Stevenson Persuasive Definitions

Mind Association
Persuasive Definitions
Author(s): Charles Leslie Stevenson
Source: Mind, Vol. 47, No. 187 (Jul., 1938), pp. 331-350
Published by: Oxford University Press on behalf of the Mind Association
Stable URL: http://www.jstor.org/stable/2250337
Accessed: 06-03-2016 12:43 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact support@jstor.org.
Mind Association and Oxford University Press are collaborating with JSTOR to digitize, preserve and extend access to Mind.
http://www.jstor.org
This content downloaded from 141.2.134.110 on Sun, 06 Mar 2016 12:43:56 UTC
All use subject to JSTOR Terms and Conditions
IV.-PERSUASIVE DEFINITIONS.
BY CHARLES LESLIE STEVENSON.
I.
A " PERSUASIVE" definition is one which gives a new conceptual
meaning to a familar word without substantially changing its
emotive meaning, and which is used with the conscious or un-
conscious purpose of changing, by this means, the direction of
people's interests.
The object of this paper is to show that persuasive definitions
are often used in philosophy, and that the widespread failure to
recognise them for what they are-the temptation to consider
them as definitions which merely abbreviate, or which analyse,
common concepts-has led to important philosophical confusions.
Before considering philosophical examples, however, it will be
helpful to consider some simpler ones, which will serve to make
clearer what persuasive definitions are.
As an initial example let us take a definition of the word
" culture ". It will be convenient to invent pure fictions about
the linguistic habits of the people to whom the definition is ad-
dressed; for this will typify the actual situation in a way that
is free from complicating irrelevancies. Let us consider, then,
a hypothetical community in which " culture " began by having
an almost purely conceptual meaning. Let us sketch the de-
velopment of its emotive meaning, show why the emotive mean-
ing led certain people to redefine the word, and examine the way
in which this redefinition achieved its purpose.
There was once a community in which "cultured " meant
widely read and acquainted with the arts.
In the course of time these qualities came into high favour.
If one man wanted to pay another a compliment, he would dwell
at length upon his culture. It became unnatural to use " cul-
ture " in any but a laudatory tone of voice. Those who lacked
culture used the word with awe, and those who possessed it
used the word with self-satisfaction, or perhaps with careful
modesty. In this way the word acquired a strong emotive
332 CHARLES LESLIE STEVENSON:
meaning. It awakened feelings not only because of its conceptual
meaning, but more directly, in its own right; for it recalled the
gestures, smiles, and tone of voice which so habitually accom-
panied it. A public speaker, for instance, was never introduced
as " a man widely read and acquainted with the arts ". He was
described, rather, as " a man of culture ". The latter phrase
had no different conceptual meaning than the former, but was
more suitable for awakening in the audience a favourable attitude.
As the emotive meaning of the word grew more pronounced,
the conceptual meaning grew more vague. This was inevitable,
for the emotive meaning made the word suitable for use in meta-
phors. Men who were not cultured, literally, were often called
so, particularly when they were admired for having some of the
defining qualities of " culture ". At first people readily dis-
tinguished these metaphorical compliments from literal state-
ments; but as the metaphors grew more frequent, the distinction
became less clear. People weren't quite sure whether a person
must know about the arts in order to be literally cultured. Per-
haps some other kind of knowledge would serve as a substitute.
Let us now suppose that one member of the community had
no wholehearted regard for mere reading, or mere acquaintance
with the arts, but valued them only to the extent that they
served to develop imaginative sensitivity. He felt that they
were not always a reliable means to that end, and on no account
the only means. It was his constant source of regret that such
mechanical procedures as reading, or visiting museums, should
win instant praise, and that sensitivity should scarcely be noticed.
For this reason he proceeded to give " culture " a new meaning.
" I know ", le insisted, " that so and so is widely read, and
acquainted with the arts; but what has that to do with culture ?
The real meaning of ' culture ', the true meaning of ' culture ',
is imaginative sensitivity." He persisted in this statement, in
spite of the fact that " culture " had never before been used in
exactly this sense.
It will now be obvious that this definition was no mere ab-
breviation; nor was it intended as an analysis of a common
concept. Its purpose, rather, was to redirectpeople's interests.
