34

THE TWO USES OF LANGUAGE

The intelligible forms of ancient poets

The fair humanities of old religion . . .
They live no longer in the faith of reason:
But still the heart doth need a language, still
Doth the old instinct bring back the old names.
Coleridge, Piccolomini

There are two totally distinct uses of language. But because the
theory of language is the most neglected of all studies they are
in fact hardly ever distinguished. Yet both for the theory of
poetry and for the narrower aim of understanding much which
is said about poetry a clear comprehension of the differences
between these uses is indispensable. For this we must look
somewhat closely at the mental processes which accompany
them.
It is unfortunate but not surprising that most of the psycho-
logical terms which we naturally employ tend to blur the
distinction. ‘Knowledge’, ‘belief’, ‘assertion’, ‘thought’, and
‘understanding’, for example, as ordinarily used, are ambiguous
 the two uses of language 245

in a fashion which disguises and obscures the point which must

be brought out. They record distinctions which are oblique to
the distinctions required, they are cross-cuts of analysis made in
the wrong place and in the wrong direction, useful enough for
some purposes no doubt, but for this present purpose very con-
fusing. We shall do well to put them out of mind for a while if
possible.
The chief departure made from current conceptions in the
sketch of the mind given in Chapter Eleven lay in the substitution
of the causes, the characters and the consequences of a mental event, for
its aspects as thought, feeling and will. This treatment was introduced
with a view to the analysis which now occupies us. Among the
causes of most mental events, we urged, two sets may be dis-
tinguished. On the one hand there are the present stimuli reach-
ing the mind through the sensory nerves, and, in co-operation
with these, the effects of past stimuli associated with them. On
the other hand is a set of quite different factors, the state of the
organism, its needs, its readiness to respond to this or that kind
of stimulus. The impulses which arise take their character and
their course from the interaction of these two sets. We must keep
them clearly distinguished.
The relative importance of the two sets of factors varies enor-
mously. A sufficiently hungry man will eat almost anything
which can be chewed or swallowed. The nature of the substance,
within these limits, has very little effect upon his behaviour. A
replete person, by contrast, will only eat such things as he
expects will taste pleasant, or regards as possessing definite
beneficial properties, for example, medicines. His behaviour, in
other words, depends almost entirely upon the character of his
optical or olfactory stimulation.
So far as an impulse owes its character to its stimulus (or to
such effects of past accompanying or connected stimuli as are
revived) so far is it a reference, to use the term which we intro-
duced in Chapter Eleven, to stand for the property of mental
246 principles of literary criticism

events which we substitute for thought or cognition.1 It is plain

that the independent internal conditions of the organism usually
intervene to distort reference in some degree. But very many of
our needs can only be satisfied if the impulses are left
undistorted. Bitter experience has taught us to leave some of
them alone, to let them reflect or correspond with external states
of affairs as much as they can, undisturbed as far as possible by
internal states of affairs, our needs and desires.
In all our behaviour can be distinguished stimuli we receive,
and the ways in which we use them. What we receive may be
any kind of stimulus, but only when the reaction we make to it
tallies with its nature and varies with it in quasi-independence of
the uses we make of it does reference occur.
Those to whom visual images are of service in considering
complex matters may find it convenient at this point to imagine
a circle or sphere constantly bombarded by minute particles
(stimuli). Within the sphere may be pictured complex mechan-
isms continually changing for reasons having nothing to do with
the external stimuli. These mechanisms by opening little gate-
ways select which of the stimuli shall be allowed to come in and
take effect. So far as the subsequent convulsions are due to the
nature of the impacts and to lingering effects of impacts which
have accompanied similar impacts in the past, the convulsions
are referential. So far as they are due to the independent motions
of the internal mechanisms themselves, reference fails. This dia-
grammatic image may possibly be of convenience to some. By
those who distrust such things it may with advantage be dis-
regarded. It is not introduced as a contribution to neurology,
and is in no way a ground for the author’s view.
1
The reader who is a psychologist will notice many points in this statement at
which elaboration and qualifications are required. For example, when we are
‘introspecting’ factors normally belonging to the second set may enter the first.
But he will be able, if he grasps the general theory, to supply these complica-
tions himself. I did not wish to burden the text with unnecessary intricacies.
 the two uses of language 247

