Empiricism, Semantics, and Ontology*

Rudolf Carnap
Revue Internationale de Philosophie 4 (1950): 20-40. Reprinted in the Supplement to Meaning and
Necessitv. A Studv in Semantics and Modal Logic, enlarged edition (University oI Chicago Press,
1956).
1. The problem of abstract entities
Empiricists are in general rather suspicious with respect to any kind oI abstract entities like
properties, classes, relations, numbers, propositions, etc. They usually Ieel much more in sympathy
with nominalists than with realists (in the medieval sense). As Iar as possible they try to avoid any
reIerence to abstract entities and to restrict themselves to what is sometimes called a nominalistic
language, i.e., one not containing such reIerences. However, within certain scientiIic contexts it
seems hardly possible to avoid them. In the case oI mathematics some empiricists try to Iind a way
out by treating the whole oI mathematics as a mere calculus, a Iormal system Ior which no
interpretation is given, or can be given. Accordingly, the mathematician is said to speak not about
numbers, Iunctions and inIinite classes but merely about meaningless symbols and Iormulas
manipulated according to given Iormal rules. In physics it is more diIIicult to shun the suspected
entities because the language oI physics serves Ior the communication oI reports and predictions
and hence cannot be taken as a mere calculus. A physicist who is suspicious oI abstract entities may
perhaps try to declare a certain part oI the language oI physics as uninterpreted and uninterpretable,
that part which reIers to real numbers as space-time coordinates or as values oI physical
magnitudes, to Iunctions, limits, etc. More probably he will just speak about all these things like
anybody else but with an uneasy conscience, like a man who in his everyday liIe does with qualms
many things which are not in accord with the high moral principles he proIesses on Sundays.
Recently the problem oI abstract entities has arisen again in connection with semantics, the theory
oI meaning and truth. Some semanticists say that certain expressions designate certain entities, and
among these designated entities they include not only concrete material things but also abstract
entities e.g., properties as designated by predicates and propositions as designated by sentences.1
Others object strongly to this procedure as violating the basic principles oI empiricism and leading
back to a metaphysical ontology oI the Platonic kind.
It is the purpose oI this article to clariIy this controversial issue. The nature and implications oI the
acceptance oI a language reIerring to abstract entities will Iirst be discussed in general; it will be
shown that using such a language does not imply embracing a Platonic ontology but is perIectly
compatible with empiricism and strictly scientiIic thinking. Then the special question oI the role oI
abstract entities in semantics will be discussed. It is hoped that the clariIication oI the issue will be
useIul to those who would like to accept abstract entities in their work in mathematics, physics,
semantics, or any other Iield; it may help them to overcome nominalistic scruples.
2. Linguistic frameworks
Are there properties classes, numbers, propositions? In order to understand more clearly the nature
oI these and related problems, it is above all necessary to recognize a Iundamental distinction
between two kinds oI questions concerning the existence or reality oI entities. II someone wishes to
speak in his language about a new kind oI entities, he has to introduce a system oI new ways oI
speaking, subject to new rules; we shall call this procedure the construction oI a linguistic
framework Ior the new entities in question. And now we must distinguish two kinds oI questions oI
existence: Iirst, questions oI the existence oI certain entities oI the new kind within the framework;
we call them internal questions; and second, questions concerning the existence or reality of the
svstem of entities as a whole, called external questions. Internal questions and possible answers to
them are Iormulated with the help oI the new Iorms oI expressions. The answers may be Iound
either by purely logical methods or by empirical methods, depending upon whether the Iramework
is a logical or a Iactual one. An external question is oI a problematic character which is in need oI
closer examination.
The world of things. Let us consider as an example the simplest kind oI entities dealt with in the
everyday language: the spatio-temporally ordered system oI observable things and events. Once we
have accepted the thing language with its Iramework Ior things, we can raise and answer internal
questions, e.g., "Is there a white piece oI paper on my desk?" "Did King Arthur actually live?",
"Are unicorns and centaurs real or merely imaginary?" and the like. These questions are to be
answered by empirical investigations. Results oI observations are evaluated according to certain
rules as conIirming or disconIirming evidence Ior possible answers. (This evaluation is usually
carried out, oI course, as a matter oI habit rather than a deliberate, rational procedure. But it is
possible, in a rational reconstruction, to lay down explicit rules Ior the evaluation. This is one oI the
main tasks oI a pure, as distinguished Irom a psychological, epistemology.) The concept oI reality
occurring in these internal questions is an empirical scientiIic non-metaphysical concept. To
recognize something as a real thing or event means to succeed in incorporating it into the system oI
things at a particular space-time position so that it Iits together with the other things as real,
according to the rules oI the Iramework.
From these questions we must distinguish the external question oI the reality oI the thing world
itselI. In contrast to the Iormer questions, this question is raised neither by the man in the street nor
by scientists, but only by philosophers. Realists give an aIIirmative answer, subjective idealists a
negative one, and the controversy goes on Ior centuries without ever being solved. And it cannot be
solved because it is Iramed in a wrong way. To be real in the scientiIic sense means to be an
element oI the system; hence this concept cannot be meaningIully applied to the system itselI.
