words

we have

when

certain definite dispositions!


j

to understand that state in such

and such a way. The theory'

of intentional acts embraces a thesis about such or other psychic


dispositions concerning the

meaning of expressions, but

at the

same time it includes a lot of other theses, such as a soberly thinking man must consider to be at least strange, theses which are
taken over as part and parcel of the theory of intentional acts.
According to Brentano (Husserl adopts

his theses), the dif|

ferences between the various types of experience consist in "the


\

manner of

reference of consciousness to a certain content" or,

Hus-

in the mediaeval scholastic terminology, in their intention.


serl says:

In

"...

der

Wahrnehmung wird etwas wahrgenommen,

in

der Bildvorstellung etwas bildlich vorgestellt, in der Aussage

etwas ausgesagt, in der Liebe etwas gehebt, in Hasse etwas gehasst,

im Begehren etwas begehrt usw. ... Nur eins halten wir als
fiir uns wichtig im Auge dass es wesentliche spezifische Verschiedenheiten der intentionalen Beziehung, oder kurzweg der In:

tention

(die

ausmacht)

den deskriptiven Gattungscharakter des


Die Weise, in der eine

gibt.

Sachverhalts diesen ihren 'Gegenstand' meint,


als die

Weise des

Urteils, das

'Aktes'

'blosse Vorstellung' eines

den Sachverhalt

fiir

ist

eine andere,

wahr oder

falsch

hah"24.

Thus, every experience in which there


lation of consciousness
to

an intentional

to

object, is

is

a reference

a re-

a certain content or, in other words,

an intentional

act.

As we

shall see,

the content of consciousness and the intentional object are one

and the same. The

difference in intention depends

ference in that content or intentional object. But

pose that
sciousness

we have
and

two things

dif-

and the

inten-

really appearing in consciousness25.

24 Husserl, op. cit., pp. 366-367.


25 Ibid., p. 371.

on the

us not sup-

here to do with a real relation between con-

object, or that the intentional act

tional object are

let

The Meanings of "Meaning"

There

is

241

only an intentional experience, the characteristic

trait

of which consists just in the given intention.

"...

Je

nach ihrer spezifischen Besonderung macht

sie

das

diesen Gegenstand Vorstellen oder das ihn Beurteilen usw. voll

und

allein aus. 1st dieses Erlebnis prasent, so ist eo ipso

liegt,

betone ich, an seinem eigenen Wesen

auf einen Gegenstand'

'Beziehung

voUzogen, eo ipso

Gegenstand 'intentional gegenwartig'

das

die intentionale
ist

ein

denn das eine und andere

besagt genau dasselbe. Und natiirlich kann solch ein Erlebnis


im Bewusstsein vorhanden sein mit dieser seiner Intention, ohne
dass

ist

Gegenstand

der

existieren

iiberhaupt

kann; der Gegenstand

Erlebnis; aber er

ist

existiert

gemeint,

und
d. h.

vielleicht

dann bloss vermeint und

ist

gar

das ihn Meinen

Wahrheit

in

nichts"26.

Husserl explains his idea by

imagine Jove. Jove


but in analysing

is

my

way of

the example of

an intentional object of what

experience

I find

how we
imagine,

that object neither in mente

simply nowhere, and yet the imagined

nor extra mentem.

It

picture of Jove

a true experience. But the most interesting

is

is

is

Husserl's assertion that nothing changes

when

the object of

intension does exist.


"Existiert andererseits der intendierte Gegenstand, so braucht
in

phanomenologischer Hinsicht nichts geandert zu

das Bewusstsein
vorgestellte
leicht

ist

Gegenstand

gar widersinnig

existiert,

ist.

oder ob er fingiert und

Fiir

viel-

Jupiter stelle ich nicht anders vor als

Bismarck, den Babylonischen Turm nicht anders

Dom,

sein.

das Gegebene ein wesenthch Gleiches, ob der

als

den Kolner

ein regelmdssiges Tausendeck nicht anders als einen regel-

mdssigen Tausendfldchnef'^''.

A strange theory,

that

it

refers to a reference of consciousness

and to an object of the intentional act, only to state


end that not only is there no object, but that there cannot

to content,
in the

26 Ibid., pp.
27 Ibid., p.

372-373.
373.

242

Selected Problems of Semantics

be one at

all,

since ex definitione there

only and exclusively

is

an act of consciousness as an intentional

Elsewhere in

act.

his

work28, Husserl explains that content expressed in the objective

meaning and object. All this is considerably


clarified by the later development of his theory. In the light of
the phenomenological epoche those judgements which do not
sense

identical with

is

'

pertain to pure consciousness and thereby to eidos, to the es-

sence of things, are subject to suspension. Eidetic reduction sus-

pends the judgements concerning the individual existence of the


object,

and transcendental reduction goes

ing

that

still

further

by suspend;

as

all
J.

is

not a correlate of pure consciousness. In

M. Bochehski

only what

is

correctly remarks, there

given to the subject^^.

referred to pure consciousness,

ent

and absolute character;

it

The

is left

real

this

of the object

world

and deprived of

way,

totally

is

\.\

independ-

its

thus becomes only an "intentional

object", a content of consciousness.

This

is

subjective idealism of the

first

water,

and seems

to

be at variance with the objective idealism of Husserl's theory


of ideal

including meanings. That duality

entities,

is

revealed

both in the development of Husserl's own theory, and in the


development of the phenomenological school. Finally, we must
concern ourselves with one point more: does Husserl somehow
bridge the gap between

his

meaning

the

objective idealism

of his concep-

and the subjective idealism of


conception of intentional acts? Does he formulate any uni-

tion of

as

an ideal

entity

form theory of meaning?


Yes, and no. Yes

in the

sense that he clearly combines

the two conceptions into a whole in a

demonstrate below.
objective

and

No -

subjective

tions are connected

manner which

shall

in the sense that the inner split into

idealism

by a bridge that

remains,
is

and both concep-

very narrow and unsteady

(from the point of view of the uniformity of the theory).


28 Ibid., p.
29 J.

52.

M. Bochenski, Contemporary European

Philosophy, Berkeley and Los


'

Angeles 1956, pp. 137-140.

The Meanings of "Meaning"

243

Husserl thus bridges the gap:

'

mannigfahigen

Die

"...

entsprechenden

Rote

"...

Bedes

Bedeutung
...

verhalt

wie etwa

den hier liegenden Papierstreifen, die

specie zu

in

Rote 'haben'

diese selbe

And

ideal-einen

Aktmomente

zu den jeweiligen Akten des Bedeutens

sieh also

die

Die

Bedeutungsintentionen.

die

Bedeutens,

zur

Einzelheiten

deutung sind natiirlich die

alle

"3o.

elsewhere:

Wunsch hegen mehrere Personen, wenn ihre


ist. Bei dem einen mag der Wunsch
sein, bei dem andern nicht, bei dem einen mit

Denselben

wiinschende Intention dieselbe

voU ausdriicklich

auf den

Beziehung

fundierenden

dem andern mehr

lich klar, bei

anschau-

Vorstellungsgehalt

oder minder unanschaulich usw.

In jedem Falle liegt die Identitat des 'Wesenthchen'


in

den beiden oben unterschiedenen Momenten,

und

Aktqualitat

auch

also

Akte

verleihenden

oben
d.h.

vorweg

offenbar
derselben

Dasselbe nehmen wir


und speziell die hedeutungsAnspruch, und zwar so, dass, wie wir es

derselben

in

Materie.

ausdriicklichen

die

fiir

in

in

ausgesprochen

das in ihnen, was das

reelle

haben,

Bedeutungsmdssiges,

ihr

phanomenologische Korrelat der

idealen Bedeutung bildet, mit ihrem intentionalen

Wesen zusam-

menfant"3i.

Thus the

situation

is

clear.

vidual intentional acts are,

According to Husserl, the indi-

as

it

were,

specifications

of ideal

meaning; their intentional objects coincide with the phenomenolconsciousness) correlate of ideal
ogical (i.e., appearing
in
meaning.

Meaning

is

thus

an

ideal

entity

manifesting

itself

in intentional acts owing to a direct, eidetic, "seeing" (Schau)

of the essence of things.

Is that

"bridge" satisfactory? Let the

adherents bother about that, since for opponents such Schonheitsfehler

We

is

trifle

compared with the basic

as

have presented Husserl's

theory

Husserl to have been a brilliant philosopher,


30

Husserl, op.

31

Ibid, pp.

cit.,

p.

420-421.

100.

thesis of the theory.

of meaning.

marked by

hold
extra-

244

Selected Problems of Semantics

ordinary precision of thought. This

one of the greatest

intellectual

who can

only Bergson

It is

is

precisely

why he has become

trouble-makers of our times4

vie with

him

in that respect.

What'

does that trouble-making consist of? In propagating metaphysianti-scientific opinions.

cal,

In Husserl's case, that occurred, so

which inevitably madeij

to speak, in the full glory of precision,

a great impression on readers.

My

attitude

meaning,
ideal

self-evident

is

adopts the standj

Husserl's philosophy

to

and unequivocal

seeing, eidetic

eidetic

entities,

who

that of a philosopher

point of dialectical materialism

my

In

of

opinion the

reduction, transcendental

reduction, intentional acts, etc., are products of anti-scientific

belated aftermath of the various systems

views,

Why?

philosophy.

mands,

which

among

:\

\[

method
and de-

the requirements of the scientific


is

hostile to all kinds of intuitions

other

things,

tion of the results of research.


all

that

communicability

inter-subjective

Nor do

said.

My

the

is

possibility

due to the

have said,

verifica-

do not maintain

of solving

compared with the


fact that I

atomism

1918), Bertrand Russell reflected


science.

am more

philosophical

concluding lecture on logical

ophy and

say that "I do not understand" what has been

"liberalism", as

positivists,

As

an unsense, mere rubbish clad in a grammatical

is

garb.

on the

(a

series

lot

In

delivered

his

in

between philos-

agree with everything

too pessimistic, but nevertheless his opinion

one and contains a

of the

sceptical about

controversies.

relation

One perhaps cannot

attitude

Russell says, his assessment of the role of philosophy


is

an

is

certainly

interesting

of truth.

between science and philoswhat you more or less know and philosof
is what you do not know. Philosophy is that part
about,
opinion
science which at present people choose to have an
"...

ophy
ophy

is

;j

First of all, because Husserl's ideas are in

of the facts under investigation and the possibility of

that

1|

idealistic

clear opposition to
in thinking,

of

believe the only difference

that science

is

but which they have no knowledge about. Therefore every ad-

The Meanings of "Meaning"

vance in knowledge robs philosophy of


formerly

had

it

become

. . .

245

some problems which


they become soluble,

And of course the moment

minds uninterestbecause to many of the people who hke philosophy, the

they
ing,

charm of it

to a large class of philosophical

consists in the speculative freedom,

in the

fact that

you can play with hypotheses. You can think out this or that which
may be true, which is a very valuable exercise until you discover what
is

when you

true; but

discover what

play of fancy in that region


that region

is

whole

true the

is

curtailed,

and you

will

fruitful

abandon

and pass on"32.

The fact that philosophical controversies continue for miland that the basic standpoints in those controversies have
survived millennia, the fact that to this day there are discussions
lennia

hke those which Schiller so well described in


"Einer, das horet

man

Xenien:

his

dem andern.
Doch keiner

wohl, spricht nach

Mit dem andern; wer nennt zwei Monologe Gesprach?"

all this

seems to confirm the truth of

at least

some of

statements, seasoned as they are with bitter irony.


if I reject

serl

and

his followers reply with equal

if

entities, if

their

eyes

which

demand

my
is

is

scientific

aplomb that

and that

am

it is

it,

my

opponents

is

them a

and

function in the class struggle.

personally think that

and methodologically

B. Russell,

1918, quoted after

anti-scien-

describe their opinions

the sociological point of view

32

wrong?

opponents adduce evidence as a proof which in


a most scientific one? I can multiply arguments

also can, if I feel forced to

justified

precisely

in the

a scientific proof of the existence of ideal

demonstrating that the standpoint of


tific; I

What then

Husserl's theory as a product of imagination, if Hus-

their standpoint

What

Russell's

correct.

ascribe to

But

it

all

from

definite

that

is

has one fundamental

"The Philosophy of Logical Atomism", in The Monist,


B. Russell, Logic and Knowledge, p. 281.

17

246

Selected Problems of Semantics

defect:

does not

it

suffice to convince the

And

opponent

if lie sticks

makes all the difference,


between philosophy on the one hand and such sciences as mathematics, chemistry or any other experimental discipline on the
assumptions.

original

to his

this

other. In the exact sciences, calculus or experiment usually set-

the question. In philosophy

tles

it

not so; this

is

is testified

byj

the history of philosophy and the present situation in philosophy.


I

do not want to engage

beheve that

this

why

in a discussion as to

it is

nor do

so,

philosophy.

fact ultimately discredits

Butj

one has to acknowledge that state of things in detail


and draw appropriate conclusions. Neither my Marxist companions nor the numerous positivist-minded Pohsh philosoat least

phers wanted to draw such conclusions.

And

the basic concluj

the

sion

is:

not

agree

of a philosophy with which one does

adherent

may

nor accused of preaching

neither be ignored

'

'

nonsense.

What
of

do

all,

then

is

a philosopher to do in such a situation? First

This leads to further consequences. Since the views with

exist.

which I do not agree are ahen to

me and even, according to my conhuman knowledge, I must,

viction, detrimental to the progress of

as a philosopher,

oppose and fight them. But since

a field in which victory

may

be

won

operate in

only by convincing

ponent by appropriate argumentation,

my

op-

must familiarize myself

with his opinions in order to be able to combat them properly.


All this

makes

my

victory only possible, but in

no way

certain.

But in order not to be at a loss on that slippery ground of philosophical problems, where one

moves

in a different

way from

on the firm ground of the exact science, one thing


necessary: one must realize not only that there are diff'erenT

where one
is

he should take cognizance of the fact that opposing views

is

ways of solving philosophical problems, but also that these are


different ways.

And

one must clearly

From the

if

one sees and understands those differences

realize which

way and

to

point of view of a maximalist, this

but in fact the requirement

is

where one
is

no small one.

is

going.

certainly not

much,

i|

The Meanings of "Meaning"

MEANING AS A RELATION

3.