The extent to which reference is interfered with by needs and

desires is underestimated even by those who, not having yet
forgotten the events of 1914–1918, are most sceptical as to the
independence of opinions and desires. Even the most ordinary
and familiar objects are perceived as it pleases us to perceive
them rather than as they are, whenever error does not directly
deprive us of advantages. It is almost impossible for anyone to
secure a correct impression of his own personal appearance or of
the features of anyone in whom he is personally interested. Nor
is it perhaps often desirable that he should.
For the demarcation of the fields where impulse should be as
completely as possible dependent upon and correspondent with
external situation, those in which reference should take prior
place from those in which it may be subordinated to appetencies
with advantage, is not a simple matter. On many views of the
good and of what should be, themselves results of subordinating
reference to emotional satisfactions, there could be no question.
Truth, it would be said, has claims prior to all other consider-
ations. Love not grounded upon knowledge would be described
as worthless. We ought not to admire what is not beautiful and if
our mistress be not really beautiful when impartially considered
we ought, so the doctrine runs, to admire her, if at all, for other
reasons. The chief points of interest about such views are the
confusions which make them plausible. Beauty as an internal
quality of things is usually involved, as well as Good the
unanalysable Idea. Both are special twists given to some of our
impulses by habits deriving ultimately from desires. They linger
in our minds because to think of a thing as Good or Beautiful
gives more immediate emotional satisfaction than to refer to it as
satisfying our impulses in one special fashion (cf. Chapter Seven)
or another (cf. Chapter Thirty-two).
To think about Good or Beauty is not necessarily to refer to
anything. For the term ‘thinking’ covers mental operations in
which the impulses are so completely governed by internal
248 principles of literary criticism

factors and so out of control of stimulus that no reference occurs.

Most ‘thinking of’ includes reference in some degree, of course,
but not all, and similarly much reference would not commonly
be described as thinking. When we drop something which is too
hot to hold we would not usually be said to have done so
through thinking. The two terms overlap, and their definitions, if
there be a definition of ‘thinking’ as commonly used, are of
different types. This is why ‘Thought’ was on an earlier page
described as marking an oblique distinction.
To return, the claims of reference are by no means easy to
adjust with other claims. An immense extension of our powers
of referring has recently been made. With amazing swiftness
Science has opened out field after field of possible reference.
Science is simply the organization of references with a view
solely to the convenience and facilitation of reference. It has
advanced mainly because other claims, typically the claims of
our religious desires, have been set aside. For it is no accident
that Science and Religion conflict. They are different principles
upon which impulses may be organized, and the more closely
they are examined the more inevitable is the incompatibility
seen to be. Any so-called reconciliation which is ever effected
will involve bestowing the name Religion upon something
utterly different from any of the systematizations of impulses
which it now denotes, for the reason that the belief elements
present would have a different character.
Many attempts have been made to reduce Science to a position
of subjection to some instinct or emotion or desire, to curiosity
for example. A special passion for knowledge for its own sake
has even been invented. But in fact all the passions and all the
instincts, all human needs and desires may on occasion supply the
motive force for Science. There is no human activity which may
not on occasion require undistorted reference. The essential
point, however, is that Science is autonomous. The impulses
developed in it are modified only by one another, with a view to
 the two uses of language 249

the greatest possible completeness and systematization, and for

the facilitation of further references. So far as other consider-
ations distort them they are not yet Science or have fallen out of
it.
To declare Science autonomous is very different from sub-
ordinating all our activities to it. It is merely to assert that so far
as any body of references is undistorted it belongs to Science. It
is not in the least to assert that no references may be distorted if
advantage can thereby be gained. And just as there are innumer-
able human activities which require undistorted references if
they are to be satisfied, so there are innumerable other human
activities not less important which equally require distorted ref-
erences or, more plainly, fictions.
The use of fictions, the imaginative use of them rather, is not a
way of hoodwinking ourselves. It is not a process of pretending
to ourselves that things are not as they are. It is perfectly compat-
ible with the fullest and grimmest recognition of the exact state
of affairs on all occasions. It is no make-believe. But so awk-
wardly have our references and our attitudes become entangled
that such pathetic spectacles as Mr. Yeats trying desperately to
believe in fairies or Mr. Lawrence impugning the validity of solar
physics, are all too common. To be forced by desire into any
unwarrantable belief is a calamity. The state which ensues is
often extraordinarily damaging to the mind. But this common
misuse of fictions should not blind us to their immense services
provided we do not take them for what they are not, degrading
the chief means by which our attitudes to actual life may be
adjusted into the material of a long-drawn delirium.1
1
Revelation Doctrines when once given a foothold tend to interfere every-
where. They serve as a kind of omnipotent major premise justifying any and
every conclusion. A specimen: ‘Since the function of Art is to pierce through to
the Real World, then it follows that the artist cannot be too definite in his
outlines, and that good drawing is the foundation of all good art.’ – Charles
Gardner, Vision and Vesture, p. 54.
250 principles of literary criticism