Those who raise the question oI the reality oI the thing world itselI have perhaps in mind not a
theoretical question as their Iormulation seems to suggest, but rather a practical question, a matter
oI a practical decision concerning the structure oI our language. We have to make the choice
whether or not to accept and use the Iorms oI expression in the Iramework in question.
In the case oI this particular example, there is usually no deliberate choice because we all have
accepted the thing language early in our lives as a matter oI course. Nevertheless, we may regard it
as a matter oI decision in this sense: we are Iree to choose to continue using the thing language or
not; in the latter case we could restrict ourselves to a language oI sense data and other
"phenomenal" entities, or construct an alternative to the customary thing language with another
structure, or, Iinally, we could reIrain Irom speaking. II someone decides to accept the thing
language, there is no objection against saying that he has accepted the world oI things. But this
must not be interpreted as iI it meant his acceptance oI a belief in the reality oI the thing world;
there is no such belieI or assertion or assumption, because it is not a theoretical question. To accept
the thing world means nothing more than to accept a certain Iorm oI language, in other words, to
accept rules Ior Iorming statements and Ior testing accepting or rejecting them. The acceptance oI
the thing language leads on the basis oI observations made, also to the acceptance, belieI, and
assertion oI certain statements. But the thesis oI the reality oI the thing world cannot be among
these statements, because it cannot be Iormulated in the thing language or, it seems, in any other
theoretical language.
The decision oI accepting the thing language, although itselI not oI a cognitive nature, will
nevertheless usually be inIluenced by theoretical knowledge, just like any other deliberate decision
concerning the acceptance oI linguistic or other rules. The purposes Ior which the language is
intended to be used, Ior instance, the purpose oI communicating Iactual knowledge, will determine
which Iactors are relevant Ior the decision. The eIIiciency, IruitIulness, and simplicity oI the use oI
the thing language may be among the decisive Iactors. And the questions concerning these qualities
are indeed oI a theoretical nature. But these questions cannot be identiIied with the question oI
realism. They are not yes-no questions but questions oI degree. The thing language in the
customary Iorm works indeed with a high degree oI eIIiciency Ior most purposes oI everyday liIe.
This is a matter oI Iact, based upon the content oI our experiences. However, it would be wrong to
describe this situation by saying: "The Iact oI the eIIiciency oI the thing language is conIirming
evidence Ior the reality oI the thing world; we should rather say instead: "This Iact makes it
advisable to accept the thing language."
The svstem of members. As an example oI a system which is oI a logical rather than a Iactual nature
let us take the system oI natural numbers. The Iramework Ior this system is constructed by
introducing into the language new expressions with suitable rules: (1) numerals like "Iive" and
sentence Iorms like "there are Iive books on the table"; (2) the general term "number" Ior the new
entities, and sentence Iorms like "Iive is a number"; (3) expressions Ior properties oI numbers (e.g.
"odd," "prime"), relations (e.g., "greater than") and Iunctions (e.g. "plus"), and sentence Iorms like
"two plus three is Iive"; (4) numerical variables ("m," "n," etc.) and quantiIiers Ior universal
sentences ("Ior every n . . . ) and existential sentences ("there is an n such that . . .") with the
customary deductive rules.
Here again there are internal questions, e.g., "Is there a prime number greater than a hundred?" Here
however the answers are Iound not by empirical investigation based on observations but by logical
analysis based on the rules Ior the new expressions. ThereIore the answers are here analytic, i.e.,
logically true.
What is now the nature oI the philosophical question concerning the existence or reality oI
numbers? To begin with, there is the internal question which together with the aIIirmative answer,
can be Iormulated in the new terms, say by "There are numbers" or, more explicitly, "There is an n
such that n is a number." This statement Iollows Irom the analytic statement "Iive is a number" and
is thereIore itselI analytic. Moreover, it is rather trivial (in contradistinction to a statement like
"There is a prime number greater than a million which is likewise analytic but Iar Irom trivial),
because it does not say more than that the new system is not empty; but this is immediately seen
Irom the rule which states that words like "Iive" are substitutable Ior the new variables. ThereIore
nobody who meant the question "Are there numbers?" in the internal sense would either assert or
even seriously consider a negative answer. This makes it plausible to assume that those
philosophers who treat the question oI the existence oI numbers as a serious philosophical problem
and oIIer lengthy arguments on either side, do not have in mind the internal question. And indeed, iI
we were to ask them: "Do you mean the question as to whether the Iramework oI numbers, if we
were to accept it, would be Iound to be empty or not?" they would probably reply: "Not at all; we
mean a question prior to the acceptance oI the new Iramework." They might try to explain what
they mean by saying that it is a question oI the ontological status oI numbers; the question whether
or not numbers have a certain metaphysical characteristic called reality (but a kind oI ideal reality,
diIIerent Irom the material reality oI the thing world) or subsistence or status oI "independent
entities." UnIortunately, these philosophers have so Iar not given a Iormulation oI their question in
terms oI the common scientiIic language. ThereIore our judgment must be that they have not
succeeded in giving to the external question and to the possible answers any cognitive content.
