THE MEANING OF THE WORD TRUTH [Footnote: Remarks at the meeting of the
American Philosophical Association, Cornell University, December, 1907.]
My account of truth is realistic, and follows the epistemological
dualism of common sense. Suppose I say to you 'The thing exists'--is
that true or not? How can you tell? Not till my statement has developed
its meaning farther is it determined as being true, false, or irrelevant
to reality altogether. But if now you ask 'what thing?' and I reply 'a
desk'; if you ask 'where?' and I point to a place; if you ask 'does it
exist materially, or only in imagination?' and I say 'materially'; if
moreover I say 'I mean that desk' and then grasp and shake a desk
which you see just as I have described it, you are willing to call
my statement true. But you and I are commutable here; we can exchange
places; and, as you go bail for my desk, so I can go bail for yours.
This notion of a reality independent of either of us, taken from
ordinary social experience, lies at the base of the pragmatist
definition of truth. With some such reality any statement, in order
to be counted true, must agree. Pragmatism defines 'agreeing' to mean
certain ways of 'working,' be they actual or potential. Thus, for my
statement 'the desk exists' to be true of a desk recognized as real by
you, it must be able to lead me to shake your desk, to explain myself
by words that suggest that desk to your mind, to make a drawing that is
like the desk you see, etc. Only in such ways as this is there sense in
saying it agrees with THAT reality, only thus does it gain for me the
satisfaction of hearing you corroborate me. Reference then to something
determinate, and some sort of adaptation to it worthy of the name
of agreement, are thus constituent elements in the definition of any
statement of mine as 'true'.
You cannot get at either the reference or the adaptation without using
the notion of the workings. THAT the thing is, WHAT it is, and WHICH it
is (of all the possible things with that what) are points determinable
only by the pragmatic method. The 'which' means a possibility of
pointing, or of otherwise singling out the special object; the 'what'
means choice on our part of an essential aspect to conceive it by (and
this is always relative to what Dewey calls our own 'situation');
and the 'that' means our assumption of the attitude of belief, the
reality-recognizing attitude. Surely for understanding what the word
'true' means as applied to a statement, the mention of such workings is
indispensable. Surely if we leave them out the subject and the object
of the cognitive relation float-in the same universe, 'tis true--but
vaguely and ignorantly and without mutual contact or mediation.
Our critics nevertheless call the workings inessential. No functional
possibilities 'make' our beliefs true, they say; they are true
inherently, true positively, born 'true' as the Count of Chambord was
born 'Henri-Cinq.' Pragmatism insists, on the contrary, that statements
and beliefs are thus inertly and statically true only by courtesy:
they practically pass for true; but you CANNOT DEFINE WHAT YOU MEAN by
calling them true without referring to their functional possibilities.
These give its whole LOGICAL CONTENT to that relation to reality on a
belief's part to which the name 'truth' is applied, a relation which
otherwise remains one of mere coexistence or bare withness.
The foregoing statements reproduce the essential content of the lecture
on Truth in my book PRAGMATISM. Schiller's doctrine of 'humanism,'
Dewey's 'Studies in logical theory,' and my own 'radical empiricism,'
all involve this general notion of truth as 'working,' either actual or
conceivable. But they envelop it as only one detail in the midst of much
wider theories that aim eventually at determining the notion of what
'reality' at large is in its ultimate nature and constitution.
X
THE EXISTENCE OF JULIUS CAESAR [Footnote: Originally printed under the
title of 'Truth versus Truthfulness,' in the Journal of Philosophy.]
My account of truth is purely logical and relates to its definition
only. I contend that you cannot tell what the WORD 'true' MEANS, as
applied to a statement, without invoking the CONCEPT OF THE STATEMENTS
WORKINGS.
Assume, to fix our ideas, a universe composed of two things only:
imperial Caesar dead and turned to clay, and me, saying 'Caesar really
existed.' Most persons would naively deem truth to be thereby uttered,
and say that by a sort of actio in distans my statement had taken direct
hold of the other fact.