" Culture " had and would continue to have a laudatory emotive
meaning. The definition urged people to stop using the lauda-
tory term to refer to reading and the arts, and to use it, instead,
to mean imaginative sensitivity. In this manner it sought to
place the former qualities in a poor light, and the latter in a fine
one, and thus to redirect people's admiration. When people
learn to call something by a name rich in pleasant associations,
PERSUASIVE DEFINITIONS. 333
they more readily admire it; and when they learn not to call it
by such a name, they less readily admire it. The definition made
use of this fact. It changed interests by changing names.
The past history of " culture " facilitated the change. The
emotive meaning of the word, it is true, had grown up because
of the old conceptual meaning; but it was now so firmly estab-
lished that it would persist even though the conceptual meaning
were somewhat altered. The old conceptual meaning was
easily altered, since it had been made vague by metaphorical
usage. The definition could effect a change in conceptual
meaning, then, which left the emotive meaning unaltered.
Thanks again to vagueness, the change seemed a " natural "
one, which, by escaping the attention of the hearers, did not
remind them that they were being influenced, and so did not
stultify them by making them self-conscious. The effectiveness
of the definition lay partly in this, and partly in the fact that it
made its results permanent by embedding them in people's
very linguistic habits.
The definition may be called " persuasive ", then, in a quite
conventional sense. Like most persuasive definitions, it was in
fact doubly persuasive. It at once dissuaded people from in-
discriminately admiring one set of qualities (wide reading and
acquaintance with the arts) and induced them to admire
another (imaginative sensitivity). The speaker wished to
attain both of these ends, and was enabled, by his definition, to
work for both at the same time.
There are hundreds of words which, like " culture ", have
both a vague conceptual meaning and a rich emotive meaning.
The conceptual meaning of them all is subject to constant re-
definition. The words are prizes which each man seeks to
bestow on the qualities of his own choice.
In the nineteenth century, for instance, critics sometimes
remarked that Alexander Pope was " not a poet ". The foolish
reply would be, " It's a mere matter of definition ". It is indeed
a matter of definition, but not a" mere "one. The word" poet"
was used in an extremely narrow sense. This, so far from being
idle, had important consequences; it enabled the critics to deny
to Pope a laudatory name, and so to induce people to disregard
him. A persuasive definition, tacitly employed, was at work in
redirecting interests. Those who wish to decide whether Pope
was a poet must decide whether they will yield to the critics'
influence-whether they will come to dislike Pope enough to
allow him to be deprived of an honorary title. This decision
will require a knowledge of Pope's works and a knowledge of
their own minds. Such are the important matters which lie
behind the acceptance of the tacitly proposed, narrow definition
of " poet ". It is not a matter of " merely arbitrary " definition,
then, nor is any persuasive definition " merely arbitrary ", if
this phrase is taken to imply " suitably decided by the flip of a
coin ".
Persuasive definitions are often recognisable from the words
" real " or " true ", employed in a metaphorical sense. The
speaker in our first example, for instance, was telling us what
" real " culture was, as distinct from the " shell" of culture.
The following are additional examples: " Charity ", in the true
sense of the word, means the giving not merely of gold, but of
understanding; true love is the communion between minds alone;
" Courage ", in the true sense, is strength against adverse public
opinion. Each of these statements is a way of redirecting in-
terests, by leaving the emotive meaning of the words unchanged,
and wedding it to a new conceptual one. Similarly we may
speak of the true meaning of " sportsmanship ", " genius ",
" beauty ", and so on. Or we may speak of the true meaning of
" selfishness " or " hypocrisy ", using persuasive definitions of
these derogatory terms to blame, rather than to praise. " True ",
in such contexts, is obviously not used literally. Since people
usually accept what they consider true, " true " comes to have
the persuasive force of " to be accepted ". This force is utilised
in the metaphorical expression " true meaning ". The hearer is
induced to accept the new meaning which the speaker intro-
duces.
Outside the confinements of philosophical theory the im-
portance of persuasive definitions has often been recognised.
In philology they receive occasional stress. Or rather, although
little attention is given to persuasive definitions, much is said
about the broad heading under which a study of them would
fall: the interplay between emotive and conceptual meanings
in determining linguistic change, and its correlation with in-
terests.