Unless and until they supply a clear cognitive interpretation, we are justiIied in our suspicion that
their question is a pseudo-question, that is, one disguised in the Iorm oI a theoretical question while
in Iact it is a non-theoretical; in the present case it is the practical problem whether or not to
incorporate into the language the new linguistic Iorms which constitute the Iramework oI numbers.
The svstem of propositions. New variables, "p," "q," etc., are introduced with a role to the eIIect that
any (declarative) sentence may be substituted Ior a variable oI this kind; this includes, in addition to
the sentences oI the original thing language, also all general sentences with variables oI any kind
which may have been introduced into the language. Further, the general term "proposition" is
introduced. "p is a proposition" may be deIined by "p or not p" (or by any other sentence Iorm
yielding only analytic sentences) . ThereIore every sentence oI the Iorm ". . . is a proposition"
(where any sentence may stand in the place oI the dots) is analytic. This holds, Ior example, Ior the
sentence:
(a) Chicago is large is a proposition.
(We disregard here the Iact that the rules oI English grammar require not a sentence but a that-
clause as the subject oI another sentence; accordingly instead oI (a) we should have to say "That
Chicago is large is a proposition.") Predicates may be admitted whose argument expressions are
sentences; these predicates may be either extensional (e.g. the customary truth-Iunctional
connectives) or not (e.g. modal predicates like "possible," "necessary," etc.). With the help oI the
new variables, general sentences may be Iormed, e.g.,
(b) "For every p, either p or not-p."
(c) "There is a p such that p is not necessary and not-p is not necessary."
(d) "There is a p such that p is a proposition."
(c) and (d) are internal assertions oI existence. The statement "There are propositions" may be
meant in the sense oI (d); in this case it is analytic (since it Iollows Irom (a)) and even trivial. II,
however, the statement is meant in an external sense, then it is non-cognitive.
It is important to notice that the system oI rules Ior the linguistic expressions oI the propositional
Iramework (oI which only a Iew rules have here been brieIly indicated) is suIIicient Ior the
introduction oI the Iramework. Any Iurther explanations as to the nature oI the propositions (i.e.,
the elements oI the system indicated, the values oI the variables "p," "q," etc.) are theoretically
unnecessary because, iI correct, they Iollow Irom the rules. For example, are propositions mental
events (as in Russell's theory)? A look at the rules shows us that they are not, because otherwise
existential statements would be oI the Iorm: "II the mental state oI the person in question IulIills
such and such conditions, then there is a p such that . . . ." The Iact that no reIerences to mental
conditions occur in existential statements (like (c), (d), etc.) shows that propositions are not mental
entities. Further, a statement oI the existence oI linguistic entities (e.g., expressions, classes oI
expressions, etc.) must contain a reIerence to a language. The Iact that no such reIerence occurs in
the existential statements here, shows that propositions are not linguistic entities. The Iact that in
these statements no reIerence to a subject (an observer or knower) occurs (nothing like: "There is a
p which is necessary Ior Mr. X."), shows that the propositions (and their properties, like necessity,
etc.) are not subjective. Although characterizations oI these or similar kinds are, strictly speaking,
unnecessary, they may nevertheless be practically useIul. II they are given, they should be
understood, not as ingredient parts oI the system, but merely as marginal notes with the purpose oI
supplying to the reader helpIul hints or convenient pictorial associations which may make his
learning oI the use oI the expressions easier than the bare system oI the rules would do. Such a
characterization is analogous to an extra-systematic explanation which a physicist sometimes gives
to the beginner. He might, Ior example, tell him to imagine the atoms oI a gas as small balls rushing
around with great speed, or the electromagnetic Iield and its oscillations as quasi-elastic tensions
and vibrations in an ether. In Iact, however, all that can accurately be said about atoms or the Iield
is implicitly contained in the physical laws oI the theories in question.2
The svstem of thing properties The thing language contains words like "red," "hard," "stone,"
"house," etc., which we used Ior describing what things are like. Now we may introduce new
variables, say "f," "g," etc., Ior which those words are substitutable and Iurthermore the general
term "property." New rules are laid down which admit sentences like "Red is a property," "Red is a
color," "These two pieces oI paper have at least one color in common" (i.e., "There is an f such that
f is a color, and . . ."). The last sentence is an internal assertion. It is an empirical, Iactual nature.
However, the external statement, the philosophical statement oI the reality oI properties -- a special
case oI the thesis oI the reality oI universals -- is devoid oI cognitive content.
The svstem of integers and rational numbers. Into a language containing the Iramework oI natural
numbers we may introduce Iirst the (positive and negative) integers as relations among natural
numbers and then the rational numbers as relations among integers. This involves introducing new
types oI variables, expressions substitutable Ior them, and the general terms "integer" and "rational
number."
The svstem of real numbers. On the basis oI the rational numbers, the real numbers may be
introduced as classes oI a special kind (segments) oI rational numbers (according to the method
developed by Dedekind and Frege). Here again a new type oI variables is introduced, expressions
substitutable Ior them (e.g., "\2" |square root oI 2|), and the general term "real number."