Among

247
(1)

the theories which interpret meaning as a relation,

those two which were Hsted


quently, these

two

will

merit special attention. Conse-

last,

now be examined

in detail,

whereas the

rest

be accorded only cursory treatment.

will

When we ask "What

does the word

mean?" we are often conword into a language we know. This


is especially so when we ask about the meaning of words in a language we do not know or about a word we do not know in a
language which in principle is known to us. If, when learning
French, I ask "What does 'la chaise' mean?", I obviously want
. . .

cerned with a translation of that

to

know the

translation of

'la

in English 'the chair' "

thereby to

when

know

Consequently

chaise' into English.

with the reply that '"la chaise' means

rest perfectly satisfied

and

am

satisfied

that

the meaning of that word.

It

ask in English about the meaning of a term

I
is

have come
also similar

do not know,

for example, a technical term. I have said "similar", because

the
is

mechanism of explanation

involved here

by

its

is

in this case

is

different.

What

not just a replacement of a word in one language

equivalent in another language, but a definition of a term.

For instance,

ask "What

such an answer as

"A

is

'differential gear' "

and

differential gear is ...", for I actually

an explanation as to what a
In these cases meaning

is

differential gear

as

much

obtain

wanted

is.

as a relation between signs,

a relation between the definiendum and the definiens, such that


the definiens

known

may

be either a verbal sign in a language that

to us or a definition of a

word given

is

in that language.

Meaning of this type is called lexical meaning. This is one the


most commonly used meanings of "meaning", very important
in practice.

When

it comes to "meaning" understood as a relation between


and the object or the sign and the thought about an
object (according to others: the idea of an object), that is that
particular meaning of "meaning" which prevails in semantics

the sign

248

Selected Problems of Semantics

on a broader

as pursued by logicians, then a critical discussion


theoretical basis

is

We

required.

analysing the last point of our

shall revert to that issue

list,

in

of meaning based on Marxist principles. For


a context that

we

shall

when

connection with the theory


it

is

only in such

be able to assess properly the merits and

demerits of the conceptions enumerated here.

The other extreme, opposite


conception of meaning,
or

all

is

to a psychological or mentalistic

tenanted by the biological conception,

those variations which connect the meaning of a sign with

such reactions of the

human body

as manifest themselves in action.

In the hght of that conception, meaning implies the relation between


the sign and the biological reaction of the
to that sign or

the sign and

in a social interpretation

human

human organism

the relation between

action in a broad sense of the term. If

we

ap-

proach the biological conception from that angle, there are


revealed the links which connect the Pavlovian theory not only

with pragmatism, operationism, and Morris's semiotic, but also


with certain neo-positivist ideas.

The
is

most

anti-mentalistic character of that conception of


clearly

marked

meaning

in the case of Pavlov's reflex theory. First

meaning and interprets the sign-situation, and consequently the communication


process, in terms of stimuli and reactions. The specific nature
of human communication, i.e., communication by means of
a phonic language, is reduced to a more complicated system of
stimuli and reactions to them this is achieved by the introduction
of an additional, higher level of stimuli as signals. It would be
of

all,

that theory eliminates the category of

superfluous to expound here the elements of the theory of conditioned reflexes;

I shall

abstain also from repeating

vations concerning the term "signal" in general,

my

reser-

and the term

"signal of signals" in particular. It will suffice to state that in the

hght of Pavlov's physiological theory, the category of meaning


(in the sense

adopted above) disappears, and the relation wliich

in the mentahstic interpretation bears the traditional

"meaning"

is

name of

tantamount to the relation between the sign ("signal"

The Meanings of "Meaning"

and the

in Pavlov's terminology)

In other words,

"meaning"

is

it

reflex associated

claimed that what

is

249
with that sign.

traditionally called

is

human organism

in fact a reflex of

to the sign

(signal).

In his theory, Pavlov was not concerned with the philosophical


aspect of the

problem of meaning, and confined himself to the

of certain

study

Nevertheles,

organism.

philosopliical

direct,

physiological

specific

his

human

of

reactions

theory has important, though in-

These

implications.

implications

appear

which solve the problem

directly in those philosophical trends

of meaning after the fashion of the biological conception.


First of

all,

mean

here pragmatism.

Not pragmatism

as

interpreted by James, but the authentic pragmatism which orig-

with Peirce and then follows various channels and com-

inates

common name

bines under the

very

much one from

ions

which

cisely

tendencies which often differ

another, from radical subjectivism to opin-

in certain points

with pragmatism that

come
is

close to materialism.

It is

pre-

connected the tendency towards

a behaviouristic interpretation of meaning.

When

the fashion of the biological

I refer to solutions after

have no sort of idea of taking over some elements


of that conception, but refer to a similarity of solutions, to their
congeniality. This applies of course to Peirce who expounded

conception,

his ideas

long before Pavlov.

In his article

"How

to

Make

our Ideas Clear" (1878), Peirce

formulated the basic concepts of his pragmatism; they included


the thesis that meaning is nothing but the practical consequence
of thought as manifested in action.
"...

action

He

The whole function of thought


...

To develop

its

(i.e.,

thinking's

have, therefore, simply to determine


for

what a thing means


3^

says:

Ch.

S. Peirce,

of Change {Selected
Press 1958, p. 123.

"How

to

is

to

A.

produce habits of

meaning, we

S.)

what habhs

simply what habits

Make Our

Writings of

is

it

involves"^^.

Ideas Clear", in Values

Charles

S.

Peirce),

produces,

it

in

Stanford

a Universe
University

250

Selected Problems of Semantics

And

then he concludes:

"It appears, then, that the rule for attaining the third

of clarity of apprehension

is

as follows: consider

what

grade

effects,'

which might conceivably have practical bearings, we conceive


the object of our conception to have. Then, our conception of
these effects

is

the whole of our conception of the object"34.

That idea Peirce next developed in the articles published in


1905 in The Monisfi^, in order, above all, to protect pragmatism
against being vulgarized

by the epigoni. The idea

itself is

very

simple: meaning amounts to effects manifested in action, definite


habits of action. Effects of what? Habits connected with what?

Peirce explains in the paragraph

on thinking. But he understood


communication

perfectly well the function of the sign in the

of thoughts and the connection between meaning and sign. Thus,

we

find here the idea

which Pavlov developed

later on,

although

the starting point of interests was quite different in each case.

That idea came to


tions within

fluence

rest later at the root

pragmatism

itself,

and

on other philosophical

of the various interpreta-

at the

same time exerted

in-

trends.

Within pragmatism, the two extreme tendencies concerning

by F. C. S. Schiller on the one hand,


and by George H. Mead on the other.
In 1920, Mind published materials from a symposium on
meaning. The principal participants in the discussion were Schiller, Russell and Joachim. Schiller, starting from a pragmatist
position, ascribed a purely subjectivist and voluntaristic character
to the theory that meaning amounts to effects manifested in
that issue were represented

human

action.

"What

meaning be neither an inherent property of objects


between objects at all, not even between
the object and a subject, but essentially an activity or attitude
taken up towards objects by a subject and energetically projected
nor a

if

static 'relation'

34 Ibid., p.
35

ism".

124.

They are the

articles

"What Pragmatism

Is?"

and

"'Issues

of Pragmat-

The Meanongs of "Meaning"

them

into

like

an a

particle,

until they, too,

begin to radiate with 'meaning'? Here,


to

25]

if

grow

active

and

anywhere, would seem

the clue to the mystery of 'meaning' "36.

lie

Schiller then

immediately ranges himself on the side of a vol-

untaristic interpretation of

meaning, and

meaning

essentially personal", so that

is

states that

"meaning

is

with respect to

relative

the whole personality of the subject37.

Another interpretation of the problem, one in conformity


with the basic assumptions of Peirce,

Mead^s. That author reveals, however,

is

given by George H.

effects

of the direct

in-

on him of the Pavlovian theory, although he speaks not


of reflexes of human organism, but of responses. At the same time.
Mead interprets the problem of meaning from the social point
of view, in the context of the communication process. A few of
his statements, those which are representative of his ideas, are
quoted below by way of example.
"Meaning arises and lies within the field of the relation between the future of a given human organism and the subsequent
fluence

behavior of this organism as indicated to another

human organism

by gesture"39.

"Meaning

is

thus not to be conceived, fundamentally, as

a state of consciousness, or as a set of organized relations existing or subsisting

mentally outside the

which they enter; on the contrary,


ly,

it

field

of experience into

should be conceived subjective-

as having existence entirely within this field

itself.

The

re-

sponse of one organism to the gesture of another in any given


social act is the
"...

meaning of that

Meaning involves a

gesture"40.

reference of the gesture of one or-

ganism to the resuhant of the social act


as adjustively
36

it

this reference

Mind, October 1920, No. 116,

37 Ibid.,
38

responded to in

indicates or initiates,

by another organism

p. 389.

pp. 390, 391.

G. H. Mead, Mind, Self and Society, Chicago 1955.

39 Ibid.,

pp. 75-76.

40 Ibid., p. 78.

252

Selected Problems of Semantics

and the adjustive response of the other organism

is

the meaninfi]

of the gesture"4i.

The contentions of that trend in pragmatism which sees


meaning as being effects manifested in human action, as being
the results of communicating thoughts by means of signs, are
in harmony with the tendencies which in the 1920's and '30's:
appeared in Russell's logical atomism, in logical empiricism',
(neo-positivism),

in

operationism

trends are interconnected,

and

influence

All

semiotic.

in

these i

one another, and follow

similar courses in their approaches to meaning. This refers above


all

to certain general conclusions

of experimenters
resentatives

tivism)

from the

(chiefly in the case

of the natural and the

who combatted,

of operationism) and repexact

of traditional philosophy. Another point

laid foundations for a

aimed

at

the

be said that

new

(neo-posi-

sciences

at least subjectively,

they succeeded in that combat, and

the metaphysics

whether or

is,

how

how

metaphysics. Yet, despite

all criticism
it

must

endeavours to find a practical interpretation of

meaning indicated an important problem.


Operationism is undoubtedly the most
respect. Its founder, Bridgman, behaves as a

interesting

physicist

in

that

who

tries

to explain, for the purpose of the discipline he represents, certain

general theoretical categories, in doing which he takes into ac-

count above

all

the great revolution effected in a

concepts by Einstein's theory of relativity.

number of

cannot say whether

and exchange of ideas between Bridganyhow, the problems of meaning


and of meaningful statements, which played such an important

there were any connections

man and

the Vienna Circle

role in the evolution of neo-positivism, stands out in full relief


precisely in the case of Bridgman's views.

volved here. First, that meaning

is

Two

equal to the

theses are in-

sum of

the opera-

tions corresponding to a given term; or, in other words, that the

*! Ibid., p. 81.

far

far they themselves

philosophy connected with positivism,

its

practical experience

The Meanings of "Meaning"

253

meaning of a term is to be sought in what one does, and not


in what one says. Second, that a question has a meaning (in the
sense that

it

is

and not purely

sensible,

when one can

verbal)

point to the operations resulting in answer to that question.

These two theses are in close harmony with the standpoint of

on the one hand, and with

Peirce,

that of the neo-positivists, to

be discused below, on the other.

Bridgman

"We

writes:

evidently

know what we mean by

what the length of any and every object


nothing more

is

required.

To

length

is,

if

we can

and for the

find the length of

an

tell

physicist

object,

we

have to perform certain physical operations. The concept of


length
is

is

therefore fixed

measured are

much

as

length

is

when

fixed: that

is,

the operations by which length


the concept of length involves as

and nothing more than the


determined. In general,

more than a

set

set

of operations by which

we mean by any concept nothing

of operations; the concept

the corresponding set

is

synonymous with

of operations. If the concept

is

physical,

as of length, the operations are actual physical operations, namely,

those by which length

is

measured; or

if

the concept

is

mental,

as of

mathematical continuity, the operations are mental opera-

tions,

namely those by which we determine whether a given ag-

gregate of magnitudes

And

is

continuous "^2.

on he thus formulates his ideas:


"For of course the true meaning of a term is to be found by
observing what a man does with it, not by what he says about
further

it"43.

That concept of meaning

is

connected with the concept of

meaningfulness of statements, which has played such an impor-

campaign against metaphysics.


question has meaning, it must be possible to

tant role in the neo-positivist

"If a specific
find operations
42 p.
p.

by which an answer may be given to

W. Bridgman, The

5.

43 Ibid., p. 7.

Logic of Modern Physics,

it ... I

believe

New York

1927,

254

Selected Problems of Semantics

many of the

that

subjects will be

questions asked about social and philosophical

found to be meaningless when examined from

the point of view of operations"44.

The question of meaningful and meaningless statements

brings

us to a certain point which will enable us to understand better


the appropriate conceptions of the neo-positivists, namely the

ambiguity of the expression: "That statement has meaning".


In one case, reference

is

made

meaning

to

in the traditional sense

of the term, which implies that a given statement conveys some


content to the listener or reader. Such statements are contrasted

with nonsense, that

is

such sequences of words as in view of their

disagreement with the grammatical rules of the language involved

"Horse by though seven

(e.g.,

prehensibility of certain

words

or in view of the incom-

still")

(e.g.,

"Woggled

diggles are mig-

ghng") convey no content. In the second case, the point in question


is

the meaningfulness of such statements as can be decided

in practice, as contrasted with unsense,

upon

such statements as

i.e.,

have the grammatical form of sentences and are comprehensible,


but cannot be decided upon in practice and consequently must

be recognized as purely verbal. Thus, unsense has a verbal meaning (that

is,

some

sense)

and

is

not a nonsense, but being an

undecidable and unverifiable statement

is

importance and cannot be considered a

devoid of practical
scientific

statement.

That distinction between sense and unsense was enlisted by the

campaign against metaphysics.


on meaning of the pragmatists and the opera-

neo-positivists in their

The

basic ideas

tionists are

found

in neo-positivism to

be one of the elements

in that oft-changing trend, non-uniform in

its

conceptions.

Historically, neo-positivism (logical empiricism)

was shaped

under the overwhelming influence of Russell's logical atomism,


chiefly
it is

though the intermediary of Wittgenstein. Consequently,

desirable to begin a presentation of the neo-positivist views

44 Ibid., pp. 28-30.

An

interesting exposition of that standpoint

found in Operational Philosophy by Anatol Rapoport,

New York

is

to be

1953.

The Meanings of "Meaning"

255

with Russell's and Wittgenstein's opinions on meaning. This


I

neither easy nor simple, since their opinions are complicated,

is

by no means uniform, and often

(cf.