Leonard Bloomfield1 presents us with a particularly clear
example: " The speculative builder has learned to appeal to
every weakness, including the sentimentality, of the prospec-
tive buyer; he uses the speech forms whose content will turn
the hearer in the right direction. In many locutions ' house '
is the colorless, and ' home ' the sentimental word. Thus the
salesman comes to use the word ' home ' for an empty shell
I Language (Henry Holt, N.Y., 1933), p. 442.
that has never been inhabited, and the rest of us follow his
style."
Hanns Oertel, having stated that " the emotional element
greatly influences the fate of some words ", points out that
"amica " came to have one sense which was synonymous with
"concubina "2 To be sure there are several reasons for this.
"Concubina" had become slightly profane, too strong for
delicate ears. And " amica " permitted a convenient ambiguity.
Any shocking thoughts could always be ascribed to those who
chose to understand the word in its less innocent sense. But a
persuasive factor must also have been involved. Tact often
required people to refer to concubines without expressing con-
tempt. The word " amica ", which retained part of its old
laudatory emotive meaning in spite of its new sense, was useful
in making concubines appear less contemptible.
Persuasive definitions are too frequently encountered, however,
to have been noticed solely by the philologists. An extremely
penetrating account, in spite of its cynical turn, is given by
Aldous Huxley, in his Eyeless in Gaza:
" But if you want to be free, you've got to be a prisoner. It's the
condition of freedom-true freedom."
" True freedom ! " Anthony repeated in the parody of a clerical voice.
"I always love that kind of argument. The contrary of a thing isn't the
contrary; oh, dear me, no ! It's the thing itself, but as it truly is. Ask
any die-hard what conservatism is; he'll tell you it's true socialism. And
the brewer's trade papers; they're full of articles about the beauty of
true temperance. Ordinary temperance is just gross refusal to drink;
but true temperance, true temperance is something much more refined.
True temperance is a bottle of claret with each meal and three double
whiskies after dinner .
" What's in a name ? " Anthony went on. "The answer is, practically
everything, if the name's a good one. Freedom's a marvellous name.
That's why you're so anxious to make use of it. You think that, if you
call imprisonment true freedom, people will be attracted to the prison.
And the worst of it is you're quite right."
II.
As has been intimated, the study of persuasive definitions falls
under a much broader heading: the correlation between termin-
ology and interests. This correlation is highly complicated.
A few observations will serve to show that our account of per-
suasive definitions deals with a severely limited aspect of it.
A change in meaning may be either a cause or an effect of a
change in interest; and persuasive definitions figure only when
the change in meaning is a cause. When it is an eflect, as when
1 Lectuires on the Study of Language (Scribner's, N.Y., 1902), pp. 304, 305.
our growing disapproval of present conditions in Germany causes
us to use " fascist " as an epithet, there is not in this situation
itself any element of persuasion; although once the word has
acquired its derogatory associations, it may be used in persuasion
later on.
Our subject is still more limited in scope than this. We are
concerned with definitions which change interests. And it is
important to note that we are concerned only with some of these
definitions. Many definitions which redirect interests are not
persuasive. Interests tend to be redirected by any definition,
so long as it at all changes the meaning of a term, or selects some
one sense to the exclusion of others. When a scientist intro-
duces a technical term, in no matter how detached a manner, he
indicates his interest in what he names-his estimation of the
importance of talking about it, or of predicting its occurrence-
and he often leads his readers to have a similar interest. It
would be quite misleading to call such definitions " persuasive ".
How, then, are they to be distinguished from persuasive defi-
nitions ?
The distinction depends upon whether the term defined has
a strong emotive meaning, and upon whether the speaker em-
ploys the emotively laden word with dynamic purposes-with
the predominating intention of changing people's interests.
Men sometimes say, " I do not care what word you use, so long
as you make my distinction; " and again, " If you are not
interested in my distinction, well and good; I shall confine my
remarks to the limited set of people who are ". Definitions given
in such a spirit are not persuasive; for although they indicate
the speaker's interests, and may happen to influence the hearer's
interests, they do not utilise emotive meaning in a deliberate
efiort to sway interests.
Such a distinction is inconveniently stringent, however, and
must be slightly qualified. When a definition is given mainly
for the purposes of distinction or classification, when it is used to
guide only those interests which (like curiosity) are involved in
making the classification understood, and when it in no way
suggests that this is -the one legitimate sort of classification, then
the definition will not be called persuasive. (This is not meant to
imply that persuasive definitions are never used in scientific
writings, nor that non-persuasive definitions are based on some
rock foundation, nor that persuasive definitions are less respect-
able than others.)