The spatio-temporal coordinate svstem for phvsics. The new entities are the space-time points. Each
is an ordered quadruple oI Iour real numbers, called its coordinates, consisting oI three spatial and
one temporal coordinates. The physical state oI a spatio-temporal point or region is described either
with the help oI qualitative predicates (e.g., "hot") or by ascribing numbers as values oI a physical
magnitude (e.g., mass, temperature, and the like). The step Irom the system oI things (which does
not contain space-time points but only extended objects with spatial and temporal relations between
them) to the physical coordinate system is again a matter oI decision. Our choice oI certain Ieatures,
although itselI not theoretical, is suggested by theoretical knowledge, either logical or Iactual. For
example, the choice oI real numbers rather than rational numbers or integers as coordinates is not
much inIluenced by the Iacts oI experience but mainly due to considerations oI mathematical
simplicity. The restriction to rational coordinates would not be in conIlict with any experimental
knowledge we have, because the result oI any measurement is a rational number. However, it would
prevent the use oI ordinary geometry (which says, e.g., that the diagonal oI a square with the side I
has the irrational value \2) and thus lead to great complications. On the other hand, the decision to
use three rather than two or Iour spatial coordinates is strongly suggested, but still not Iorced upon
us, by the result oI common observations. II certain events allegedly observed in spiritualistic
seances, e.g., a ball moving out oI a sealed box, were conIirmed beyond any reasonable doubt, it
might seem advisable to use Iour spatial coordinates. Internal questions are here, in general,
empirical questions to be answered by empirical investigations. On the other hand, the external
questions oI the reality oI physical space and physical time are pseudo-questions. A question like:
"Are there (really) space-time points?" is ambiguous. It may be meant as an internal question; then
the aIIirmative answer is, oI course, analytic and trivial. Or it may be meant in the external sense:
"Shall we introduce such and such Iorms into our language?"; in this case it is not a theoretical but a
practical question, a matter oI decision rather than assertion, and hence the proposed Iormulation
would be misleading. Or Iinally, it may be meant in the Iollowing sense: "Are our experiences such
that the use oI the linguistic Iorms in question will be expedient and IruitIul?" This is a theoretical
question oI a Iactual, empirical nature. But it concerns a matter oI degree; thereIore a Iormulation in
the Iorm "real or not?" would be inadequate.
3. What does acceptance of a kind of entities mean?
Let us now summarize the essential characteristics oI situations involving the introduction oI a new
kind oI entities, characteristics which are common to the various examples outlined above.
The acceptance oI a new kind oI entities is represented in the language by the introduction oI a
Iramework oI new Iorms oI expressions to be used according to a new set oI rules. There may be
new names Ior particular entities oI the kind in question; but some such names may already occur in
the language beIore the introduction oI the new Iramework. (Thus, Ior example, the thing language
contains certainly words oI the type oI "blue" and "house" beIore the Iramework oI properties is
introduced; and it may contain words like "ten" in sentences oI the Iorm "I have ten Iingers" beIore
the Iramework oI numbers is introduced.) The latter Iact shows that the occurrence oI constants oI
the type in question -- regarded as names oI entities oI the new kind aIter the new Iramework is
introduced -- is not a sure sign oI the acceptance oI the new kind oI entities. ThereIore the
introduction oI such constants is not to be regarded as an essential step in the introduction oI the
Iramework. The two essential steps are rather the Iollowing. First, the introduction oI a general
term, a predicate oI higher level, Ior the new kind oI entities, permitting us to say Ior any particular
entity that it belongs to this kind (e.g., "Red is a propertv," "Five is a number"). Second, the
introduction oI variables oI the new type. The new entities are values oI these variables; the
constants (and the closed compound expressions, iI any) are substitutable Ior the variables.3 With
the help oI the variables, general sentences concerning the new entities can be Iormulated.
AIter the new Iorms are introduced into the language, it is possible to Iormulate with their help
internal questions and possible answers to them. A question oI this kind may be either empirical or
logical; accordingly a true answer is either Iactually true or analytic.
From the internal questions we must clearly distinguish external questions, i.e., philosophical
questions concerning the existence or reality oI the total system oI the new entities. Many
philosophers regard a question oI this kind as an ontological question which must be raised and
answered beIore the introduction oI the new language Iorms. The latter introduction, they believe,
is legitimate only iI it can be justiIied by an ontological insight supplying an aIIirmative answer to
the question oI reality. In contrast to this view, we take the position that the introduction oI the new
ways oI speaking does not need any theoretical justiIication because it does not imply any assertion
oI reality. We may still speak (and have done so) oI the "acceptance oI the new entities" since this
Iorm oI speech is customary; but one must keep in mind that this phrase does not mean Ior us
anything more than acceptance oI the new Iramework, i.e., oI the new linguistic Iorms. Above all, it
must not be interpreted as reIerring to an assumption, belieI, or assertion oI "the reality oI the
entities." There is no such assertion. An alleged statement oI the reality oI the system oI entities is a
pseudo-statement without cognitive content. To be sure, we have to Iace at this point an important
question; but it is a practical, not a theoretical question; it is the question oI whether or not to accept
the new linguistic Iorms. The acceptance cannot be judged as being either true or Ialse because it is
not an assertion. It can only be judged as being more or less expedient, IruitIul, conducive to the
aim Ior which the language is intended. Judgments oI this kind supply the motivation Ior the
decision oI accepting or rejecting the kind oI entities.4
Thus it is clear that the acceptance oI a linguistic Iramework must not be regarded as implying a
metaphysical doctrine concerning the reality oI the entities in question. It seems to me due to a
neglect oI this important distinction that some contemporary nominalists label the admission oI
variables oI abstract types as "Platonism."5 This is, to say the least, an extremely misleading
terminology. It leads to the absurd consequence, that the position oI everybody who accepts the
language oI physics with its real number variables (as a language oI communication, not merely as
a calculus) would be called Platonistic, even iI he is a strict empiricist who rejects Platonic
metaphysics.