Wittgenstein)

marked by

even radical evolutions.

two interpretations of meaning as a relation.


is evidently influenced by pragmatism and
namely
that which maintains that the meaning
behaviourism,
of signs is to be sought by observing how they are being used.
In his article "On Propositions: What They Are and How They
Mean"45 (1919), which later became the foundation of the 1920
Russell has

At

one of them

least

symposium, Russell makes

explicit reference to behaviourism^^,

and adopts the standpoint of that theory concerning the so-called


demonstrative use of language, consisting in pointing out the
properties of actually existing enviromment.

pressions of a language

means here

a) to use

To understand
them

ex-

in appropriate

and b) to act in a definite way on hearing them. Between


the sign and its meaning there is a causal relation in the sense
that the signs are evoked by appropriate events, and also in the
sense that they in turn evoke other events, namely human action.
conditions,

Russell concludes quite in the behaviourist spirit (as noted

by F. C.
images)

S.

its

of meaning-

not necessary to 'understanding' of a word that a person

should 'know what

word means

it

means', in the sense of being able to say

so-and-so'.

vague; but the meaning

less

his conception

"It is

'this

who opposed

Schiller,

is

word has a meaning, more or

only to be discovered by observing

first, and the meaning is distilled out of it.


word to its meaning is, in fact, of the nature
and there is no more reason why a person using

use ; the use comes

The

relation of a

of a causal law,
a

word

is

for a planet

ler's

45

meaning than there


moving correctly to be conscious of Kep-

correctly should be conscious of

which

is

its

law."

In B. Russell, Logic and Knowledge, pp. 285-320.

46 Ibid., p. 291.

256

Selected Problems of Semantics

"He

[a man] 'understands' a word, because he does the right


Such 'understanding' may be regarded as belonging to
the nerves and brain, being habits which they have acquired
while the language was being learnt. Thus understanding in this

thing.

sense

may be

But in

reduced to a mere physiological causal law"47.

his theory

of meaning Russell did not maintain a con-

sistently behaviourist position. In thinking, verbal signs are

images retained by

would say

their function consists in describing

memory

or produced by imagination (we

rather: reproduced

and produced

"These two ways of using words


as the use of
it

words

And

may

in 'thinking'. This

depends upon images, cannot be

lines.

used

and

for "narrative" purposes,

way of using words,

fully dealt

most

this is really the

ideas).

be spoken of together
since

with on behaviourist

essential function of words:

that primarily through their connection with images they bring

us in touch with what

operate without the

is

remote in time or space.

medium of images

this

When

seems to be a

coped process. Thus the problem of the meaning of words


duced to the problem of the meaning of images" '^^.

they
telesis

re-

Thus, in Russell's case, the behaviourist interpretation of

meaning
of signs

is

is

accompanied by a second one, whereby the meaning


reduced to images or reproduced and produced ideas.

While in the
field

of the

first

conception meaning belongs to the

of physiological reactions of

of the second

Yet

light

in

it

human

organism, in the light

belongs to the sphere of psychological experience.

both cases the sign

is

linked with meaning by the bonds

in

no way uniform, is however dommeaning with action

of a causal relation.

The whole conception,


inated by the pragmatist
and of seeking meaning in

ideas of linking

action. Replying to F. C. S. Schiller's

objections at the 1920 symposium, Russell wrote:

"Meaning, in

my

view,

a characteristic of

is

'signs' are sensible (or imaginal)


47 Ibid., pp.

300-301.

48 Ibid., pp.

302-303

(italics

A.

'signs',

phenomena which cause


S.).

and

actions

The Meanings of "Meaning"

257

appropriate, not to themselves, but to something else with which

The

they are associated.

what

is

possibility of action with reference to

not sensibly present

is

one of those things that might

be held to characterize mind"49.


This point

is still

clearer in the case of

a disciple of Russell's,
latter
if

who

and continental philosophers; the

we take

Ludwig

Wittgenstein,

acted as intermediary between the


clarity

into account the evolution of his views

the greater

is

from Tractatus

Logico-Philosophicus to Philosophical Investigations (that book,

pubhshed posthumously, presents Wittgenstein's views as shaped


over a period of

some dozens of

In this Tractatus, he
images,

still

years).

represents the conception of meaning-

He

although even there different tunes can be heard.

says for instance:


"3.

262

appHcation.

What does not get expressed in the sign is shown by its


What the signs conceal, their apphcation declares"50.

In Philosophical Investigations, the identification of meaning

with the application of the sign

is

already effected consistently.

Wittgenstein even comes to the conclusion that in the words,


apart from the

way

in

which they are

used, there

is

nothing

else

that might be called meaning^i. His standpoint can be reduced


to the following:

"43 For a large class of cases

though

not for

all

which we employ the word 'meaning' it


meaning of a word is its use in the language"52,

can be defined thus

in

the

Wittgenstein does not repeat the ideas of the pragmatists

and the behaviourists

on the same plane as

in spite of the fact that his reasoning


theirs.

Meaning

is

and action, but an action of a particular kind, namely


which consists in using the sign in speech.
sign

49

Mind, October 1920, No. 116,

50

L. Wittgenstein, Tractatus Logico-Philosophicus,

51

L. Wittgenstein, Philosophical Investigations, Oxford

52 Ibid., p.

20.

is

the relation between


that

p. 402.

London

1933.

1953.

p.

3.

258

Now

Selected Problems of Semantics

the neo-positivist conception of meaning becomes fully

comprehensible against the background of

all

these views. It

for us a matter of secondary importance whether

we have

is

to

do with congeniality or merely with an ordinary reception of


views of other trends and other thinkers. We only wish to

and to show its variations in


them for a conception of the

high-light a certain standpoint,

order to extract what

is

typical in

meaning of "meaning".

Among

the former

members of

lem was analysed by, in

the Vienna Circle that prob-

Moritz Schlick and Rudolf

particular,

Carnap.
Schlick raised the problem of
Erkenntnislehre,

and

meaning

later reverted to

and papers published

it

in

his

Allgemeine

in the various articles

Gesammelte

in the collection

Aufsdtze.

"Meaning and
adopted by the Vienna Circle
as a whole, was that the meaning of a sign is the same as the
method of its verification. Hence a transition to the concept
of unsense that is, statements which have the grammatical
Verification"53.

I refer in particular to his article

Schlick's starting point, later

form of a sentence but are devoid of sense since they are unveriSchhck combined that typically operationist conception

fiable.

with the view of Wittgenstein


verbis) that the

(to

whom

he referred expressis

meanings of expressions are manifested in the way

which they are used in speech. That combination of the two


conceptions appears most clearly in the article "Meaning and
in

Verification",

mentioned above.

"Thus, whenever
it

mean?', what

we

we ask concerning
expect

is

a sentence, 'What does

instruction as to the circumstances

meaning of
a sentence amounts to stating the rules according to which the
sentence is to be used, and this is the same as stating the way in
which it can be verified (or falsified). The meaning of a proposiin wliich that sentence

tion is the
53 First

54 In

method of

is

to be used

...

Stating the

its verification''''^^.

published in Philosophical Review, Vol. 44, 1936.

M. SchUck, Gesammelte

Aufsdtze, 1926-1936, Vienna 1938, p. 340.

The Meanings of "Meaning"

The

situation

is

somewhat

259

different in the case of

Carnap.

He, too, had undergone a considerable evolution from the period


of logical syntax,

when he

in

meaning

asserted that a logic of

superfluous, to the period of semantics (Introduction

which he exphcitly renounced that

to

is

Semantics)

assertion. In the course of

that evolution the concept of "meaning" appeared in his works:


originally

he interpreted meaning exclusively as the designatum

name

of the

(cf.

The International

his introductory article to

Encyclopedia of Unified Science, entitled "Logical Foundations

of the Unity of Science"). Yet throughout that period Carnap

was

interested in something else,

namely in the meaningfulness

of sentences, understood as their capacity for verification in some


form. This

is

a fascinating problem in the evolution of neo-positiv-

ism, most clearly marked in the evolution of Carnap's views.


The requirement of a full verifiability of sentences was connected
with the standpoint on meaning of Wittgenstein^s and Schlick.
The requirement of falsification appeared under the influence of
Popper. Later on, Carnap adopted a more moderate position

and Meaning) in that he required only a gradual

{Testability

and not a

confirmation,

full

verification

or falsification.

But,

as indicated with reference to Bridgman's views, these are only

and not fully coincident.


a brief mention of the concept of meaning as

related matter,
Finally,

rep-

resented by Morris's semiotic.

So

far, I

have more than once indicated a similarity between

such and such views on the one hand and pragmatism and behav-

we have

do with an obvious connection with such views, since semiotic was born in the union of

iourism on the other; here

to

neo-positivism with pragmatism. This

which we have

emerge

in

previously

called

the

is

why

semiotic with particular clarity.

preted in terms of physiological reactions of

and the term


55 L.

itself

is

most

Wittgenstein. Tractatus

....

willingly

4.003.

those tendencies

biological

conception,

Meaning

human

avoided.

is

inter-

organism,

260

Selected Problems of Semantics

In referring to Morris's views,

it is

of the Theory of Signs^^ and those formulated

Language and Behaviour. In

Signs,

more modest work


in that

is

much

make a

necessary to

between the opinions contained in

distinction

my

on

later

clear

Foundations

his

in his

book

opinion, his earlier and

better than his system as developed

book. In Foundations, Morris

sets forth

a reasonable con-

ception of meaning. Stressing the ambiguity of that term, he

opposes

firmly

Platonizing

all

with his conception, meaning


process

of

semiosis

is

interpretations.

In conformity

a term that belongs to the social

("sign-situation"

terminology).

other

in

That term becomes comprehensible only

in the context of such

process and the terms related to the latter concept.

"Nothing
such only

is

intrinsically a sign or a sign vehicle,

is

so far as

something through

its

mediation. Meanings are not to be located

any place in the process of semiosis but are to be

as existences at

characterized in terms of
is

but becomes

permits something to take account of

it

this,

process as a whole.

'Meaning'

a semiotical term and a term in the thing-language; to say that

there are meanings in nature

of entities on a par with

is

not to afl&rm that there

trees, rocks,

a class

is

organisms, and colours,

but that such objects and properties function within processes

of semiosis" 57.

The aspect

quite different

is

damental work on signs

when

marked

it

comes

ficant results. I fully agree with the criticism

by

Max

Black^s

and

J.

to Morris's fun-

by vast claims and


of

its

insigni-

ideas as

made

Kotarbinska59.

Morris wanted to handle the problem of the sign from the


behaviourist point of view, and therefore postulated a complete

renunciation of the category of "meaning".


56

In International Encyclopedia of Unified Science, Vol.

versity of

Chicago

1,

No.

2,

Uni-

Press, 1938.

57 Ibid., p. 45.
58

and

M.

Black,

Pfii/oiopfiy,
59 J.

"The Semiotic of Charles Morris",

New York

in

M.

Black, Language

1949.

Kotarbinska, "Poj?cie znaku" [The Concept of Sign], in Studia

Logica, 1957, Vol.

6,

pp. 57-133.

The Meanings of "Meaning"

"The term 'meaning'

is

26]

among

not here included

the basic

terms of semiotic. This term, useful enough at the level of every-

day analysis, does not have the precision necessary for

scientific

Accounts of meaning usually throw a handful of putty


at the target of sign phenomena, while a technical semiotic must
provide us with words which are sharpened arrows. 'Meaning'
analysis.

signifies

any and

all

phases of sign-processes (the status of being

a sign, the interpretant, the fact of denoting, the significatum),

and frequently suggests mental and valuational processes as well;


hence

it is

desirable for semiotic to dispense with the term

to introduce special terms for the various factors

and

which meaning

to discriminate''^^.

fails

The declaration

is

reasonable. But the execution?

One has

to

agree with Black that nothing remains of that strident announce-

ment except the declaration itself. Morris replaces the term


"meaning" by the term "significatum" (in doing which he avails
himself of the fact that in English there are two synonymous
words, "to

mean" and

the conditions under

"to signify") as being the equivalent of

which a sign denotes something ^i, and then

distinguishes between the individual variations

ficatum" according to the dispositions of the


to react in definite

time to

It is

ways to preparatory

sum up what been

of that "signi-

human organism

stimuli.

said so far

and

to

draw more

general conclusions.

To avoid

my

misunderstandings,

must emphasize once more

been to show a certain type of solution


of the problem of meaning. Consequently, I do not claim that
the views described here belong to one and the same philosophical
that

trend;

sole object has

on the contrary, there are sometimes considerable difNor do I claim that the theory of meaning

ferences between them.


60

Ch. Morris, Signs, Language and Behavior,

61

Ibid., p.

New York

1946, p. 19.

18.

18

262

is

Selected Problems of Semantics

characteristic of such trends;

on the contrary,

it is

sometimes

opposition to other theories within the same system (as in the


case of neo-positivism). I only maintain that there

type of solution of the problem of meaning which

and

to various philosophical schools

a certain

is
is

common,

trends.

That solution was born of opposition to metaphysical conwhere meaning is transformed into

cepts of the Husserlian type,

a sort of ideal entity, or against mentaUstic or psychologistic'


concepts, which locate

meaning

man's

in

exclusively to psychological categories.

by the concept of meaning as a

and
form of a

spiritual life

and

resort

Such views are opposed;

specific relation

between the sigm

certain reflexes of organism, or a conscious reaction in the:|

evoked by the given

definite action

sign.

Thus, in the

of that conception, "meaning"' means the equivalent

light

a relation between the sign and the reaction

but with

briefly

less precision,

it

evokes, or, more!

of the action evoked by the sign.

Psychological categories are here replaced by categories of objective behaviour, the action

of

human

organism, and in this

way we

acquire objective data which enable us to define meanings and


differentiate

between them, data which can be observed

inter-

subjectively.

How
It

is

that conception to be evaluated?

has two unquestionable

virtues.

First,

the opposition to

Platonizing metaphysics and idealistic mentalism gives rise to


a tendency to interpret meaning as a relation sui generis. Second,

same source

the

gives rise to a tendency to interpret

in terms of objective reactions of

Yet
ories

and

it

their

action, tend, at least to

is

above

all

all,

these the-

foundation the relation between sign

some

extent, to fetishize the sign, as

indicated above. In other words, they

ing

meaning

organism or conscious action.

also has obvious shortcomings. First of

which take as

social

relation

fail

to notice that

between

mean-

men who

act

and who communicate with one another. Next, while developing


out of an opposition to a one-sided interpretation of meaning,
they

fall

off!

into the other extreme: they simplify the

problem

in the

'

The Meanings of "Meaning"


behaviourist spirit. There
by,

among

is

no doubt

263

has been pointed out

(as

others, Russell) that the behaviourist

when

completely

is

it

fails

and interpretation

abstract subjects, such as the understanding

of a philosophical treatise, a lyrical poem,

Of

approach

required to explain communication on

course, one might

draw attention

etc.

number of other

to a

shortcomings, inconsistencies or simple mistakes in the conceptions discussed above. But


criticizing

erroneous views

That

positive conception.