We must now proceed to a further point. Persuasive defi-
nitions redirect interests by changing only the conceptual meaning
of an emotively laden term, allowing the emotive meaning to
remain roughly constant. Clearly, the opposite change is equally
important and prevalent: the emotive meaning may be altered,
the conceptual meaning remaining constant. This latter device
is no less persuasive. In fact, the same persuasive force can often
be obtained either by the one linguistic change or by the other.
In our initial example of " culture ", for instance, the speaker
used a persuasive definition. He might equally well have re-
iterated statements such as this: " Culture is only fool's gold;
the true metal is imaginative sensitivity ". This procedure
would have permitted " culture " to retain its old conceptual
meaning, but would have tended to make its emotive meaning
derogatory; and it would have added to the laudatory emotive
meaning of " imaginative sensitivity ". The same purpose
would have been served in this way that was served by the
persuasive definition. The qualities commonly referred to by
" culture " would still be placed in a poor light, and imaginative
sensitivity in a fine one; but this would have been effected by a
change in emotive meaning, rather than in conceptual meaning.
Cases of this last sort must be excluded from our account
of persuasive definitions. Although persuasive, they are not
secured through definition, but rather by, one's gestures and tone
of voice, or by rhetorical devices such as similes and metaphors.
It is expedient to restrict the word " definition " to cases where
conceptual meaning alone is being determined, or where, at least,
this aspect predominates. We must not forget, however, that
many statements which change mainly the emotive meaning of
words may, in a wider sense, be called " definitions "; and that
they, no less than persuasive definitions in our strict sense, may
easily be confused with statements that are not persuasive.
(For example: "By ' conscience ' is meant the voice of destiny.")
The remarks of the last several pages may be summarised as
follows: Persuasive definitions, so far from explaining the whole
interrelationship between terminology and interests, deal only
with the cases where change in terminology causes change in
interest, where emotive meaning and dynamic usage are involved,
and where the terminological change is in conceptual meaning
only.
There is one further clarifying remark that deserves mention.
The redirection of people's interests obviously depends upon
much more than emotive meaning. It depends as well upon
dynamic usage: upon the vigour of the speaker, his gestures,
his tone of voice, the cadence of his accompanying sentences,
his figures of speech, and so on. It is further conditioned by the
temperament of the hearers, their respect for the speaker, their
susceptibility to suggestion, their latent prejudices and ideals-
and indeed, by their factual beliefs, for a sudden change in
men's beliefs prepares the way (though often with a " lag ")
for a redirection of interests. Persuasion is seldom effective
unless the hearers are already on the point of changing their
interests. A persuasive definition may then be important as a
final impetus to the change, and as a mnemonic device, imbedded
in language, for keeping the change permanent. In dwelling
upon definitions, then, and upon the function of emotive meaning,
we have stressed but one aspect of persuasive situations. There
are excellent reasons for this stress, however. Emotive meaning
is a fairly stable element amid the widely varying set of factors
upon which effective persuasion depends, and although a partial
factor, is often essential. When a man redefines an emotively
laden term, moreover, he is very frequently endeavouring to
persuade, and takes care that the other factors necessary to
successful persuasion are fulfilled. Emotive meaning is a re-
liable sign of persuasion-permits it to be noticed. This is
important in the case of definitions, where persuasion, however
legitimate and vital in itself, can so easily acquire a spurious
appeal by masking itself in the guise of a logical analysis.
III.
SHaving explained what persuasive definitions are, let us now
see how they are important to philosophy.
We can readily begin by considering philosophic definitions of
the word " philosophy " itself. Ramsey defines it as a system
of definitions. Van der Leeuw defines it as an attempt to pene-
trate behind appearances. Their divergence is no terminological
accident. " Philosophy " is a dignified term, and each man
reserves it for the inquiry he most wishes to dignify.
Consider the word " Reality ". Philosophers often seek not
reality, but Reality, or rather, true Reality. But " true Reality ",
like " true culture ", is easily defined in many different ways,
with many different persuasive effects. Were the shadows in
Plato's cave Cc real " shadows ? Were there " real " shadows of
horses and men, as distinct from the imaginary shadows of
centaurs ? It will not do to express it so. "Real " is too im-
pressive a term to be used in describing shadows and flux; so
it must be given a restricted sense which makes it predicable
only of the eternal patterns. (When " Reality " is used by the
mystics, the effects of a tacit persuasive definition become even
more obvious.)