If we knew enough it might be possible that all necessary

attitudes could be obtained through scientific references alone.
Since we do not know very much yet, we can leave this very
remote possibility, once recognized, alone.
Fictions whether aroused by statements or by analogous
things in other arts may be used in many ways. They may be
used, for example, to deceive. But this is not a characteristic use
in poetry. The distinction which needs to be kept clear does not
set up fictions in opposition to verifiable truths in the scientific
sense. A statement may be used for the sake of the reference, true or
false, which it causes. This is the scientific use of language. But it
may also be used for the sake of the effects in emotion and
attitude produced by the reference it occasions. This is the emotive
use of language. The distinction once clearly grasped is simple.
We may either use words for the sake of the references they
promote, or we may use them for the sake of the attitudes and
emotions which ensue. Many arrangements of words evoke
attitudes without any reference being required en route. They
operate like musical phrases. But usually references are involved
as conditions for, or stages in, the ensuing development of attitudes,
yet it is still the attitudes not the references which are import-
ant. It matters not at all in such cases whether the references
are true or false. Their sole function is to bring about and
support the attitudes which are the further response. The ques-
tioning, verificatory way of handling them is irrelevant, and
in a competent reader it is not allowed to interfere. ‘Better
a plausible impossibility than an improbable possibility’ said
Aristotle very wisely; there is less danger of an inappropriate
reaction.
The differences between the mental processes involved in the
two cases are very great, though easily overlooked. Consider
what failure for each use amounts to. For scientific language a
difference in the references is itself failure: the end has not been
attained. But for emotive language the widest differences in
 the two uses of language 251

reference are of no importance if the further effects in attitude

and emotion are of the required kind.
Further, in the scientific use of language not only must the
references be correct for success, but the connections and rela-
tions of references to one another must be of the kind which we
call logical. They must not get in one another’s way, and must be
so organized as not to impede further reference. But for emotive
purposes logical arrangement is not necessary. It may be and
often is an obstacle. For what matters is that the series of atti-
tudes due to the references should have their own proper organ-
ization, their own emotional interconnection, and this often has
no dependence upon the logical relations of such references as
may be concerned in bringing the attitudes into being.
A few notes of the chief uses of the word ‘Truth’ in Criticism
may help to prevent misunderstanding:
1. The scientific sense that, namely, in which references, and
derivatively statements symbolizing references, are true, need
not delay us. A reference is true when the things to which it
refers are actually together in the way in which it refers to them
Otherwise it is false. This sense is one very little involved by any
of the arts. For the avoidance of confusions it would be well if
the term ‘true’ could be reserved for this use. In purely scientific
discourse it could and should be, but such discourse is uncom-
mon. In point of fact the emotive power which attaches to the
word is far too great for it to be abandoned in general discus-
sion; the temptation to a speaker who needs to stir certain emo-
tions and evoke certain attitudes of approval and acceptance is
overwhelming. No matter how various the senses in which it
may be used, and even when it is being used in no sense what-
ever, its effects in promoting attitudes will still make it
indispensable; people will still continue to use the word with the
same promiscuity as ever.
2. The most usual other sense is that of acceptability. The
‘Truth’ of Robinson Crusoe is the acceptability of the things we are
252 principles of literary criticism