4.

One of

the

is

it

seems that the best way of

to formulate a

I shall

now

MEANING AS A RELATION

ways of

interpreting

more

satisfactory

attempt.

(2)

"meaning"

is

to conceive

of meaning as of a specific relation between the persons

communicate with one another.


I

It is

who

within this conception that

intend to expound the Marxist standpoint on the issue under

discussion.

My

declaration requires certain explanation in order to avoid

misunderstandings:

have to explain what

expound the Marxist standpoint."


It might mean, first, that the author

is

is

meant here by "to

going to present the

opinions of the Marxist classics on the given subject, drawing

on the appropriate

writings. This

is

not the point in this case since

no theory of meaning is to be found in the works of the Marxist


classics. They simply did not tackle the subject, except to the
extent of chance remarks on language and meaning.
Alternatively it might mean that the author wants to subject the
problem concerned to an analysis from the Marxist point of

making use of the Marxist method. This is in fact my intention and this is how my declaration is to be understood. But
from this definite deductions can be made. The study of a problem
from a certain methodological standpoint does not imply a moview,

nopoly of correct solutions. Not only because a researcher

may

264

Selected Problems of Semantics

commit an

error

and be mistaken

same method and

different people, in using the

same

theoretical

different,

The

may

assumptions,

and even

some

in

results of research

in his analysis, but also because

from the

starting

in concrete

matters reach

respects contradictory, conclusions.

and the conclusions drawn from research

are determined not only by the methodological

and

theoretical

assumptions made, but also by knowledge concerning the prob-

lem involved, general knowledge which forms the background


of the problem under investigation, individual talent for research,
creative inventiveness, etc.

Be

that as

it

may, from the

fact that

two persons adopt similar basic assumptions, that they speak


in the same manner, it does not follow that they must always
say the same thing, that is, come to identical research results.
This refers in particular to such
controversial issues as

is

difiicult, intricate

and extremely

the problem of meaning. Consequently,

by declaring that I want to expound the Marxist point of view


on that issue I wish to say only this that I start from Marxist
assumptions. I do not in any wise claim that what

be "authentic" Marxism and that every Marxist

me on

I shall

say will

at variance

that subject should be anathematized. I find

it

the

with

more

necessary to stress that point because the literature of the subject

based

on Marxist

studies, since

principles

(primarily

theoretical

hnguistic

Marxist philosophical studies on language, and in

on meaning and the theory of signs, are very scanty)


I would fully endorse and on
the other hand includes items from which I should like vigorously
to dissociate myself. Hence, I have to admit that the problem
particular

does not include a single item which

is

controversial and to present

my own

view as one of the pos-

sible solutions.

A. Meaning as a relation between men who communicate with


one another
I

have often repeated the thesis that

all effective

analysis

of the signs and meaning should start from an analysis of the

265

The Meanings of "Meaning"

of communication or, in other words of sign-

social process

I shall begin from that point.


The problem of meaning appears wherever we have to do with

Consequently,

situation.

human communication. In this sense,


relation between men who communicate

signs in the process of

meaning

is

a definite

with one another.

we

are concerned

It is that meaning of "meaning" with which


now; other meanings of that term are outside

the scope of our interest.

meant when we say that meaning is a definite social


relation? This means more or less: someone wants to incite
someone else to action, to inform him about his thoughts, feelings, etc., and with that end in view resorts to a sign a gesture,

What

is

a word, an image,
the

if

i.e.,

etc. If

the intended

appropriate thoughts have

to the other party (as

eff'ect

has been achieved,

fact

in

been conveyed

can be inferred from the reply or other

behaviour of that other party), then

we

say that the meaning of

the sign has been understood by the hearer.

What we

call

meaning

a complicated social process takes place, a process

appears where
which we have referred to above in analysing the concept of

The following elements are indispensable for


two persons (classes of persons) who communicate with one another, that is, who think;
(2) that something to which the sign refers; (3) the sign by means

sign-situation.

the occurrence of that process: (1)

of wliich thoughts are conveyed. But material objects or events

become
with
are

signs only

men who

somehow

of signs,

when they
them as

use

enter into definite intricate relations


signs;

with reality to which they

referred (as names, pictures, etc.); with the system

language, within which they function.

i.e.,

in such a context that

It is

only

an object or event becomes a sign or, in


if we do not believe in the

other words, has a meaning. Thus,

mysticism of inherent meanings, meanings as ideal

which

their

material

admit that for

all

vehicles

participate,

it

is

entities,

we have

in

to

the ambiguity of the term "meaning" (even in

the narrower sense of the word, with which


here),

then

we

are concerned

always a certain system of social relations which

is

266

Selected Problems of Semantics

human

involved. Similarly,
it

is

cognition

is

a social relation, since

by

relation betv^een the cognizing subject (but shaped

in society)

life

and the object of cognition; what we call "reflection"


of cognition), etc. The question practically consists
appreciation of what relation it is which is called mean-

(in the theory

in a better
ing, or

what system of

There

relations.

are specific relations

men and

between

another;

between signs and

reality ;

the elements of the

all

men who communicate one with


reality; between men and signs;

between the

sign-situation:

between

between signs and other signs in a

cer-

tain system of signs. These are relations of various types, situated,

as

were, on various planes, above

it

plane and on the plane of


is

human

always human communication which

It is

all

on the psychological
and actions. But it

attitudes
is

decisive in such relations.

always in that context that the sign-situation, the sign and

meaning are comprehensible. The separation of a fragment


of such relations (the relation between the sign and its designatum
the relation between signs; the relation between the producer

of signs and the signs,


poses and

is

treat such a

as

we have

necessary for research purit is

not permissible to

fragment as an independent whole, since

this involves,

seen above, the danger of various "fetishisms".

Thus, meaning

between

may be

etc.)

of course permissible; but

men on

first,

is,

a relation or a system of relations

the psychological plane.

We may

also refer to

a psychological aspect of meaning, namely the relation between

men who
that

is

which

act

and

feel,

refer to

some

A closer explanation
of

its

and who communicate with one another,

comprehendingly convey to one another their thoughts

elements,

reality in the broadest sense

of the term.

of the sense of that relation, and an analysis

would require volumes. For

virtually

each of

these elements could be the subject matter of a separate

mono-

graph: beginning with the issue of a social individual,

human

individual shaped in society,

action and thought,

is

at

whose every

step,

i.e.,

both as regards

once individual and social; through

the problem of reality as the subject of cognition

and communi-

The Meanings of "Meaning"

267

cation through the issue of the sign as the intermediary in that


process; to the problem of the interpretation of the process of

comprehending and communication

individual and social


can be seen that we touch here upon extremely
complicated problems which moreover are inseparably inter-

psychology.

in

It

we must of

connected. Thus

necessity use concepts

which are

move with apparent

not explained to the very end, and must

on very slippery ground.


With this reservation in mind, we may say that, on the
psychological plane, by meaning we understand that which
ease

perform the function of an intermediary

enables the sign to

human communication,

in the process of

that

is,

in the process

of conveying thoughts from one person to another. That "something"

an

is

intricate

system of interhuman relations whereby

a material object becomes a sign.


further be

made

An

exphcit reservation must

here to the effect that

origin of that relation, nor explain

how

we
it

neither refer to the

happens that material

objects and events can perform, under definite conditions, such


comphcated cognitive and communicative functions; we merely
state certain facts.

A
of

quite different view point

human

actions

is

made

and behaviour. Here,

possible by the analysis


too,

we

are concerned

with definite relations between men, the same as those referred


to above, but seen in the aspect of objective

human behaviour

in practical, mental, etc., action.

In other words,
to the sphere of

For in

it

human

might be said that meaning belongs both


action and to the sphere of human thought.

two spheres are inseparably connected one with


another. If we notice these two aspects of the meaning relation,
we are protected against mentahsm, which is nothing more nor
less

fact these

than ascribing absolute importance to one aspect of the sign-

situation.

In an objective interpretation, mentalism leads to the

conception of meanings as ideal entities


tation,

it

in a subjective interpre-

leads to psychologism, which believes

subjective property of

autonomous mental

meaning to be

processes.

268

Selected Problems of Semantics

To

my

treat

opinion,

meaning as
is

definite

interhuman relations (which,

in

extremely important for a correct analysis of the

way

complicated issue) in no

protects us against the ambiguity

from that

of the term. For, starting

theoretical assumption, one

may by "meaning" understand either the whole of the relations


which make up the sign-situation (the process of semiosis),
or a fragment of those relations (the relation between the sign

on the one hand and the

object or the thought about the object

the other), or the designatum or the denotatum of the sign

on

(i.e.,

an object of communication regardless of whether it actually


exists or not, or such an object existing in reality), or the relation
between the sign and the system of signs (language) or between
the sign and the signs or another language, etc. In the literature

of the subject

all

these meanings of "meaning" appear in

or less pure form. In our analysis, too,


tion

to

an

ambiguous

so since

we

this

in

the term "meaning". There

is

sense

we might draw
of the

word

nothing detrimental in

are here concerned not with a

it,

more
atten-

use

the

pedantic

of

more

distinc-

meaning of "meaning",
but rather with a theoretical view of the foundation on which all
these meanings rest. Now, to explain that aspect of the problem
requires the adoption of some definite standpoint on the issue of the
tion between the various shades of the

origin of meaning, that

is,

the origin of that specific property

which transforms material objects and events into

signs, turning

and events into extremely important intermediaries


in the communication of men among themselves and if it may
be so expressed with themselves (since a mental monologue
is but a form of a dialogue).
these objects

B. The origin of meaning

The problem of the


tributing to that origin,

origin of meaning, of the factors conis

of the thesis that meaning

important not only for the explanation


is

a definite social relation

it is

equally

important for a proper interpretation and solution of the very


difficult issue

of the relation between meaning and notion.

The Meanings of "Meaning"

is

269

This reveals the third plane

on which the problem of meaning

namely the

logical plane. In this connection

to be examined,

must

emphasize that both the above approach to the problem

my

and

among

logicians.

Objec-

which might possibly be raised on that score would not be

tions

justified, since

as that of

This

meaning deviate

further explanations of the origin of

considerably from customs prevailing

a one-sided formal logical analysis of such problems

meaning

is testified

is

certainly detrimental to the issue at stake.

by such

authorities fully recognized in logic circles

and Wittgenstein.
In his well-known work published in 1919, "On Propositions:
What They Are and How They Mean," Russell thus assessed
as Russell

the contribution of logicians to the analysis of meaning:

"Logicians, so far as I know, have done very

towards

little

explaining the nature of the relation called 'meaning', nor are

they to blame in

psychology"

Many

this,

since the

problem

one for

essentially

is

62.

years later,

when

his life

was already drawing

to a close,

Wittgenstein wrote:

"23

compare the multipHcity of the tools


ways
and of the
they are used, the multiplicity of
kinds of word and sentence, with what logicians have said about
It is interesting to

in language

the structure of language, (including the author of the Tractatus


Logico-Philosophicus)"^^.

two statements have in common is the understanding of the fact that the approach to the problems of language,

What

meaning,

these

etc.,

exclusively

from the point of view of formal logic,


Anyhow it must be said that a pro-

impoverishes the entire issue.


test against

psychologism

of men's psychic

life

a separation of issues
tations in

62

human

may

not lead to separating the issues

from psychology. Far


of social life from that

may

less
life

and

action.

In B, Russell, Logic and Knowledge,

p. 290.

*3 L. Wittgenstein, Philosophical Investigations, p.

12.

it

its

lead to

manifes-

270

Selected Problems of Semantics


\

In asking about the origin of meaning, we ask about the

which appear

origin of those interhuman relations

munication process, we ask

how

it

and events perform the function of


emotions,

This

etc.

is

mological question;

of the logicians
or take

sign without

meaning?
out

first

it is

who

first

And

vehicles of

human

com-

thoughts,

par excellence a psychological and

episte-

in principle outside the sphere of interest

either abstract

from the problem of meaning

how can one

as given. But

it

in the

happens that material objects

develop the theory of the

answering the question, what are sign and

in turn,

how can one answer

investigating the origin of sign

that question with-

and meaning, that

is

without tackling problems belonging to epistemology, psychology

and sociology? Of course, not every discipline concerned with


signs and meanings must handle these problems in all their aspects,
but no discipline, including formal

comprehension of such problems

logic,

if it

may

aspire to a full

passes over in silence the

fundamental questions.
In reply to the question posed above (what are sign and meaning?)

it

must be made clear that the

origin of

meaning

is

connected

with the social practical activity of men, as shaped in history,


activity

an inseparable part of which

Let us recall here that

is

the process of thinking.

said in the preceding chapter concerning

verbal sign to other types of signs

meaning

what has been

in conformity with

the

when

relation

of the

refer simply to

have in mind the meaning of the phonic language

and of the verbal

signs,

whereas the meanings of other signs are

treated as derivative with respect to signs of the phonic language,


as specific translations of the latter. In view of the inseparable,

organic links between language and thinking, the question as


to the origin of

meaning

tion as to the origin of

question

able to raise
in,

intimately connected with the quesprocesses. But then that

in all the issues of epistemology

lets

aspects of social

ed

is

human thought

all

life.

Here again

must

and the

related

regret that I shall not be

the issues involved by the problem

we

are interest-

but shall have to confine myself to the principal ones.

The Meanings of "Meaning"

When men communicate

one with another by means of

signs,

sign-situation consists in the fact that the cognitive content

tlie

which

so to speak, the private property of a thinking indi-

is,

vidual,
is

271

becomes

socially

communicable by means of

signs.

This

so because the signs in question are similarly understood on

a social level.

The mechanism of that "being

similarly understood"

consists in that the sign (above all the verbal sign)


(the nature of that connection will
level

connected

is

be discussed below) on a social

with similar thought processes and with similar reactions

in the

form of action

(these are the psychological

ioural aspects of meaning). Thus,

the origin of

standing

we may

meaning and about the

(interpretation)

and the behav-

equally well ask about

origin of a similar under-

These are in

of the sign.

two

fact

approaches to the same problem.