Why did Spinoza, so anxious to free thinking from anthropo-
morphism, nevertheless tempt his readers to anthropomorphism
by using the word " God " ? Why did he not speak always of
" The One Substance " ? One points, of course, to the political
and social forces of the times, which made a semblance of ortho-
doxy imperative. But assuredly this is not all. The word
" God " arouses, as if by magic, the very deepest of feelings.
By giving the word a new conceptual meaning, Spinoza was
enabled to direct its emotional force away from the old anthro-
pomorphic fictions, and centre it upon Substance, which he so
earnestly thought would be a more rewarding object for all our
wonder and humility. Had he said, " There is no God; nothing
but Substance and its Modes ", he would have spoken what he
believed, provided " God " was used in the popular sense. But
this would have been poor economy of the emotions. It would
have taken away the object of men's wonder and humility, pro-
viding no substitute; and so these feelings would have died, to
the great impoverishment of emotional life. The persuasive
definition of a word was needed to preserve emotional vitality.
The change in the meaning of " God " was too abrupt, however,
to escape notice. Spinoza " the atheist " was long in giving
place to Spinoza " the God-intoxicated man "; for the supporters
of orthodoxy were not slow to see that his God was God in emotive
meaning only.
These remarks are not to be misconstrued as cynical. To
point out persuasion is not necessarily to condemn it, nor to
identify all persuasion with that of a mob-orator. It is impera-
tive, however, to distinguish between persuasion and rational
demonstration.
Let us now proceed to a more recent issue. Positivism
achieved its wide appeal before Carnap's " principle of tolerance ",
and achieved it largely through the statement, " Metaphysics
is without meaning ". But isn't this remark surprisingly like
that of the nineteenth-century critics, who said that Pope was
" not a poet " ? The Positivists were stating an unquestionable
truth, in their sense of " meaning ", just as the nineteenth-century
critics were, in their sense of " poet ". The truth of such state-
ments, however, is utterly beside the point. Controversy hinges
on the emotive words that are used. Shall we define " meaning "
narrowly, so that science alone will receive this laudatory title,
and metaphysics the correspondingly derogatory one of " non-
sense " ? Shall our terminology show science in a fine light,
and metaphysics in a poor one ? Shall we, in short, accept this
persuasive definition of "meaning " 2 This is the question,
though well concealed by the dictum that definitions are
" merely arbitrary ".
But this conclusion deserves careful qualification. We must
remember that the nineteenth-century critics, to return to the
analogy, were not condemning Pope with sheer bombast. They
were also making a distinction. Their narrow sense of " poet "
had the function of stressing, in the reader's attention, certain
features common to most poetry, but lacking in Pope's. Per-
haps they meant to say this: "We have long been blind to
fundamental differences between Pope's work and that of a
Shakespeare or Milton. It is because of this blindness alone
that we have been content to give Pope a laudatory title. Let
us note the difference, thien, and deprive him of the title." The
contention of the Positivists will easily bear the same interpre-
tation. Perhaps they meant to say: " We have long been blind
to the fundamental differences between the use of sentences in
science and their use in metaphysics. It is because of this
blindness alone that we have been content to dignify metaphysics
with such titles as 'meaningful'. Let us define 'meaning',
then, in a way that will at once stress these fundamental differ-
ences, and deprive metaphysics of its title." When thus stated
the Positivistic thesis has not only heat, but light, and is not to
be scorned. And yet, perhaps there is still too much heat for
the amount of light. It is of no little service to stress the ways
in which metaphysics has been confused with science; and to
the extent that Positivists have done this, their " conquest of
metaphysics " has not depended upon exhortation. But do
their distinctions take us more than half way to a full rejection
of metaphysics ? Are we led to go the other half by the word
"nonsense ", defined so that it may cast its objectionable
emotive meaning upon metaphysics, without being predicated
of it untruthfully ?