A brieI historical remark may here be inserted. The non-cognitive character oI the questions which
we have called here external questions was recognized and emphasized already by the Vienna
Circle under the leadership oI Moritz Schlick, the group Irom which the movement oI logical
empiricism originated. InIluenced by ideas oI Ludwig Wittgenstein, the Circle rejected both the
thesis oI the reality oI the external world and the thesis oI its irreality as pseudo-statements;6 the
same was the case Ior both the thesis oI the reality oI universals (abstract entities, in our present
terminology) and the nominalistic thesis that they are not real and that their alleged names are not
names oI anything but merely flatus vocis. (It is obvious that the apparent negation oI a pseudo-
statement must also be a pseudo-statement.) It is thereIore not correct to classiIy the members oI the
Vienna Circle as nominalists, as is sometimes done. However, iI we look at the basic anti-
metaphysical and pro-scientiIic attitude oI most nominalists (and the same holds Ior many
materialists and realists in the modern sense), disregarding their occasional pseudo-theoretical
Iormulations, then it is, oI course, true to say that the Vienna Circle was much closer to those
philosophers than to their opponents.
4. Abstract entities in semantics
The problem oI the legitimacy and the status oI abstract entities has recently again led to
controversial discussions in connection with semantics. In a semantical meaning analysis certain
expressions in a language are oIten said to designate (or name or denote or signiIy or reIer to)
certain extra-linguistic entities.7 As long as physical things or events (e.g., Chicago or Caesar's
death) are taken as designata (entities designated), no serious doubts arise. But strong objections
have been raised, especially by some empiricists, against empiricists, against abstract entities as
designata, e.g., against semantical statements oI the Iollowing kind:
(1) "The word 'red' designates a property oI things";
(2) "The word 'color' designates a property oI properties oI things";
(3) "The word 'Iive' designates a number";
(4) "The word 'odd' designates a property oI numbers";
(5) "The sentence 'Chicago is large' designates a proposition."
Those who criticize these statements do not, oI course, reject the use oI the expressions in question,
like "red" or "Iive"; nor would they deny that these expressions are meaningIul. But to be
meaningIul is not the same as having a meaning in the sense oI an entity designated. They reject the
belieI, which they regard as implicitly presupposed by those semantical statements, that to each
expression oI the types in question (adjectives like "red," numerals like "Iive," etc.) there is a
particular real entity to which the expression stands in the relation oI designation). This belieI is
rejected as incompatible with the basic principles oI empiricism or oI scientiIic thinking.
Derogatory labels like "Platonic realism" "hypostatization," or "'Fido'-Fido principle" are attached
to it. The latter is the name given by Gilbert Ryle8 to the criticized belieI, which, in his view, arises
by a naive inIerence oI analogy: just as there is an entity well known to me, viz. my dog Fido,
which is designated by the name "Fido," thus there must be Ior every meaningIul expression a
particular entity to which it stands in the relation oI designation or naming, i.e., the relation
exempliIied by "Fido"-Fido. The belieI criticized is thus a case oI hypostatization, i.e., oI treating as
names expressions which are not names. While "Fido" is a name, expressions like "red," "Iive,"
etc., are said not to be names, not to designate anything.
Our previous discussion concerning the acceptance oI Irameworks enables us now to clariIy the
situation with respect to abstract entities as designata. Let us take as an example the statement:
(a) "'Five' designates a number."
The Iormulation oI this statement presupposes that our language L contains the Iorms oI
expressions which we have called the Iramework oI numbers, in particular, numerical variables and
the general term "number." II L contains these Iorms, the Iollowing is an analytic statement in L:
(b) "Five is a number."
Further, to make the statement (a) possible, L must contain an expression like "designates" or "is a
name oI" Ior the semantical relation oI designation. II suitable rules Ior this term are laid down, the
Iollowing is likewise analytic:
(c) "'Five' designates Iive."
(Generally speaking, any expression oI the Iorm "'. . .' designates . . ." is an analytic statement
provided the term ". . ." is a constant in an accepted Iramework. II the latter condition is not
IulIilled, the expression is not a statement.) Since (a) Iollows Irom (c) and (b), (a) is likewise
analytic.