Meaning and

and consequently simthought processes (but not only such the same apphes

ilar

similar

to

similar understanding,

reaction in behaviour),

are intercoimected.

shall

ignore the nature of that connection and shall concentrate on the

By what

actual similarity of thought processes.

by what

is

it

is it

evoked, and

conditioned? Certainly by a considerable number

of factors, including the relation: the cognizing subject


object

the sign, which

relation since

between

its

it

comes to the forefront. It is an intricate


complex of relations taking place

various elements: the subject

The meaning of
(the

the

in fact a

is

the object, the subject

tive

thought, which

the sign
is

term "reflection"

the object, the sign

the sign.

cannot be separated from cogni-

a subjective reflection of objective reahty


is

used here in

sense, not here discussed). In a sense,

its

specific philosophical

meaning and the

reflection

of the object in thought can be said to coincide.

Does

all this

agree with the thesis that meaning

is

a relation?

Certainly. It has already been stated that reflection in the sense

of a (cognitive) act

Thus, meaning

is

is

specific relation also.

moulded

objective reality in the

in the process of reflection of the

human mind. But

here again

we

are not

272

Selected Problems of Semantics

concerned with any simple, one-directional relation, for meaning


is

in a sense a

human mind
time

product of the cognitive process of reflection in the

of the object to which the sign

refers.

But

same

at the

an element, and even an instrument, of that process, since

it is

without the sign there

is

not only no communication, but no

process of thinking and cognition in general (in particular, because

without the sign, and especially the verbal sign,


to attain to that level of generalization
is

it is

impossible

and abstraction which

necessary for thinking in terms of ideas).

Thus

at the root of the origin of

meaning there

same

the

is

process of a generalizing reflection of reality 64, as in the case

of thinking in terms of ideas. As in the case of the cognitive


process, the process of reflection, the practical experience of human
social history rests at the root of the origin of

Man
forming

it.

This thesis, which

may

Marxist epistemology,
that if

we

accept

appear

trivial.

"trifles" as

But

philosophy.

process of

From the
human

human

us bear in

mind

consciousness. These

point of view of Marxist philosophy,

its

When we have

all)

to

an

its origin, its

goal,

truth ahke. Consequently, that philosophy

new viewpoint on

(verbal signs above

it is

at the root of the

is

cognition, in the sense of

criterion of

acquires a

the origin of the meaning of signs

and discovers a new aspect of such

do with a name of an empty class, that

has a designatum, but no real denotatum,

such imaginary objects as fauns,


perties,

let

which are infrequently to be met with in contemporary

uncontestable fact that practical experience

64

trans-

the existence of material reality

reflection of that reahty in

are theses

and the

and

the principal assumption of

is

then that implies the recognition of at least

it,

two such philosophical


and the

meaning.

acquires knowledge of reality by influencing

e.g.,

when

devils, centaurs, etc.,

such as heroism, here too there

is

is

reference

origin,

name that
made to

is

or to abstract pro-

a reflection of reality, although

an indirect way. Ideas of imaginary objects consist of fragments of reality,


and abstract properties are reflections of properties, relations, attitudes,

in

behaviour,

etc.,

that are

common

to elements of a set of objects,

vidual material objects (see pp. 219220).

i.e..

indi-

The Meanings of "Meaning"

form of cognitive and semantic

Intricate social conditions in the

on human

processes are based

273

social practice and, conditioned

by that practice in a multiple manner, develop on

When

its basis.

practice intervenes in semantic relations,

it

accom-

is

panied by the historical element as one of the important factors


of the origin of those relations.

which

a problem

who

pologists

is

We

are here concerned with

known

perfectly well

and anthro-

to linguists

study linguistic meaning. If meaning

is

genetically

conditioned by social practice, then historical changes in that


practice

must have repercussions

which

conditions. This

it

studies

is

which often base on

in the field of semantic relations

confirmed by linguistic semasiological


this fact their hypotheses

the evolution of linguistic meanings (this

is

concerning

especially true of the

sociologically-minded trends, such as Meillet's school). This

is

also a natural explanation of the thesis concerning the national


specific

features

of languages, which refers above

all

to their

semantic aspect in the broad, linguistic sense of the term.

Only such a genetic approach to the problem of meaning,


which enables us to interpret

it

as specific interhuman relations

human

giving rise to the reflection of the objective reahty in

human

minds, conditioned by practical


sible to solve the

activity,

makes

pos-

question of the arbitrary nature of the sign and

of conventionalism in linguistic matters. The question


ing not only

it

from the

linguistic

is

interest-

but also from the philosophical

much

point of view. (The latter issue, which has wrought so

this refers above all


proves most telhngly

havoc in philosophy

to neo-positivism

broadly understood

that

aspect

may

the

genetic

not be disregarded in problems of meaning).

Attention was directed in the previous chapter to the problem

of the arbitrary nature of the sign.

De

Saussure's thesis that there

no natural hnk between the linguistic sign (including its semantic aspect) and the reality to which that sign refers, is, as we know,
is

in

conformity with the philosophical standpoint

the times
verbis

known

since

of Plato. That standpoint was supported expressis


in Capital in connection with the

by Marx when he wrote

Selected Problems of Semantics

^74
analysis of value:

from the quahties


nothing in

"The name of a thing is something distinctj


of that thing" ^5. But that concept has

common

with the view that language

with rules fixed by convention,

is

game

and consequently something

may be changed in an arbitrary fashion (reference is here


made to natural languages). This is understood by linguists
as well as by sociologists and psychologists who are interestec

that

in the study of language problems. I shall refer here to the opinion^

of the Soviet psychologist whose judgement in these matters


1

hold in very high esteem, namely

sign

fixed arbitrarily, the

is

S. L.

word has

its

Rubinshtein:

own

history,

"...

The

owing to

own life, independently of us" 66. De Saussure,


as a linguist who was interested in the study of actual social
conditions, also made an explicit reservation against interpret-

which

hves

it

its

ing his opinions in the spirit of conventionalism. Such an inter-

had

pretation

C.

its

say only as regard philosophy.

Meaning and notion


In order to understand well what

about meaning,

it

is

we mean when we speak

necessary to analyse the relation between

and that for two reasons. First, it will


help us to understand better the meaning of "meaning". Second,

the verbal sign and notion,

it

uncover certain deeply rooted idealistic views


between meaning and notion, views which, as we

will enable us to

on the

relation

shall see,

can be encountered even in Marxist

Theoretically, there are

the meaning of a

either

two

circles.

possible solutions to that problem:

word and a notion

are two different

phenomena between which such or other relations may exist, or


we have to do with the same phenomenon examined from dif-

65

K. Marx,

Moscow
66

C.

Capital,

Vol.

I,

Foreign

Languages Publishing House,

1954, p. 100.
JI.

Py6nHmTeHH, Ocnoebt
MocKBa 1946, p.

eral Psychology],

ooiiien nciixo.ioih'tiii

405.

[Principles of

Gen-

The Meanincjs

ferent aspects

The

first

oi

"Mkaning'"

275

and connected with the cognitive process

as a whole.

opinion prevails in the literature of the subject, including

Marxist works. In

my

opinion

is

it

defend the second

opinion and propose to devote this section to


Usually, the problem

and

essentially erroneous

based on a mystification. Consequently,

its

defence.

formulated thus: meaning

is a hnwhereas
guistic category,
notion belongs to the sphere of mental
processes, and as such to the field of interest of logic, psychology

is

and epistemology. Now, I claim that such a division of the field


which the categories of meaning and notion belong is er-

to

roneous, and that both meaning and notion belong


in

some

respect to

at

least

the field of thought (mental) processes

properly understood, which by no means excludes specific


guistic
It is

interest

in

the

also possible to find such formulations as: "the

expresses notion", "the

lin-

problem of meaning.

word

and meaning are interconnected,

the root of meaning", "notion

but they do not coincide",

etc.

word

realizes notion", "notion rests at

Now,

I assert that

such a

distinc-

meaning and notion, and even opposition of meaning


to notion, is pure speculation completely lacking any confirmation
in the sciences which investigate the corresponding mental and
linguistic processes, and is a tribute paid to the myths of Platonism

tion between

and nominahsm. That tribute


ophers

who

is

often paid even by those philos-

themselves issue a call to fight against idealistic

mythologization of the problem.

Are the opinions mentioned above


fact? Indeed, they are.

There

is

to

be encountered in

nothing particularly interesting

or strange in the fact that, for instance, the intuitionists sharply


its meaning to "true" cognition. Likeno reason to be surprised that Husserl separates
notion from meaning, and even opposes one to the other. Nor
do we find anything unexpected in, for example, Schlick's neonom-

oppose the word and

wise, there

is

inahsm. Let them so think; such are their views, and the conception of the relation between meaning and notion
tion

from others of

their conceptions.

is

no devia-

276

Selected Problems of Semantics

The

situation

is

different

when

comes to the Marxists.


some of them,

it

If the objections raised above also pertain to

then this
is

an

is

interesting fact and, to say the least, strange. It

worth explaining how people who declare themselves to be

adherents of dialectical materialism

come

ideahstic views in this or that version.

objectively to

And

claim that

expound
it is

so,

and consider the present controversy to be a "family quarrel".


Documentation to that controversy can be found e.g., in Zvegintsev's Semasiologia^'^

who

salonitsky68,

and

in articles

by Kovtun and

the various

relate

S.

A. Fes-

opinions held by Soviet

authors on the issue. I wish to emphasize that (except for a few


somewhat secondary points) I fully agree both as regards
his criticism and the formulation of his own views with the
opinion of P. S. Popov, expounded lucidly in his article "Znacheniye slova

poniatiye"^^.

fact, when
we speak of meaning and notion, we speak of different phenomena and categories, phenomena and categories belonging
to different spheres of human activity? Such an idea is suggested

Let us begin with the question as to whether in

by the majority of the authors mentioned

in this connection.

Let us take Zvegintsev as an example:


"In

all

those cases

when

the sign of equality

is

placed between

notion as a logical category and meaning, words are deprived

of

all

those specific characteristics which transform them into

elements of language

...

"As has been pointed out already, the forms of the mutual
relations between the two phenomena, closely inter-connected
notion and word should be understood as the relation
67

See

68 JI.

e.g.

C.

pp. 110-112, 138 et passim.

KoBTyH, 0 SHaneHHH cjiosa" [On

Bonpocbi H3biK03HanuH 1955,

JVe

5;

Meanings of Words]

C. A. OeccAJioHHUKHii,

in

"053op HHxe-

parypbi no BonpocaM cbhsh asbiKa h MbimjieHHa" [A Review of the Literature Concerning the Connections Between Language and Thinking], in BonpOCbl H3blK03HaHUH,
69

1953,

JV"o

Bonpocbi .H3biK03HauuH

3).

1956,

JVs

6.

The Meanings of "Meaning"

277

of mutual influence, and not as one functioning instead


in place of, the other. It

precisely in the

is

of,

or

mutual influence of

word and notion that the formation and development of both


notion and the meaning of the word takes place, but each of these
phenomena is subordinated to its own regularities, follows its
own path, so that they carmot be examined as equivalent phenomena" 70.
I pronounce or hear pronounced the word "horse" and at
the same time I experience a thought connected with that fact.
They say that when I experience that act of thought I understand the meaning of the word "horse", but that there also appears in

it

As

the notion "horse".

the above quotation

from

Zvegintsev's work shows, there are authors (probably the major-

who assert in addition that in such cases we have to do with


phenomena which, although somehow interconnected, are different one from another.
ity)

Wliile

still

for the time

being leaving undecided the issue

of the identity of the meaning of the


state that in

word and

relationships occurring in the cognitive process

human

Any
as

and

we may
intricate

in the

com-

But these relationships are always con-

munication process.
nected with

notion,

both interpretations we have to do with

psychic

life,

they are cognitive relations.

attempt to overstep the framework of such relations leads

was the case of Husserl

directly to the metaphysics

of

Platonic ideal entities. Consequently, whoever chooses to avoid

Platonism must ask himself: what


tion

and meaning

if in

of reflection by the
signs? Before

we

is

mind of

the objective reality by

pass to conclusions,

the thesis that notion

the difference between no-

both cases we have to do with the relation

let

us hear those

and meaning are

means of

who

defend

different categories.

There are two principal variations of their argumentation.


Starting

from apparently opposing positions they are

based on a

70

common

B. A. 3BernHueB,

in fact

foundation.

CeMacuoAozuH [SemasiologyJ,

p.

142.

19

Selected Problems ok Semantics

278

Such authors as Zvegintsev

in his

Semasiologia quoted above,

or Gorsky in his articles published in Voprosy

assert

filosofii'^^,

that the difference consists in the fact that a scientific notion


is

ordinary (universally accepted) meaning (of

richer than the

words) since

it

contains

the essential traits of

all

which

including the regularities

designata,

its

govern such designata. They

do not deny that ordinary meanings of words coincide with


ordinary (every-day) notions, but the existence oi scientific notions

and

eliminates, in their eyes, the possibility of treating notions

meanings as equivalent categories. The adherents of similar opinion


occasionally add that words live long and their meanings (they
claim) do not change, whereas notions are changeable and
as

richer

and the world

science

found

criticism of such opinions can be

quoted above;

fully agree

with

Kovtun's

in

fundamental

his

grow

knowledge develop.

scientific

article

arguments.

Thus the argumentation of those who would separate notions from meanings is based principally on the claim that in
each of these phenomena we have to do with some other cognitive content. In conformity with the reasoning adduced above,
that content

would be broader

in the case of notions. Let this

be illustrated by a quotation from Semasiologia


"If

we

means

ing of a word, this

of a word

identify notion as a logical category with the

is

that

we assume

and

reflected the entirety of general

mean-

meaning

that in the

essential traits

of a definite class of object, with due consideration of the


cate connections of,

and

tablished by science at a given level of


''I

]X-

n. ropcKiiH,

its

1952,

Role of Language

Xs
in

4,

and

"O

po.nH

of

a3biKa

in

no3HaHHM"

now

[f this

H paSBHTMH nOHTHw"".

Concepts],

the Process of Cognition], ibid.,

be added that Gorsky has changed his opinion and

my

development,

"K Bonpocy 06 06pa30BaHHH

[Concerning the Formation and Development

^uAoco^uu,

intri-

relations between, these traits, as es-

1953,

JnTq

2.

Bonpocbi

[On
It

the

must

defends the view, which

a correct one, that the meaning of a word coincides with the


[The Role
content of the concept; see his "Pojib ssbiKa b no3HaHHH"
in

eyes

is

of Language in the Process


u .H3biK

of

Cognition],

in

the collection

[Thinkmg and Language], MocKsa 1957, pp.