The same question arises even when metaphysics is denied
" cognitive " meaning only. " Cognitive " is used to mean " em-
pirically verifiable or else analytic ", and with exclusive lauda-
tory import. Hence the Positivistic contention reduces to this:
" Metaphysical statements are neither empirically verifiable nor
analytic; hence they are not respectable." If metaphysicians
answer, " Our statements, even though neither empirically veri-
fiable nor analytic, are still respectable ", they are scarcely to be
led away from their position by mere exhortation.
Metaphysical impulses are too strong for hortatory treatment;
they are inhibited by it without being removed. If metaphysics
is wholly to give place to science in our- esteem, this can come
only from a closer scrutiny of both metaphysics and science.
Inquiries into verification and syntax make a good beginning,
but they are not the only points for study. It would be well to
consider how words which suggest graphic images and meta-
phors are used in the sciences, and contrast their function there
with their function in metaphysics; or to examine the psycho-
logical needs and specific confusions which lead people to think
that metaphysics is necessary. Such inquiries would direct our
attitudes toward metaphysics in a more permanent and illuminat-
ing fashion; they would shape our attitudes by clarifying and
augmenting our beliefs. If an adverse attitude to metaphysics
were prepared for in this manner, the word " nonsense ", per-
suasively defined, would be helpful in crystallising the attitude.
Such a programme seems more promising than that of the meta-
physicians. It is a pity, then, to hide its real complexity by
using a persuasive definition prematurely.
IV.
Let us now turn to ethics, with particular attention to the word
"justice ", as defined in Plato's Republic.
The first book of the Republic, it will be remembered, is largely
taken up with an argument between Socrates and Thrasymachus.
Socrates is the victor, and yet he is not content. " I have gone
from one subject to another ", he says, " without having dis-
covered what I sought first, the nature of justice. I left that
inquiry and turned away to consider whether justice is virtue
and wisdom, or evil and folly." (354, Jowett.)
Was this argument about the " virtue or evil " of justice really
an unwarranted digression ? In the light of our previous dis-
cussion, we cannot agree that it was. The argument had the
important function of determining whether or not " justice "
was to retain its laudatory emotive meaning; and this was
essential to the subsequent developments of the dialogue. When
a man is about to give a persuasive definition (and we shall see
in a moment that Socrates was) he must make sure that the
emotive meaning of the term defined is well established. Other-
wise a definition which was intended to illuminate a conceptual
meaning under a laudatory title will end by obscuring it under
a derogatory one. The word " justice ", which is a little too
stern to be wholly pleasing, is in danger of becoming derogatory,
and particularly so when men like Thrasymachus (with a
persuasive technique like that mentioned on page 337 above)
are using their oratoripal ability to make the word derogatory.
Socrates must praise justice, then, before he defines " justice ".
The question about the meaning of " justice " reappears in the
fourth book. The two intervening books have redirected our
interests by a moving description of the ideal state. These new
interests must be rendered permanent. This can be done by
dignifying the more significant aspects of the state under laudatory
titles. Of the four laudatory terms which Socrates mentions,
" wisdom ", " courage ", " temperance ", and " justice ", the
first three are readily made to serve this purpose, without great
change in their conceptual meaning. The remaining term must
be reserved for whatever else needs dignity. And so the defi-
nition of " justice " is found. "Justice of the state consists
of each of the three classes doing the work of its own class."
(441).
The persuasive character of this definition-the fact that it
forms a part of a spirited plea for a new class system, a beautiful
and inspired kind of aristocratic propaganda-can scarcely be
denied. The usual meanings of " justice" must give place to
the " true " one, to the meaning which needs the dignity of a
laudatory name.
This account would strike Plato as decidedly unfamiliar.
Yet he would disagree with it much less fundamentally than
may at first appear. Let us follow his own account, stressing
such points as bear analogy to the present one.
Plato would have agreed that the usual meaning of " justice"
was only a point for departure. We must fashion our definition
not after the common conception of justice, but after justice itself
-after the eternal Idea of justice, which we have beheld in a life
before birth, and can now- know only through careful recollection.
A definition based on common usage would disclose merely the
imperfect recollection of the Idea, as grasped by men bound to
the world of opinion.