Thus it is clear that if someone accepts the Iramework oI numbers, then he must acknowledge (c)
and (b) and hence (a) as true statements. Generally speaking, iI someone accepts a Iramework Ior a
certain kind oI entities, then he is bound to admit the entities as possible designata. Thus the
question oI the admissibility oI entities oI a certain type or oI abstract entities in general as
designata is reduced to the question oI the acceptability oI the linguistic Iramework Ior those
entities. Both the nominalistic critics, who reIuse the status oI designators or names to expressions
like "red," "Iive," etc., because they deny the existence oI abstract entities, and the skeptics, who
express doubts concerning the existence and demand evidence Ior it, treat the question oI existence
as a theoretical question. They do, oI course, not mean the internal question; the aIIirmative answer
to this question is analytic and trivial and too obvious Ior doubt or denial, as we have seen. Their
doubts reIer rather to the system oI entities itselI; hence they mean the external question. They
believe that only aIter making sure that there really is a system oI entities oI the kind in question are
we justiIied in accepting the Iramework by incorporating the linguistic Iorms into our language.
However, we have seen that the external question is not a theoretical question but rather the
practical question whether or not to accept those linguistic Iorms. This acceptance is not in need oI
a theoretical justiIication (except with respect to expediency and IruitIulness), because it does not
imply a belieI or assertion. Ryle says that the "Fido"-Fido principle is "a grotesque theory."
Grotesque or not, Ryle is wrong in calling it a theory. It is rather the practical decision to accept
certain Irameworks. Maybe Ryle is historically right with respect to those whom he mentions as
previous representatives oI the principle, viz. John Stuart Mill, Frege, and Russell. II these
philosophers regarded the acceptance oI a system oI entities as a theory, an assertion, they were
victims oI the same old, metaphysical conIusion. But it is certainly wrong to regard my semantical
method as involving a belieI in the reality oI abstract entities, since I reject a thesis oI this kind as a
metaphysical pseudo-statement.
The critics oI the use oI abstract entities in semantics overlook the Iundamental diIIerence between
the acceptance oI a system oI entities and an internal assertion, e.g., an assertion that there are
elephants or electrons or prime numbers greater than a million. Whoever makes an internal
assertion is certainly obliged to justiIy it by providing evidence, empirical evidence in the case oI
electrons, logical prooI in the case oI the prime numbers. The demand Ior a theoretical justiIication,
correct in the case oI internal assertions, is sometimes wrongly applied to the acceptance oI a
system oI entities. Thus, Ior example, Ernest Nagel in his review9 asks Ior "evidence relevant Ior
aIIirming with warrant that there are such entities as inIinitesimals or propositions." He
characterizes the evidence required in these cases -- in distinction to the empirical evidence in the
case oI electrons -- as "in the broad sense logical and dialectical." Beyond this no hint is given as to
what might be regarded as relevant evidence. Some nominalists regard the acceptance oI abstract
entities as a kind oI superstition or myth, populating the world with Iictitious or at least dubious
entities, analogous to the belieI in centaurs or demons. This shows again the conIusion mentioned,
because a superstition or myth is a Ialse (or dubious) internal statement.
Let us take as example the natural numbers as cardinal numbers, i.e., in contexts like "Here are
three books." The linguistic Iorms oI the Iramework oI numbers, including variables and the
general term "number," are generally used in our common language oI communication; and it is
easy to Iormulate explicit rules Ior their use. Thus the logical characteristics oI this Iramework are
suIIiciently clear while many internal questions, i.e., arithmetical questions, are, oI course, still
open). In spite oI this, the controversy concerning the external question oI the ontological reality oI
the system oI numbers continues. Suppose that one philosopher says: "I believe that there are
numbers as real entities. This gives me the right to use the linguistic Iorms oI the numerical
Iramework and to make semantical statements about numbers as designata oI numerals." His
nominalistic opponent replies: "You are wrong; there are no numbers. The numerals may still be
used as meaningIul expressions. But they are not names, there are no entities designated by them.
ThereIore the word "number" and numerical variables must not be used (unless a way were Iound
to introduce them as merely abbreviating devices, a way oI translating them into the nominalistic
thing language)." I cannot think oI any possible evidence that would be regarded as relevant by both
philosophers, and thereIore, iI actually Iound, would decide the controversy or at least make one oI
the opposite theses more probable than the other. (To construe the numbers as classes or properties
oI the second level, according to the Frege-Russell method, does, oI course, not solve the
controversy, because the Iirst philosopher would aIIirm and the second deny the existence oI the
system oI classes or properties oI the second level.) ThereIore I Ieel compelled to regard the
external question as a pseudo-question, until both parties to the controversy oIIer a common
interpretation oI the question as a cognitive question; this would involve an indication oI possible
evidence regarded as relevant by both sides.