82,

85.

Mbttu.ieuiie

The Meanings of "Meaning"

is

279

to a certain extent justified with respect to scientific terms,

which by means of words appearing here


mathematical or chemical formulae
generalizations,

it

fix

in the capacity

of

the results of scientific

certainly cannot be applied to

words of ordin-

ary language. But scientific terms have a limited range of use

and often never leave a very narrow

of

circle

specialists.

It

is

therefore quite clear that the mutual interaction between notion

and the meaning of a word should not be investigated by


the example of terms" 72.

ref-

erence to
It

is

obvious that there

is

an

difference between

essential

an every-day, colloquial meaning, or a manner of understanding


of the word "horse", and a
ing notion.

Anyone can

scientific definition

easily

ring in ordinary language dictionaries

paedias the same item

Probably only

make

this

unjustified

compared with

of the correspond-

check that for himself by

and

refer-

in scientific encyclo-

"horse". But what are the consequences?

that

we

wrong

are referring to

comparisons.

notions

Scientific

places

and

should

be

terms and their meanings, and words

scientific

used in ordinary language and their meanings, with their cor-

responding ordinary notions.

We

shall not here

concerning the definition of notion, but

it

go into

details

self-evident that

is

we reject the existence of notions as certain ideal entities,


we must seek them in the field of reflection by human mind
of the objective reality; but we have to bear in mind that it is
since

then

a specific reflection, inseparably connected with the verbal sign

(anyhow

for Marxists,

and

also for the

overwhelming majority

of linguists, notions without words are merely products of imagination).


different

There are different acts of


notions.

it

and

reflecting

it

context of that reflection.


geon,

"^2

when he

and there are

also

These differences are conditioned not only

describes

which determines our way


minds but also by the social
Even a most eminent veterinary sur-

by our knowledge of the world


of seeing

reflection,

in our

military

parade involving horses,

B. A. 3BerHHueB, CeMacuojioeun [Semasiology], p.

143.

280

Selected Problems of Semantics

does not, at that moment, have in mind a


horse and does not develop in his mind

scientific

horse,

its

diseases, its

anatomy,

when lecturing on thei

tions.

Thus, there are ordinary;

etc.

common) and

notions (in the sense of being very

notion of the

the essential character-

of that animal, as he undoubtedly does

istics

of

all

no-

scientific

There are also ordinary meanings of words and meanings]

scientific

terms.

It

is

and

obvious

easily

comprehensible;

and

that scientific notions differ by the breadth, depth

preci-

sion with which they reflect reahty from those contents which

appear in ordinary meanings of words. But in no case


sible to find essential differences

between a

scientific

is it

pos-

notion and

the meaning of a scientific term.

The second type of argumentation


between notion and meaning

in favour of the diff"erence

as indicated above, from:

starts,

an opposite position: its adherents claim that the meaning ofi


a word is broader than notion since it includes emotional, aesthetic, etc., elements which have no part in notion ^3. Here, too,
scientific notion, a product of scientific abstraction, is compared
with ordinary meaning of a word or even with an entirety of
psychic processes taking place in a person's

with a given statement, with the


etc.,

entire

who

context of such processes. Those

behave

like

people of

whom Marx

an abstract notion of a

in

etc.

aesthetic,

defend such a view

notion which disregards

fruit in general, a

it,

in connection

once said that having formed

the concrete shapes of an apple, a pear,


that fruit-in-general, taste

mind

emotional,

etc.,

would

scientific

like to

notion

touch

in fact

is

a sense semantically broader, and in a sense semantically

narrower than an ordinary meaning of a word. But


repeat once

more

the

issue

evaporates

if

let

we compare

us

ho-

73 That opinion is defended e.g. by B. M. BorycjiaBCKHH in his work "CjioBo H noHHTHe " [Word and Concept] in the collection Mbiiujieuue u ft3biK
[Thinking and Language], pp. 245 ff. It is worth while mentioning that Bogu-

slavsky also accepts the view that a concept

is

extensionally broader than the

meaning of a word, and uses these arguments to show that a concept and the
meaning of a word are two different categories, belonging to various fields.

The Meanings of "Meaning"

281

mogeneous categories: a scientific notion witli a scientific term


and on ordinary meaning of a word with an ordinary notion.

The argument

that words have a long life and have unchangwhereas notions are changeable and continue

meanings,

ing

to develop, is equally groundless, for

to be

proved

assumes what has

it

Moreover, such an assumption

a word.

still

a difference between notion and the meaning of

and unacceptable

for, e.g., a linguist,

meanings of words change as

much

as notions.

changes and the law which governs them

Thus the argumentation which

notoriously wrong

is

who knows

is

very well that

The study of such

daily bread to him.

have found in the

literature

of the subject in favour of the thesis concerning the difference

between notions and meanings of words turns out to be com-

would be difficult to accept conviction that noand meanings of words belong to different fields and form
different categories. Let us now see what are the philosophical

pletely futile. It

tions

consequences of the thesis which


It

word

ing of a

by statements

am

criticizing.

are different categories in usually accompanied


like

"The word

expresses the notion",


etc.

has been said before that view that notion and the mean-

"Notion

realizes the notion",


lies

at

the root

"The word

of meaning",

Dozens of examples might be adduced to support that point


contained in the work by P. S. Popov, quoted

(the criticism

above,

is

Now,

extremely instructive in that respect).

to resist the impression that the authors of such

it

is

hard

and similar

formulations pay, whether they want to or not, a double ransom

on the one hand to


idealism,

to substantiate

stated above, the

definite

and related objective

and on the other (indirectly), to nominalism. I shall

now endeavour
As

linguistic hypostases

my

opinion.

meaning of a word

interhuman cognitive

relations,

is

tantamount to

inseparably connected

with the act of cognition, and therefore with a psychic act. I say

"There are horses over there" and


in the
I

same way

understand that sentence

as does the person I address: in other words.

experience the understanding of the meanings of those words.

Selected Problems of Semantics

282

What

implied, then,

is

if e.g.,

word "horses"

the

realizes or ex-

presses the notion "horses", that that notion rests at the root

of the meaning of that word, etc.? The sense of such a statement


is

miivocal: apart from meaning as a cognitive relation, which

always appears in some and someone's psychic

acts, there is also

namely notion, which exists before meaning and


independently of meaning, since otherwise it could not "rest at

such an

entity,

the root, "be realized,"


idealism of the

beyond

water, and

first

The opinion
is

"be expressed,"

that

meaning

an objective category,

etc.

objective

is

|_

help.

a subjective,

is

This

and that notion

very popular, and due above

is

all

to

logicians.

had enough courage

Russell

to admit to Platonizing in his

mathematical and logical conceptions based on such "entities"


as

numbers,

relations,

classes,

Not

etc.

all

courage, but the influence of Platonism

logicians

share his

widespread in con-

is

temporary logic and the study of the foundations of mathematics.


This is due to two causes. One of those causes is the succumbing

and warnings)

(regardless of all declarations

that

ses,

(e.g.,

common

the

is

to linguistic hyposta-

suggestion that where there

must be a

"class", "notion", etc.) there

ed by that name. The other

suggestion due to one's

is

is

name

real entity designat-

own mental

constructions, a state of being entangled in the meshes of one's

own
I

abstraction process.
shall begin

not

of judgement:

as

(judgements

the

Like

the

v/ith notion,

because

issue (perhaps

in

it

acts

process

is

less

but with the

of judgement,

logical

more

still

glaring

popularized) of the two forms

and

propositions

as

sense).

of thinking,

the

communication process

takes place not through detached, isolated words, but through


sentences.

Certainly,

in

e very-day

speech we do

occasionally

resort to single words, e.g., "Attention!", "Avalanche!", "Lord!",


etc.,

but these are obviously undeveloped, abbreviated forms of

sentences (in other words, sentence equivalents).


too, serves the purpose of expressing thoughts,

The
and

sentence,
it

is

only

The Meanings of "Meaning"

283

words acquire meaning by estabone of the many meanings connected

in the context ofa sentence that

given case,

lishing, in the

with a given sound. There

is

of course the traditional grammatical

formulation that the sentence consists of words and the asser-

many

tion of

notions,

but

language

(e.g.

logicians that

the sentence

is

a relation between

Wundt) and philosophers of


Marty, Mcinong) have long since come to consider
psychologists

(e.g.

and propositions as the two fundamental

sentences

units

of

language and thought, and consequently they explain words and

them

notions by placing

in the contexts of sentences.

We

utter a

for instance "There are horses over there"


and at
same time we think that there are horses over there we thus

sentence
the

Some

experience an appropriate act of judgement.


that there

ment an

also a proposition (in the

is

sich) the expression

What

sense

logical

a judge-

the sentence.

is

the origin of that judgement in the logical sense

is

The need

proposition)?

(i.e.,

of which

authors assert

to

what Peirce

interpret

"token" and "type" in inscriptions.

Now

written: "There are horses over there".

here

called

the sentence

is

read that inscription,

"consume"

experience an appropriate judgement and thereby

that inscription as a "token" (in Carnap's terminology, "signevent").

But the inscription remains, and with

possibihty of

its

being "consumed"

as a "type" (Carnap:

the fact

many

"sign-design"),

times.

which

that whenever someone pronounces

is
it

it

It

remains the
then appears

characterized by

comprehendingly,

he thereby experiences a definite judgement, namely, that there

Each individual case of "consumption"


combined with the individual specific ex-

are horses over there.

of that inscription

is

perience of a given judgement; the experience of a veterinary

surgeon will

differ

experience of a
etc.

from

that of a municipal tax-collector,

and the

cowboy will differ from that of a horse-racing fan.


on the type and scope of knowledge about

All this depends

horses,

individual past experience, emotional associations, and

the like.

Yet

to

all

in all

such cases there

is

some content

that

is

common

those different individual judgements, something which

284

Selected Problems of SEMA>rncs

accounts for the fact that every person

who knows

stands the meaning of the sentence in question.

English under-

Why

that so?

is

There are two explanations possible. One of them simply

form of experiencing the meaning of the words spoken

to a similar

out in a given sentence context, characteristic of

who know
from

the language involved. That

brain, etc. But there

changing

the

all

those persons

phenomenon

is

in

were, the standard of

ment. This

is

also the other explanation:

which

itself,

all

there is some judgesome entity and is, as it


and experiences of that judge-

utterances

is

metaphysics of the Platonic type, as endorsed by

is

many. There are mathematicians who claim that


sible to build the set theory

as real ideal entities,

it

be possible to prove

and there are

(in

my

logicians

who maintain

entity

It

opinion, quite irrefutably) that

here to do with a mystification, that one's

own

acts of judgement,

is

exist

that

would

we have

logical construc-

by abstraction of those elements that are

tion, obtained

not pos-

is

without admitting that classes

propositions as real ideal entities are indispensable.

many

of the

apart from

judgements,

individual

ment immutable

to

explained

various aspects by grammar, the physiology

its

refers

common

transformed into an independent

which moreover appears as the standard of those actual

acts of judgement.

But

all this

does not seem adequately con-

vincing, so that the unconvinced include even adherents of

ma-

terialism.

Now

the situation

is

analogous in the case of word and no-

tion, the only difference being that notion stealthily

both

roles,

which in the case of the

ment are distinguished


notion, too,

it

is

pair: proposition

Yet

terminologically.

in

performs

judgecase of

the

possible to distinguish terminologically the act

of experiencing notion, and notion in the logical sense. The


latter is a

product of abstraction, a

construction
in reality:

reflecting

common

sometliing

specific cognitive

which

properties or traits,

idealist

objectively

common

characteristic of a given class of objects.

can be a notion. Only an

and

logical

appears,

regularities, etc.,

But no material

entity

of a peculiar kind, one

who

The Meanings of "Meaning"

hypostatizes a product of his

an independent
.

What then is

entity,

285

own mind and

can accept such an

the origin of such views

when

transforms

it

into

entity.

held by materialists.

and by Marxists in particular? They arise from the fear of a nominalist distortion

more detail.
As opposed

of the problem. This will

now be

explained in

to conceptual realism, the nominalistic doctrine

stated traditionally that general concepts are flatus vocis,

trary verbal constructions.

of ideal entities)
of

an

and

Both

arbi-

materialistic tendencies (negation

subjective-idealistic

objective correlate of notions,

e. g.,

tendencies (negation

in the case of Berkeley)

used to refer to nominalism. In the contemporary philosophical

neonominalism opened the door wide for the subjective-

context,
idealist

tendency. Hence, Marxist philosophers see in a nominahstic

interpretation of notions (e.g., in the interpretation of Schlick,

who simply

negates them) a danger of subjective idealism.

For

rightly so.

And

in that interpretation the cognitive relation of

reflection of the object

by the subject disappears, as

also does the

objective correlate of notion as a generahzing reflection of reality;

what remains is the flatus vocis and a subjective


of mental processes and their products.

inter-

pretation

In his endeavour to escape that danger, the materialistically-

minded philosopher occasionally


arms of conceptual realism.

falls

into

the

wide-spread

Consequently, the conclusions of these considerations concerning what

is

said to be a difference between

meaning and

notion are as follows: not only are there no positive arguments


in favour

of the conception that meaning and notion belong

to different fields, but even a negative

argument discloses

clearly

the idealistic consequences of such a conception.

Yet

its

defenders have in

lems which

we

mind

do. not eliminate

of such or other solutions. Is

it

certain real theoretical prob-

by proving the incorrectness


not possible, then, to suggest

a rational solution of that problem, a solution

dangers of ideahsm? In

my

opinion, this

is

free

possible.

from the

286

Selected Problems of Semantics

The
ogism

between

controversy
in

modern philosophy

and psychol-

anti-psychologism

focused around the interpreta-

is

tion of psychic processes and the products of such.

It is

a fact

that spiritual processes (we are here interested mainly in mental

processes) are always eigenpsychisch that

vidual.

Thus they belong

is,

are always "private",

and processes of a given

are always hic-et-nunc processes

to the sphere of psychic existence

indi-

and

are subject to psychological analysis without which they cannot

be understood. But at the same time, it is a fact that there is


something in those processes which goes beyond the strictly

men

"private" sphere, that

experience certain thoughts in a similar

manner, that "private" psychic processes include elements which

somehow to be found in all men who know the given language


and have a certain knowledge concerning the world; these eleare

ments are
etc.