told, their acceptability in the interests of the effects of the

narrative, not their correspondence with any actual facts
involving Alexander Selkirk or another. Similarly the falsity of
happy endings to Lear or to Don Quixote, is their failure to be
acceptable to those who have fully responded to the rest of
the work. It is in this sense that ‘Truth’ is equivalent to
‘internal necessity’ or rightness. That is ‘true’ or ‘internally
necessary’ which completes or accords with the rest of the
experience, which co-operates to arouse our ordered response,
whether the response of Beauty or another. ‘What the Imagin-
ation seizes as Beauty must be Truth’, said Keats, using this
sense of ‘Truth’, though not without confusion. Sometimes it
is held that whatever is redundant or otiose, whatever is not
required, although not obstructive or disruptive, is also false.
‘Surplusage!’ said Pater, ‘the artist will dread that, as the run-
ner on his muscles’1 himself perhaps in this instance sweating
his sentence down too finely. But this is to make excessive
demands upon the artist. It is to apply the axe of retrench-
ment in the wrong place. Superabundance is a common
characteristic of great art, much less dangerous than the
preciousness that too contrived an economy tends to produce.
The essential point is whether what is unnecessary interferes or
not with the rest of the response. If it does not, the whole
thing is all the better probably for the extra solidity which it
thereby gains.
This internal acceptability or ‘convincingness’ needs to be con-
trasted with other acceptabilities. Thomas Rymer, for example,
refused to accept Iago for external reasons: ‘To entertain the
audience with something new and surprising against common
sense and nature, he would pass upon us a close, dissembling
rascal, instead of all open-hearted, frank, plain-dealing Souldier,
a character constantly born by them for some thousands of years

1
Essay on Style, p. 19.
 the two uses of language 253
in the World.’ ‘The truth is,’ he observes, ‘this author’s head was
full of villainous, unnatural images’.1
He is remembering no doubt Aristotle’s remark that ‘the artist
must preserve the type and yet ennoble it’, but interpreting it in
his own way. For him the type is fixed simply by convention and
his acceptances take no note of internal necessities but are gov-
erned merely by accordance with external canons. His is an
extreme case, but to avoid his error in subtler matters is in fact
sometimes the hardest part of the critic’s undertaking. But
whether our conception of the type is derived in some such
absurd way, or taken, for example, as from a handbook of
zoology, is of slight consequence. It is the taking of any external
canon which is critically dangerous. When in the same connec-
tion Rymer objects that there never was a Moorish General in the
service of the Venetian Republic, he is applying another external
canon, that of historic fact. This mistake is less insidious, but
Ruskin used to be particularly fond of the analogous mistake in
connection with the ‘truth’ of drawing.
3 Truth may be equivalent to Sincerity. This character of the
artist’s work we have already touched upon briefly in connection
with Tolstoy’s theory of communication (Chapter Twenty-
three). It may perhaps be most easily defined from the critic’s
point of view negatively, as the absence of any apparent attempt
on the part of the artist to work effects upon the reader which do
not work for himself. Too simple definitions must be avoided. It
is well known that Burns in writing ‘Ae fond kiss? was only too
anxious to escape Nancy’s (Mrs. Maclehose’s) attentions, and
similar instances could be multiplied indefinitely. Absurdly
naïve views upon the matter2 exemplified by the opinion that
Bottomley must have believed himself to be inspired or he
would not have moved his audiences, are far too common. At the

1
A Short View of Tragedy.
2
Cf. A. Clutton-Brock, The Times, 11th July 1922, p. 13.
254 principles of literary criticism

level at which Bottomley harangued any kind of exaltation in the

orator, whether due to pride or to champagne, would make his
stuff effective. But at Burns’s level a very different situation
arises. Here his probity and sincerity as an artist are involved;
external circumstances are irrelevant, but there is perhaps
internal evidence in the poem of a flaw in its creating impulse.
Compare as a closely similar poem in which there is no flaw,
Byron’s ‘When we two parted’.
 35
POETRY AND BELIEFS

What I see very well is the wide-spread, infinite harm of putt-

ing fancy for knowledge (to speak like Socrates), or rather of
living by choice in a twilight of the mind where fancy and
knowledge are indiscernible.
Euripides the Rationalist

It is evident that the bulk of poetry consists of statements which

only the very foolish would think of attempting to verify. They
are not the kind of things which can be verified. If we recall what
was said in Chapter Sixteen as to the natural generality or vague-
ness of reference we shall see another reason why references as
they occur in poetry are rarely susceptible of scientific truth or
falsity. Only references which are brought into certain highly
complex and very special combinations, so as to correspond to
the ways in which things actually hang together, can be either
true or false, and most references in poetry are not knit together
in this way.
But even when they are, on examination, frankly false, this is
no defect. Unless, indeed, the obviousness of the falsity forces