There is a particular kind oI misinterpretation oI the acceptance oI abstract entities in various Iields
oI science and in semantics, that needs to be cleared up. Certain early British empiricists (e.g.,
Berkeley and Hume) denied the existence oI abstract entities on the ground that immediate
experience presents us only with particulars, not with universals, e.g., with this red patch, but not
with Redness or Color-in-General; with this scalene triangle, but not with Scalene Triangularity or
Triangularity-in-General. Only entities belonging to a type oI which examples were to be Iound
within immediate experience could be accepted as ultimate constituents oI reality. Thus, according
to this way oI thinking, the existence oI abstract entities could be asserted only iI one could show
either that some abstract entities Iall within the given, or that abstract entities can be deIined in
terms oI the types oI entity which are given. Since these empiricists Iound no abstract entities
within the realm oI sense-data, they either denied their existence, or else made a Iutile attempt to
deIine universals in terms oI particulars. Some contemporary philosophers, especially English
philosophers Iollowing Bertrand Russell, think in basically similar terms. They emphasize a
distinction between the data (that which is immediately given in consciousness, e.g., sense-data,
immediately past experiences, etc.) and the constructs based on the data. Existence or reality is
ascribed only to the data; the constructs are not real entities; the corresponding linguistic
expressions are merely ways oI speech not actually designating anything (reminiscent oI the
nominalists' flatus vocis). We shall not criticize here this general conception. (As Iar as it is a
principle oI accepting certain entities and not accepting others, leaving aside any ontological,
phenomenalistic and nominalistic pseudo-statements, there cannot be any theoretical objection to
it.) But iI this conception leads to the view that other philosophers or scientists who accept abstract
entities thereby assert or imply their occurrence as immediate data, then such a view must be
rejected as a misinterpretation. ReIerences to space-time points, the electromagnetic Iield, or
electrons in physics, to real or complex numbers and their Iunctions in mathematics, to the
excitatory potential or unconscious complexes in psychology, to an inIlationary trend in economics,
and the like, do not imply the assertion that entities oI these kinds occur as immediate data. And the
same holds Ior reIerences to abstract entities as designata in semantics. Some oI the criticisms by
English philosophers against such reIerences give the impression that, probably due to the
misinterpretation just indicated, they accuse the semanticist not so much oI bad metaphysics (as
some nominalists would do) but oI bad psychology. The Iact that they regard a semantical method
involving abstract entities not merely as doubtIul and perhaps wrong, but as maniIestly absurd,
preposterous and grotesque, and that they show a deep horror and indignation against this method,
is perhaps to be explained by a misinterpretation oI the kind described. In Iact, oI course, the
semanticist does not in the least assert or imply that the abstract entities to which he reIers can be
experienced as immediately given either by sensation or by a kind oI rational intuition. An assertion
oI this kind would indeed be very dubious psychology. The psychological question as to which
kinds oI entities do and which do not occur as immediate data is entirely irrelevant Ior semantics,
just as it is Ior physics, mathematics, economic;, etc., with respect to the examples mentioned
above.10
5. Conclusion
For those who want to develop or use semantical methods, the decisive question is not the alleged
ontological question oI the existence oI abstract entities but rather the question whether the rise oI
abstract linguistic Ioms or, in technical terms, the use oI variables beyond those Ior things (or
phenomenal data), is expedient and IruitIul Ior the purposes Ior which semantical analyses are
made, viz. the analysis, interpretation, clariIication, or construction oI languages oI communication,
especially languages oI science. This question is here neither decided nor even discussed. It is not a
question simply oI yes or no, but a matter oI degree. Among those philosophers who have carried
out semantical analyses and thought about suitable tools Ior this work, beginning with Plato and
Aristotle and, in a more technical way on the basis oI modern logic, with C. S. Peirce and Frege, a
great majority accepted abstract entities. This does, oI course, not prove the case. AIter all,
semantics in the technical sense is still in the initial phases oI its development, and we must be
prepared Ior possible Iundamental changes in methods. Let us thereIore admit that the nominalistic
critics may possibly be right. But iI so, they will have to oIIer better arguments than they have so
Iar. Appeal to ontological insight will not carry much weight. The critics will have to show that it is
possible to construct a semantical method which avoids all reIerences to abstract entities and
achieves by simpler means essentially the same results as the other methods.
The acceptance or rejection oI abstract linguistic Iorms, just as the acceptance or rejection oI any
other linguistic Iorms in any branch oI science, will Iinally be decided by their eIIiciency as
instruments, the ratio oI the results achieved to the amount and complexity oI the eIIorts required.
To decree dogmatic prohibitions oI certain linguistic Iorms instead oI testing them by their success
or Iailure in practical use, is worse than Iutile; it is positively harmIul because it may obstruct
scientiIic progress. The history oI science shows examples oI such prohibitions based on prejudices
deriving Irom religious, mythological, metaphysical, or other irrational sources, which slowed up
the developments Ior shorter or longer periods oI time. Let us learn Irom the lessons oI history. Let
us grant to those who work in any special Iield oI investigation the Ireedom to use any Iorm oI
expression which seems useIul to them; the work in the Iield will sooner or later lead to the
elimination oI those Iorms which have no useIul Iunction. Let us be cautious in making assertions
and critical in examining them, but tolerant in permitting linguistic forms.
Notes
* I have made here some minor changes in the Iormulations to the eIIect that the term "Iramework"
is now used only Ior the system oI linguistic expressions, and not Ior the system oI the entities in
question.