Now

psychic

too,

somehow connected with

are

and psychic experiences, but cannot be explained

life

exclusively

of "notion", "meaning",

in the categories

classified

these elements,

of subjective experience.

in terms

If

represents the tendency to treat the products of

only as eigenpsychisch (and

from the

did represent that tendency), then


theoretically,

and

is

the

standpoint

latter

historical point of
is

view

it

a conception inadequate

not without reason exposed to the attacks

of anti-psychologism (by which


that

it

psychologism

human thought

do not

is

at all

automatically

want to imply
correct
and

flawless).

Both psychologism and anti-psychologism thus find a real


basis in the analysis of human spiritual processes and such of
their products as notions, meanings, etc.

two trends can

successfully expose

At the same

time, these

one another, each pointing

out the one-sidedness and limitations of the other; consequently,

both are unacceptable to a sober researcher. Psychologism usually


adopted the standpoint of subjective idealism and was then
helpless in face of the

of psychic acts.
is

criticized

On

problem of regularity and communicability

the other hand, traditional anti-psychologism

by those who,

in

the

name of

scientific

sobriety.

The Meanings of "Meaning"

come out

against the fideistic conceptions of the ghosts of "no-

or meanings as "ideal entities".

tions in themselves"

The

of the thought process reveals the apparent

analysis

The thought process

character of the conflict.


process, that
it

it

is

a cognitive

always someone's thought about something,

is, it is

always transforms the sense data and

that

287

is

objective in the sense

provides information about reality which exists objectively

and independently of human mind. Of course, an epistemological

discussion only begins here

a discussion which draws a clear

of division between the adherent of such or other theory of

line

and the opponents of

reflection

are very

that theory (whose opinions too

much diff'erentiated). But one fact remains beyond diswe insist on some metaphysical conception and

pute (unless

deliberately reject the testimonies of science, above all psychology

and physiology); that

fact

tive process takes place

that the thought process as a cogni-

is

not only with the help of linguistic means

(verbal signs), but also in an organic unity with linguistic proces-

One might

ses.

quite well use interchangeably the expressions

and "to experience linguistic processes", since in both


cases we refer to the same process of thinking, the only difference being that of a stress on one of its aspects. For there is no
"to think"

separate thought process and separate process of linguistic experience, but there

and

ing

is

always the homogenous process of think-

linguistic experience.

From our

point of view, this

cess of thinking

and

is

the focal issue here.

linguistic experience will

The pro-

be neither analysed

nor explained as yet; we shall revert later on to that problem

which
also

is

to

of immense importance not only to epistemology, but


linguistic

and

to

semantics in

its

broader theoretical

aspect. But certain conclusions must be drawn here and now.

We

always think in terms of integrated thoughts which take


on the form of this or that proposition consequently, of

That sentence does not express the proposition


usually formulated but is inseparable from it: a prop-

a sentence.
as

it

is

osition

without

its

given verbal

form

cannot

be

expressed.

Selected Problems of Semantics

288

and cannot even be formulated

and sentence are not

in one's mind. Thus, proposition

different categories, or

even

distinct

and

separate wholes, in spite of the fact that propositions (and judge-

by logic and psychology, and sentences,


and logic. The thought process alv^ays appears
a verbal form more strictly, it is always linguistic. It is not

ments)

by
in

studied

are

linguistics

true that

we

experience independently a proposition as

our

thought about something, and some language processes in the


sense of understanding the appropriate words and their combinations. Neither
linguistic

it is

true that the thought

expression

sentence

is

(sentences

with

is later

on joined by

appropriate

its

meaning).

not any expression of any thought or proposition

because no such situation

is

possible in

which a thought, or

a proposition, without words (even in one's mind) develops


first

and

is

by the words which express

later joined

Now, what has been


cesses,

and about such

indivisible units of those processes as

proposition-sentences, refers

verbal expression (that


signs).

is,

it.

said about the integral thought pro-

in

toto to

such elements as notion-

a verbal sign or a combination of such

disregard here the intricate problem of the relations

between proposition and notion and between sentence and word,


since that transcends the limits of our present analysis. I shall

only focus attention on that aspect of the problem which was


the starting point of this discussion, namely the relation between

notion and verbal expression (a verbal sign or a combination

of such

signs), the relation

the meaning of a

between the content of a notion and

word (which by way of abbreviation

is

called

the issue of the relation between notion and meaning).


It is

necessary to begin with the statement that there are no

on the one hand and experiencing a corresponding meaning on the other. When we
pronounce with comprehension the word "horse", this is accompanied by a thought process which is the process of underseparate acts of experiencing a notion

standing that word, or the processes of experiencing

Nothing

else

its

meaning.

which would be a separate experience of the notion

*i;|

The Meanings of "Meaning"

appears in

'"horse"

real

cognitive

289

processes.

then

Is

notion

a fiction, as was claimed by, for instance, Schlick? By no


I

mean by notion that same product of


when examined from the

of reahty which,

communication,

is

of

number of

meaning

and

human

significant consequences,

for the interpretation of the interrelations

all

notion

a generahzcd reflection
point of view of

called "meaning".

This statement implies a


first

means

former

(the

between

understood

being

here

not as an act of experiencing a notion, but as the content of


a notion). This reveals their identity, which

of the same cognitive process

And what

is

usually veiled by

approached from various

are the difficulties which

that a certain mental process

The

is

terminology and obscured by the fact that the analysis

different

is

may

arise

analysed as to

issue of differences in content

its

aspects.

from the

fact

various aspects?

between notion and meaning

As has been pointed out above, these differences


appear only when different cognitive processes come in question
is

rather simple.

(for instance,

a scientific notion differs from the ordinary meaning

of a word, but does not differ from the meaning of that word

when used

as a scientific term). This point

may now

be consid-

ered elucidated.

There

is,

how

however, a serious problem

still

requiring explana-

from an individual act of experiencing a notion or a meaning to the intersubjective communicability of that act and to its repetition within a language
community? That problem rests at the root of the speculations
on propositions, notions, and meanings, as ideal or intentional
tion

is it

possible to pass

entities, etc.

Let us again begin with the statement that every spiritual


process,

and consequently every thought

eigenpsychisch. In that sense

time

it is

an objective process

the relation "the subject


tion,

which

is

it

is

process,

subjective.

in the sense that

is

But
it

the material object".

individual,

at the

same

always involves

That conten-

a very essential point in the Marxist theory of

reflection, will help us solve the

problems which

interest us.

Selected Problems of Semantics

2(K)

First of

the problem of the repetition of psychic proces-

all,

connected with definite linguistic expressions (the mechanism

ses

of that relation

is

disregarded here). If

we emphasize only
namely the

subjective aspect of the cognitive process,


it

eigenpsychisch (and this

is

terference of

some

usually

is

problem becomes

trends), the

done by psychologistic

in fact insoluble

without the

in-

which are the ideal pattern of

ideal entities

of judgement, the intentional object,

all acts

the

fact that

etc.

But

it

suffices

to take into account the fact that cognitive acts of subjects which

have the same perceptive apparatus, acts relating to one and

same

the

object, are the

One has

reasons.

same

(or analogous) for

most natural

here to agree fully with the argumentation

of the naturalists that to understand that repetition of psychic


acts

as

T"

no "transcendental

ideal entities construed


real

ideal

When

objects,

we may add

necessary;

is

that no

by conceptual realism, no propositions


etc.,

are

necessary

either.

say "There are horses over there", and those words

are heard by

someone who understands

English, then both of

us understand those words as the information that there are

horses

in

the

which

place

these words in the

indicate.

same way,

since

Both of us understand

we know

the verbal signs

concerned, and these signs are connected in our minds with

corresponding perceptive

reproductive

or

representations

de-

veloped in the process of a cognitive reflection on reality by our


minds. In each individual case, these cognitive acts include lots

of other experiences connected with our knowledge of the object


involved, and with personal emotions which develop as a reac-

tion to the given object (reminiscences, artistic tastes, different

systems of values,

etc.).

But in each of these acts there

is

always,

apart from the elements which account for the differences between

them

in view of their specifically individual nature,

an element

which makes them similar owing to which the understanding


of the given words
This
listeners

applies

is

similar,

not only to

of a statement

made

though not
a

"common"

identical.

understanding

by

hie et nunc, but also to those cases

The Meanings oh "Meaning"

which the statement

iia

as a "sign-event". Let us suppose that

is

above

experience

entity

which

is

periences
state of

are

will experience

He

of

as has been stated

will

appropriate

the

because there

understand the

proposition

and

any proposition as an ideal

is

actualized in individual experiences, but

men who understand

in

(knowledge

and meanings of the individual words, and

psychically

Why? Not

notions.

comprehendingly

a psychic act similar to ours.

sense of the sentence


will

nol

have to do with

still

assumed) or whoever in the future pronounces them

comprehendingly, that person

we

"There are horses over there". Whoever reads

the inscription

that inscription in the future

English

"sign-design" and

treated as a

is

!iJM

because

a given language, similar cognitive ex-

statement (the normal

connected with a given

minds being assumed), which

is

in turn

due to the

reflec-

tion of the same reahty by perception apparatuses which are built

and function in the same way. The thing as such


although the mechanism of the process
It

does not

suffice to refer to

is

quite simple,

extremely intricate.

is

language rules the observance of

which conditions a similar understanding of statements. That the


observance of language does that,
true, since a further

explanation

of those language rules? If

explanation then

we again

we

is

is

true;

required.

rest

but only partially

What

is

the origin

with that partial

satisfied

mystify the problem, this time from

and open the door wide for conventionalism. Such was the case of the numerous positivist interpretations of the issue with which we are concerned.
the position of nominalism,

Those who are


chic

life

in favour of transforming products of psy-

into ideal entities, refer in their arguments to the dif-

ferences between,

on the one hand, a psychic

act treated as a

whole and connected with experiencing a certain statement


(or with the perception of definite objects, also connected with
definite

the

language-and-thought

processes),

understanding of certain meanings

on the

notions

other,

connected

we ask ourselves or other people such


"What does the word ... mean?" or "What is ...?"

with that statement when


a question as

and
or

292

Selected Problems of Semantics

must be

It

realized

clearly

that

the defenders of the

al-

and notion, who treat either


eigenpsychisch and correspondingly

leged difference between meaning

the former or the latter as

transform the second element of the pair "notion


into

something which

meaning"

given intersubjectively (as a logical

is

or other entity), have in fact another relation in mind: namely

on the one hand, a concrete psychic act (in the sense


on the other, such products of conscious
reflection, which abstract from many factors, as no-

that between,

of a

full

experience) and,

cognitive

tion or meaning.

When

utter the sentence

statement

is

heard by people

but

"Pigs are mammals" and that


who understand the language I use

differ as to their origin, religion, education, etc.,

then while

they understand that statement in the same

way they may

considerably in the ways they experience

it.

differ

For the psychic

act of understanding that statement will include not only their

knowledge of pigs and mammals

(differing widely

in

e.g.,

case of a village veterinary surgeon and a

London

but also their emotional reactions connected

e.g.,

the

sales girl),

with rehgious

prejudicies (a Christian vs. a Muslim), aesthetic evaluations and/or

associations (one person

may

think about a dirty hog wal-

lowing in mud, and another about an amusing, pink, rolhcking


piglet),

certain

personal reminiscences,

etc.

There

is

no doubt

that every psychic act, including those which are par excellence
cognitive acts, has not only a cognitive content (intellectual description),

(moral

but also a content consisting of personal emotions

valuation,

aesthetic

assessment,

If that

etc.).

psychic act connected with a given statement

called

is

integral

meaning

or notion (both cases can be found in the literature of the subject),

then of course meaning or notion so understood

is

transformed

into something "private" not only in the sense of the subjective

nature of the experience involved, but also in the sense of incommunicability.

It

is

not to be wondered

at,

then,

a case to a "private" meaning or notion one

"pubhc" partner

in the

that

in

such

may oppose

its

form of the other element of the pair

The Meanings of "Meaning"

"meaning-notion".

But

this

is

293

only a terminological arbitra-

which merely complicates the

issue, for in the proper and


words neither "notion" nor "meaning"
are names of the relation with which the complete psychic act
is connected. "Notion" and "meaning" are names of the relation

riness

original sense of these

with which are connected certain particular acts, based on the specific

process of abstraction which

is

indispensable

answer the question "What does the word

...

if

one

is

to

mean?" or "What

is ...?"

All this should not be understood to imply that a conscious

process of abstraction takes place every time, and that only

we have

do with meaning or notion. Such conscious


processes are very rare, and in principle are elements of research
then

work. Usually,

to

takes place through spontaneous learning

all this

of a given language, in
a given society
that society.

is

wliicli

process the social experience of

transmitted tlirough language to a

Hence we

remain unconscious of

member of

usually do not notice that process and

it.

On

the other hand, scientific (psycholo-

gical, philosophical, linguistic, etc.)

research reveals the selective

and abstract character of such categories of the cognitive process


as notion

and meaning. Of course, the case of

and meanings of

scientific

scientific

notions

terms seems different from that of

ordinary notions and meanings in ordinary language. Although


that difference

when

may

be of great cognitive significance (especially

comes to understanding the regularities governing a given


phenomenon), nevertheless we have to do only with a difference
in the degree of abstraction. Notion and meaning are always
a result of the work of the mind on some sense data, they are
always products of abstraction and selection, and hence they
it

always
its

differ

from the cognitive process taken as a whole, with

emotional elements,

etc.

Thus notion and meaning are products of abstraction performed on the cognitive process by means of verbal signs. The
character of notions and meanings ordinary in one case, and
scientific in the other depends on the character of that abstrac.
20

294

Selected Problems of Semantics

and consequently

notions and mean-

lion.

The

ings,

always originates in the intricate system of social relations,

said abstraction,

perceived

as

by human consciousness.

a given thought-and-language product

all

According to whether
is

interpreted

from the

point of view of the thought process or the language process


(that

is,

according

or the other),

it

to

whether we emphasize the one aspect

appears either as a notion (the content of a notion)

or as a meaning of the word. There

is

no other

difference between

notion and meaning (of the same type).

Such an interpretation does not lose anything dear to the


heart of the gnoseologist, the linguist or the logician.

The gnoseolform

finds here reference of the cognitive process (in the

ogist

of notion or meaning) to objective reality; the linguist, the

and

living content of the cognitive process

in the

meaning of words; the

which

is

logician, the exact

full

manifested

and

precise

content of a scientific notion or a scientific term, which he develops

form of an adequate verbal or real definition. And all


does without ideal entities and the nominalist interpretation

in the
this

of words and notions as the flatus

vocis.