This point of agreement seems slight, and outweighed by the
theory of recollection. But let us look more closely~. How did
Plato decide whether his recollection was correct ? Did he
consider it correct when he reached a conception which satisfied
his deepest, inmost aspirations ? Did the dialectical method
serve only to clarify his mind, so that his aspirations could be
directed to something articulate ? It is difficult to think of
any other answer. Plato aspired to the Ideas; but this was not
a consequence of some miraculous power of attraction which 'the
Ideas possessed. It was a matter of analytic necessity. Any-
thing which was not an object of his aspirations was not called
an Idea. If this is so, then our account is again close to his. If
he had consciously been making a persuasive definition, he would
still have selected, as the conceptual meaning of " justice ", the
object of these same aspirations. Nothing else would have been
granted the laudatory name. We have retained the factors which
led Plato to make his definition, without retaining the poetic
realm of the Ideas, whose function, indeed, was only to adorn
his procedure, not to alter its outcome.
If Plato's work had been less Utopian, more satirical, he would
have had recollections not from one realm of Ideas, but from two.
The first realm would have been the dwelling-place of the gods,
as described in the Phaedrus; and the second the dwelling-place
of the " author of evil " who makes his unexpected appearance
in the tenth book of the Laws. Just as aspirations would be the
criteria for correct recollection from the first realm, so aversions
would be the criteria for correct recollection from the second.
The theory of definition would then be less closely confined to the
laudatory terms. Recollection could function likewise for the
derogatory ones. But it would be of vital importance, in defining
the derogatory terms, to confine the recollection to the second
realm. The most serious philosophical errors would come from
a failure to recollect from the " correct " realm, where the
correctness of the realm would depend on the emotive meaning
of the term defined.
We must return, however, to the definition of " justice "
Plato's definition was persuasive; but this is far from being
exceptional. Later definitions of " justice ", with but few ex-
ceptions, are equally persuasive. They exert a different, kind of
influence, of course. Not all philosophers are aristocrats. But
they do exert an influence.
Let us consider Bentham's definition. " 'Justice', in the
only sense which has meaning [!], is an imaginary personage,
feigned for the convenience of discourse, whose dictates are the
dictates of utility, applied to certain particular cases." 1 More
simply stated, " This is a just law " is a hypostatic way of say-
ing, " This law contributes to the greatest happiness of the
greatest number ". Such a definition may not immediately
strike us as being persuasive, since so many of us are willing
to be led in its direction. Yet its stress on mere numbers, its
stress on counting the poor man's happiness side by side with
the rich man's, clearly marks a plea for greater democracy.
The definition propagated the ideals of a great Liberal.
1 Principles of Morals and Legislation, 1789, ch. x, sect. xl, n. 2.
By a " just " wage for labourers, it may be suggested, is meant
the wage that anticipates what labourers would get eventually,
through operation of the laws of supply and demand, if only
there were a perfect market in the economic sense. This defi-
nition conceals its persuasion quite well, making it seem to have
the detachment of a purely scientific economics. But it is a
plea, though slightly compromised, for the operation not of
economic laws, but of " natural " economic laws-that is to say,
for the operation of economic laws as they could be stated if the
purely competitive, " devil take the hindermost ", aspects of
industry were guaranteed. So you will find this definition more
pleasing to those who thrive under the present industrial con-
ditions than to those who do not.
" Justice " can be defined in a great many, ways, always with-
out shocking the lexicographers. An eye for an eye, and a
tooth for a tooth ? The keeping of contracts, merely ? The
king's will ? The distribution of social wealth in accordance
with the amount of labour that each man does ? We have a
wide choice of meanings, and freedom, within wide conventional
limits, to invent new ones. Which meaning we choose, however,
is no trivial matter; for we shall dignify that meaning by a
laudatory title. To choose a meaning is to take sides in a social
struggle.
It is curious to note that theorists have all been perturbed by
the uncertainty of ethics, and have caught glimpses, even in
moments of philosophical calm, of the element of persuasion
involved. They sought to avoid this by defining their terms,
hoping to give greater rigour and rationality to their inquiries.
Yet, ironically enough, these very definitions involved the same
persuasion; and in a way that veiled and confused it, by making
it appear to be purely intellectual analysis;
V.
The examples we have considered, whether from metaphysics,
theology, epistemology, or ethics, indicate that persuasive defi-
nitions are far from rare in philosophy, and that failure to recog-
nise their persuasive character has been responsible for much
confusion. But what, essentially, is the nature of this con-
fusion ? Largely this: Blindness to persuasion has fostered a
misunderstanding of the kind of disagreement that motivates
many disputes; and in consequence has led people to support
their contentions by far too simple a method, or to seek a definitive
method of proof where none is possible.