1 The terms "sentence" and "statement" are here used synonymously Ior declarative (indicative
propositional) sentences.
2 In my book Meaning and Necessitv (Chicago, 1947) I have developed a semantical method which
takes propositions as entities designated by sentences (more speciIically, as intensions oI
sentences). In order to Iacilitate the understanding oI the systematic development, I added some
inIormal, extra-systematic explanations concerning the nature oI propositions. I said that the term
"proposition" "is used neither Ior a linguistic expression nor Ior a subjective, mental occurrence, but
rather Ior something objective that may or may not be exempliIied in nature. . . . . We apply the
term 'proposition' to any entities oI a certain logical type, namely, those that may be expressed by
(declarative) sentences in a language" (p. 27). AIter some more detailed discussions concerning the
relation between propositions and Iacts, and the nature oI Ialse propositions, I added: "It has been
the purpose oI the preceding remarks to Iacilitate the understanding oI our conception oI
propositions. II, however, a reader should Iind these explanations more puzzling than clariIying, or
even unacceptable, he may disregard them" (p. 31) (that is, disregard these extra-systematic
explanations, not the whole theory oI the propositions as intensions oI sentences, as one reviewer
understood). In spite oI this warning, it seems that some oI those readers who were puzzled by the
explanations, did not disregard them but thought that by raising objections against them they could
reIute the theory. This is analogous to the procedure oI some laymen who by (correctly) criticizing
the ether picture or other visualizations oI physical theories, thought they had reIuted those theories.
Perhaps the discussions in the present paper will help in clariIying the role oI the system oI
linguistic rules Ior the introduction oI a Iramework Ior entities on the one hand, and that oI extra-
systematic explanations concerning the nature oI the entities on the other.
3 W.V. Quine was the Iirst to recognize the importance oI the introduction oI variables as indicating
the acceptance oI entities. "The ontology to which one's use oI language commits him comprises
simply the objects that he treats as Ialling . . . within the range oI values oI his variables." "Notes on
Existence and Necessity," Journal of Philosophv, Vol. 40 (1943), pp. 113-127; compare also his
"Designation and Existence," Journal of Philosophv, Vol. 36 (1939), pp. 702-709, and "On
Universals," The Journal of Svmbolic Logic, Vol. 12 (1947), pp. 74-84.
4 For a closely related point oI view on these questions see the detailed discussions in Herbert
Feigl, "Existential Hypotheses," Philosophv of Science, 17 (1950), pp. 35-62.
5 Paul Bernays, "Sur le platonisme dans les mathematiques" (LEnseignement math., 34 (1935), 52-
69). W.V. Quine, see previous Iootnote and a recent paper "On What There Is," Review of
Metaphvsics, Vol 2 (1948), pp. 21-38. Quine does not acknowledge the distinction which I
emphasize above, because according to his general conception there are no sharp boundary lines
between logical and Iactual truth, between questions oI meaning and questions oI Iact, between the
acceptance oI a language structure and the acceptance oI an assertion Iormulated in the language.
This conception, which seems to deviate considerably Irom customary ways oI thinking, is
explained in his article "Semantics and Abstract Objects," Proceedings of the American Academv of
Arts and Sciences, 80 (1951), 90-96. When Quine in the article "On What There Is," classiIies my
logistic conception oI mathematics (derived Irom Frege and Russell) as "platonic realism" (p. 33),
this is meant (according to a personal communication Irom him) not as ascribing to me agreement
with Plato's metaphysical doctrine oI universals, but merely as reIerring to the Iact that I accept a
language oI mathematics containing variables oI higher levels. With respect to the basic attitude to
take in choosing a language Iorm (an "ontology" in Quine's terminology, which seems to me
misleading), there appears now to be agreement between us: "the obvious counsel is tolerance and
an experimental spirit" ("On What There Is," p. 38).
6 See Carnap, Scheinprobleme in der Philosophie, das Fremdpsvchische und der Realismusstreit,
Berlin, 1928. Moritz Schlick, Positivismus und Realismus, reprinted in Gesammelte Aufsat:e, Wien,
1938.
7 See Introduction to Semantics (Cambridge, Massachusetts, 1942); Meaning and Necessitv
(Chicago, 1947). The distinction I have drawn in the latter book between the method oI the name-
relation and the method oI intension and extension is not essential Ior our present discussion. The
term "designation" is used in the present article in a neutral way; it may be understood as reIerring
to the name-relation or to the intension relation or to the extension-relation or to any similar
relations used in other semantical methods.
8 Gilbert Ryle, "Meaning and Necessity," Philosophv, 24 (1949), 69-76.
9 Ernest Nagel, "Review oI Meaning and Necessity," Journal of Philosophv, 45 (1948), 467-72.
10 WilIrid Sellars ("Acquaintance and Description Again", in Journal of Philosophv, 46 (1949),
496-504; see pp. 502 I,) analyzes clearly the roots oI the mistake "oI taking the designation relation
oI semantic theory to be a reconstruction oI being present to an experience."
Transcribed into hypertext by Andrew Chrucky, Sept. 19, 1997.