So to the conclusions:
(1) The thesis that notion and meaning are
gories

with

different

contents

is

not

arguments which are supposed to support that


to

different cate-

acceptable,

because the
thesis

prove

be in error, and the adoption of that thesis would imply adop-

tion of objective idealism.


(2)

On

the contrary,

it

is

to be concluded that notion

and

the difference between

meaning are identical as to content, and


them consists only in the fact that the same cognitive process
is interpreted from two points of view, different but inseparably
linked with one another
case,

and

that

that of the thought process in one

of the language process in the other.

(3) Differences as to content, said to

appear as between mean-

and notion, are in fact differences between a scientific and


an ordinary notion, between the meaning of a scientific term
ing

and the meaning of words as used

in

ordinary language.

The Meanings of "Meaning"

In order to avoid misunderstandings and logical

(4)
it

295

necessary to distinguish as to content between:

is

slips,

integral

psychic acts which along with cognitive elements include emotional

and other elements; ordinary notions and meanings of words;


scientific notions and meanings of scientific terms, which like
the ordinary ones, are results of a specific abstraction process

applied to an integral psychic process and are therefore

com-

municable.

D. The mechanism of the links between sign and meaning

The

and meaning can be

theories of the sign

classified

by

taking as the criterion of division the views on the character

of the finks between sign and meaning (this

used by
If

J.

we do

the procedure

is, e.g.,

Kotarbinska in her paper on the theory of the

so,

we obtain two

which see the

links

large groups of such theories

between sign and meaning

and those which see

it

in

some

sign).

those

in association,

specific intentional

act.

This

obviously refers to the psychic aspect of the problem, the psychological

ly

mechanism of the

do not think that

of the sign, but

links

between sign and meaning.

this is the

it is

main or

certainly one of those issues

to gain a better insight into the problems of sign

That

is

why

it

is

worthy of

personal-

decisive issue in the theory

which help us

and meaning.

analysis.

reference to the links between sign

and meaning should

As

not restrict the problem to the meanings of verbal signs.

and hence

we already know, verbal

signs are of a special nature

in their case the links

between sign and meaning are

and

different

This

is

from

specific

the case of the remaining categories of signs.

so because, not to look for other reasons,

all

other signs

are in a sense derivative with respect to verbal signs,


in the last analysis, translated into verbal signs. All

and

are,

proper signs

except the verbal signs resort to an actual, ready meaning which

is,

them" and is connected with a sign understood


autonomously, that is with a definite material object or maso to speak, "behind

296

Selected Problems of Semantics

The situation is quite different with the verbal signs


which have no meanings "behind them", are organically linked
with their respective meanings and in this sense are, as it is often
terial event.

expressed, "transparent to meaning".

All the proper signs (again except for the verbal signs-analysis

of that point will be

made

meanings by associative
This

with their respective

later) are linked

links.

As we know, proper signs


and are in a way conventional. Definite meannotions, if you prefer it that way) are combined in a puris

explained quite simply.

are artificial signs


ings (or

manner with material sign-vehicles, that


The sign (in that narrower sense) and meaning are here autonomous, as has been
rightly pointed out by Zvegintsev in his cited work on the signnature of language 74; moreover, they are not bound by the rules
posive and deliberate
is

signs in the narrower sense of the term.

of the system concerned

(e.g.,

syntax,

That

etc.).

form of the sign-vehicle (i.e., sign in the narrower


be changed in those cases without involving a change
For

instance, there

is

why

is

sense)

the

may

in meaning.

a road sign in the form of a triangular

board, yellow with red edges, with a black curve in the middle.

Whoever knows the road

signs

also

knows

"Beware! Sharp bend in the road!" There

is

that this means:

nothing to prevent

from changing the shape and the colour of the board and the
the meaning of the
road sign "Beware! Sharp bend in the road!" will not be changed,
us

shape and the colour of the central drawing

provided that the convention remains unchanged. The meaning


of the expression "Beware! Sharp bend in the road!"

is

con-

veyed to us by the sign adopted by a convention. But that meaning


exists regardless

of the given form of the sign, as the meaning

of a definite verbal expression which

we combine with

that material sign-vehicle that exists independently,

and

has, besides

74 B.

its

meaning, other values as well

A. SBennmcB, ITpodjieMa SHaKosocinu sihiKa

the Sign-nature of Language],

MocKBa

1956.

this

or

autonomously
(e.g.,

[The

aesthetic

Problem

of

The Meanings of "Meaning"

297

etc.). On what basis, then, do we


combine the image of the sign with the thought "Beware Sharp
bend of the road!" On the basis of association which we have

values such as shape, colour,

developed in the practice of using the given language, in a close


connection with the knowledge of convention (from which
does not follow that everything what
sign belongs to

its

it

associate with a given

meaning).

similar analysis

can be performed in every case of a

a symbol, or a substitutive sign sensu


case

we

stricto

that

is,

signal,

in every

when the material form of the sign is joined to a ready meaning


when the meaning of a sign

of an expression, in other words,


is

given not directly but through the intermediary of a verbal

sign.

The situation changes when we proceed to analyse the verbal


Then the associationist conception can hold its ground

signs.

in special cases only.


First of all, in learning a language.

The
a

trifle

what

child learns a language

by association. Let me

St.

Augustine

says

in

refer,

testimony of a saint. This

spitefully, perhaps, to the

is

Confessions:

his

my

would
would
point
make some particular sound, and
at or move towards some particular thing: and from this I came
to realize that the thing was called by the sound they made when
they wished to draw my attention to it. That they intended
this was clear from the motions of their body, by a kind of natural language common to all races which consists in facial expressions, glances of the eye, gestures, and the tones by which
the voice expresses the mind's state for example whether
"So

began to

reflect.

[I

observed that]
as they

things are to be sought, kept,


I

made

elders

it,

thrown away, or avoided. So, as

heard the same words again and again properly used in

ferent phrases, I
fied;

and forcing

them

to express

75 St.

dif-

gradually to grasp what things they signi-

came
my mouth

my own

to the

same sounds,

began to use

wishes" ^s.

Augustin, Confessions, trans, by F.

J.

Sheed,

London

1943, p. 11.

298

Selected Problems of Semantics

Psychologically, there

a language than by
especially

when

child in action

it

is

way of
comes

no other way for an infant to learn


The meanings of words,

associations.

to abstract terms, are learned

combined with speaking, by numerous

by the

repetitions

of specific definitions in use.

When an

adult

who

already

knows some language

learning

is

a foreign language he also has to do with associations, but of


a quite different type. In his case, there

is

an association of sounds

of foreign language words with the words of his mother tongue.

That

is

why

in the initial stage of learning the foreign language

he thinks in his native language and translates into the foreign

language he

is

learning.

(foreign) language only

He

will,

when he

of course, really

ceases to translate

know

that

and begins

to think in that newly-learned language.

But here ends the application of the associationist theory


to the verbal signs. All attempts to interpret the meanings of

words
is

in

terms of that theory are complete

Meaning

failures.

then treated as the images, representations, thoughts associated

with the sounds of words. Such was the standpoint of Russell


at the time of the

well-known discussion

in

Mind

(he then

com-

pared language processes to the dancing of the bears provoked

by the tune which formerly used to be played when the bears


were placed on a hot floor); a similar position was at one time
held by Wittgenstein, and

among

In the criticism of that opinion,


forth by

Polish hnguists too,


I

e.g.

Szober.

agree with the arguments set

Ajdukiewicz. The interpretation of the verbal signs

as sounds with

which certain independently existing thoughts

are associated

due to a complete disregard of the nature of lan-

is

guage processes and thought processes. Not only are there no


thoughts existing independently of speech sounds (which

is

closely

connected with the role of verbal signs in the abstraction process

on the level of thinking in terms of notions), but also there is


no thought independent of a system of such signs, that is the
syntax of the language concerned,

etc.

All attempts to reduce

meanings of words to associations with images of objects or

The Meanings of "Meaning"

with

extremely

representations,

become

to names,

primitive

absolutely unacceptable

290
even

when

with
it

lespecl

comes

to

words that are not names and to more complicated combinations


of signs in sentences.
Sohdarity with the criticism of associationism by no means

amounts to

phers as Ajdukiewicz

the form of the intentional con-

critics in

As

ception of meaning.

to

with the alternative proposal advanced

solidarity

by the majority of the

have

said, in the case

of such philoso-

consider his acceptance of intentionalism

be a misunderstanding due to a wrong interpretation of Hus-

serl's

standpoint. I have previously striven to present that stand-

point in the most faithful

and unbiased manner, namely by


now see what Ajdukiewicz said

quoting Husserl himself. Let us

on that subject

in his

"On

paper

the

"According to Husserl, that

Meaning of Expressions"

'act

of meaning', or the use

of a given phrase as an expression of a certain language, consists


in the fact that a sensory content

appears in consciousness, by

means of which one might think

visually

about that phrase,

should that content be joined by an appropriate intention directed


to that phrase.

But when a given phrase

is

used as an expression

belonging to a certain language, then that sensory content


joined by
one, which

is

another intention, not necessarily a representative


is

however

than that phrase

in principle directed to

itself.

something other

Together with the sensory content in

makes up a uniform

question, that intention

experience,

but

neither the experiencing of that sensory content, nor that intention


is

a complete, independent experience. Both the one and the other

are non-independent parts of the experience as a whole.

ing of a given expression (as a type)


serl,

would

The mean-

be, according to

Hus-

the type under which that intention joined to the sensory

content must

fall if

the given phrase

is

to

be used as an expres-

sion belonging precisely to that language"76.

The misunderstanding
sible,
"76

clears.

that whole passage has


Ajdukiewicz, op.

cit.,

To put

it

except for the

pp. 19-20.

as mildly as pos-

word "intention"

Selected Problems of Semantics

300
little

common

in

with Husserl's intentionalist conception.

not permissible to lose on the

way

all

It is

the luggage of Husserl's

metaphysics, without which the interpretation of that author's

opinions becomes quite arbitrary and alien to phenomenology.

That

why

word "intention" has with Ajdukiewicz a difmeaning from that it has with Husserl. It seems that such

is

ferent

Polish

the

authors

who

Ajdukiewicz and those

as

follow

him

(Czezowski, Kotarbinska, and others) are in error in admitting


that their views are related to Husserl's.

Ajdukiewicz really means by "intention"

When we
we

analyse what

find rather certain

connections with behaviourist semiotic. Ajdukiewicz has in mind


rather a certain readiness to use language in such

and such way

and not otherwise, and consequently he means certain psychic


dispositions. This is not a phenomenological conception; the
idea has nothing in

common

with phenomenology, but plays an

important role with Morris and others. But that


point:

my

intention

is

is

a secondary

not to discuss that issue or to

criticize

the relative views of Ajdukiewicz, since the problem under dis-

cussion

is

par excellence a psychological one, even in the physio-

and such issues are not decided


by philosophizing. There also comes a moment in which the
term

logical aspect of that

method of imparting
its

precision to ideas or notions encounters

natural limit. Such

performed

all

rejected all the

words,
links

is,

at the

moment when we have already


is when we have

the preliminary operations, that

wrong

etc., precisely

theories, given precision to the sense of

the case of the issue of the nature of the

between speech sounds and

their meanings.

can provide an answer to the question


between sounds and meanings,
ly consist? It is

in

now under

No

deduction

what does the hnk

investigation, actual-

a matter which requires experimental studies, and

as such belongs above all to experimental psychology

physiology of the brain.

and the

philosopher can at the best adopt

a certain attitude towards such or other proposals as to

how

to solve the problem, proposals submitted by such experimental


disciplines,

and that

is all. I

think

it

reasonable to confine myself

The Meanings of "Meaning"

to the statement that the link between

verbal signs

301

sound and meaning

in the

a connection sui generis and does not consist

is

For the

in association.

rest,

as has been said

it

belongs to

psychology and physiology. Certainly, the results of research


in those disciplines are
ular,

it

not entirely satisfactory so far in partic;

does not seems possible to accept the Pavlovian hy-

pothesis concerning the second system of signals as definitively

formulated and proved, and consequently as one which fully ex-

and the mechanism of the


functioning of thought-language. All this is true. But philosplains the nature of the verbal signs

made by

ophers can only wait for further progress to be


perimental sciences in the

5.

field

they are interested

ex-

in.

THE LINGUISTIC APPROACH TO THE ISSUE OF MEANING


References to

linguistic

meaning are frequently found

in

the literature of the subject. That might suggest that linguistics

has

its

own,

distinct,

theory of meaning. This

misunderstanding, to be refuted at the outset.

which

ing

ing?

How

above

is it

all

answers such questions as

connected with sign?,

etc.

an obvious

is

theory of mean:

What

is

mean-

does not and cannot

vary according to the discipline concerned.

On

the contrary,

every author, regardless of his speciality and consequent approach


to the problem, if

formulates

What,

it

as a general theory with universal validity.

then,

Simply what

he formulates a theory of meaning, always

is

lies

behind

the

indicated in the

term

title

"hnguistic

meaning"?

of this section, namely an

approach to the issue of meaning from the point of view of linguistic interest, that is of those questions which are theoretically
significant in specifically linguistic research.

This

is

as natural

and comprehensible as the

fact that the logician, the sociologist,

investigates linguistic

problems from the point of view of

etc.,

his specific interest.

in

What, then, are the theoretical aspects of linguistic interest


the issue of meaning? They have been outhned in a most gen-

302

Selected Problems of Semantics

manner

eral

in the

first

when

part of this book,

the nature and

scope of semantic interests of hnguistics was tentatively ex-

But

plained.

meaning was one of the many


comes back as the issue, which may

there, the issue of

issues involved

at present

it

shed additional light on the questions we are discussing.

As opposoed
guist

is

and the

to the philosopher

logician, the lin-

what meaning

in principle not interested in

wants to know what happens to meaning, that

what language

units)

it is

an

attribute, in

mean something, how meaning


that

it

is

changes,

but he

what way verbal

etc. I

agree with

signs

Quine'^'''

possible to study the regularities of something even

without knowing what that something


ers

is,

of what (of

is,

knew very

what planets

well the

is

movements of

(e.g.,

ancient astronom-

planets without

knowing

are).

According to Quine, there are three principal spheres of the


linguistic interest in the issue
(1)

(2)

of meaning:

the study of those forms which have


lexicography the study of synonyms,

grammar

i.e.,

meaning;
expres-

sions which have similar meanings; thus the subject matter of

lexicography

is

to

identify

synonymous expressions
(3)
ysis)

in a

meanings, that

is

to

list

pairs of

language or in a pair of languages</