But have my words so certainly denoted THAT Caesar?--or so certainly
connoted HIS individual attributes? To fill out the complete measure
of what the epithet 'true' may ideally mean, my thought ought to bear
a fully determinate and unambiguous 'one-to-one-relation' to its own
particular object. In the ultrasimple universe imagined the reference is
uncertified. Were there two Caesars we shouldn't know which was meant.
The conditions of truth thus seem incomplete in this universe of
discourse so that it must be enlarged.
Transcendentalists enlarge it by invoking an absolute mind which,
as it owns all the facts, can sovereignly correlate them. If it intends
that my statement SHALL refer to that identical Caesar, and that the
attributes I have in mind SHALL mean his attributes, that intention
suffices to make the statement true.
I, in turn, enlarge the universe by admitting finite intermediaries
between the two original facts. Caesar HAD, and my statement HAS,
effects; and if these effects in any way run together, a concrete medium
and bottom is provided for the determinate cognitive relation, which, as
a pure ACTIO IN DISTANS, seemed to float too vaguely and unintelligibly.
The real Caesar, for example, wrote a manuscript of which I see a
real reprint, and say 'the Caesar I mean is the author of THAT.' The
workings of my thought thus determine both its denotative and its
connotative significance more fully. It now defines itself as neither
irrelevant to the real Caesar, nor false in what it suggests of him. The
absolute mind, seeing me thus working towards Caesar through the cosmic
intermediaries, might well say: 'Such workings only specify in detail
what I meant myself by the statement being true. I decree the cognitive
relation between the two original facts to mean that just that kind of
concrete chain of intermediaries exists or can exist.'
But the chain involves facts prior to the statement the logical
conditions of whose truth we are defining, and facts subsequent to it;
and this circumstance, coupled with the vulgar employment of the terms
truth and fact as synonyms, has laid my account open to misapprehension.
'How,' it is confusedly asked, 'can Caesar's existence, a truth already
2000 years old, depend for its truth on anything about to happen now?
How can my acknowledgment of it be made true by the acknowledgment's own
effects? The effects may indeed confirm my belief, but the belief was
made true already by the fact that Caesar really did exist.'
Well, be it so, for if there were no Caesar, there could, of course,
be no positive truth about him--but then distinguish between 'true' as
being positively and completely so established, and 'true' as being so
only 'practically,' elliptically, and by courtesy, in the sense of
not being positively irrelevant or UNtrue. Remember also that Caesar's
having existed in fact may make a present statement false or irrelevant
as well as it may make it true, and that in neither case does it itself
have to alter. It being given, whether truth, untruth, or irrelevancy
shall be also given depends on something coming from the statement
itself. What pragmatism contends for is that you cannot adequately
DEFINE the something if you leave the notion of the statement's
functional workings out of your account. Truth meaning agreement with
reality, the mode of the agreeing is a practical problem which the
subjective term of the relation alone can solve.
NOTE. This paper was originally followed by a couple of paragraphs meant
to conciliate the intellectualist opposition. Since you love the word
'true' so, and since you despise so the concrete working of our ideas,
I said, keep the word 'truth' for the saltatory and incomprehensible
relation you care so much for, and I will say of thoughts that know
their objects in an intelligible sense that they are 'truthful.'
Like most offerings, this one has been spurned, so I revoke it,
repenting of my generosity. Professor Pratt, in his recent book, calls
any objective state of FACTS 'a truth,' and uses the word 'trueness'
in the sense of 'truth' as proposed by me. Mr. Hawtrey (see below, page
281) uses 'correctness' in the same sense. Apart from the general evil
of ambiguous vocabularies, we may really forsake all hope, if the term
'truth' is officially to lose its status as a property of our beliefs
and opinions, and become recognized as a technical synonym for 'fact.'