These methodological confusions have so far been evident
only by implication, and must now be treated more explicitly.
Let us proceed by indicating the actual complexity in methodology
which persuasive definitions introduce; for the extent to which
this complexity has been overlooked will then become obvious,
without further mention. It will be convenient to confine our
attention to the example of " justice "; but it must be remem-
bered, of course, that the same considerations arise for any case
which involves a term that is subject to persuasive definition.
The summary of methodology will be parallel to that given in
a previous paper.' The pattern of analysis there exemplified
by " good ", however, is slightly difierent from the one here
exemplified by c" justice ". The same methodological considera-
tions reappear, but we must recognise them in their new guise,
and amid additional complications.
Two men disagree about whether a certain law is just. Let
us examine the several forms which their argument may take.
(1) Suppose that both men use " just " with the same con-
ceptual meaning, namely: leading to consequences A and B.
The argument may then be resolved by use of the empirical
method. The disputants have only to see whether the law in
question leads to these consequences.
This simple case is seldom found, however. We have seen
that " justice " is constantly subject to persuasive definition,
with the result that different people come to use it in different
senses.
(2) Suppose, then, that the first man uses " just " to refer to
A and B, and the second man uses it to refer to B and C. Sup-
pose further that B is the only point of disagreement. In this
case the disputants will probably proceed without noticing the
discrepancy in their terminology, and will again find the em-
pirical method adequate. The outcome of the argument will
depend upon whether the law is or is not found to lead to B.
(3) Let us next make the same supposition as immediately
above, save that C, rather than B, is the sole point of disagreement.
The discrepancy in terminology will then probably be realised.
Yet the argument may proceed, and in some cases may be settled
empirically. If the second man, who uses " just " to refer to B
and C, is the one who denies the justice of the law, his opponent
may refute him by showing empirically that the law does lead
to C. (B is already agreed upon, by hypothesis.) " You are
I " The Emotive Meaning of Ethical Terms ", MND, vol. xlvi, N.S.,
No. 181.
23
refuted ", the first man will say, " even according to your own
faulty conception of justice."
This case raises a point which demands particular attention.
The first disputant did not refer to C, in his initial statement,
and the second disputant denied the justice of the law on account
of C alone. Hence the initial statement of the first man was at
no time contradicted by his opponent. Yet the first man will feel,
even after the discrepancy in terminology is clearly realised, that
he has been opposed from the very beginning. He will feel the
need of refuting his opponent's statement, as though this were
necessary to support his own. Why is this the case ?
This question seems puzzling only because we have attended
exclusively to conceptual meaning. We have been tacitly as-
suming that the disputants were pure scientists, motivated by a
detached curiosity. If our example is to be typical of the
majority of actual ones, this assumption is wholly unwarranted.
The use of " just " and " unjust " clearly indicated that one
disputant wasfor the law, and the other against it. They argued
for this reason, not because they were statistically minded.
They were disagreeing in interest. Each had a different kind of
interest in the law, and neither was content to let the other's
interest remain unchanged.' This kind of disagreement is
evident more from emotive meaning than from conceptual mean-
ing. The fact, then, that the conceptual meaning of the first
disputant was not contradicted did not lead him to feel that his
position was unchallenged. He wanted his opponent not merely
to acknowledge certain consequences of the law, but likewise to
praise it; and his opponent would not be praising it if he called
it " unjust ", no matter what conceptual meaning he assigned to
the term.
The disagreement in interest is most easily seen in cases like (3),
but a moment's consideration will show that it is equally present
in cases (1) and (2). The use of the laudatory term " just"
in the earlier cases indicated that they too were concerned with
whether or not the law was to be favoured. A, B, and C were
involved, of course, but no more so than in the third case, and
they were relevant for the same reason-relevant because the
disagreement in interest, which motivated the argument, was
rooted in a disagreement in belief. In other words, the dis-
putants would have the same kind of interest in the law if only
they resolved their opposing beliefs about these consequences of
it. In the first cases these opposing beliefs were about conse-
1 For a fuller analysis of disagreement in interest, see MIND, vol. xlvi,
N.S., No. 181, p. 27.
quences which both disputants referred to conceptually by the
word " just ". In the third case they were about something
which only one referred to by " just ". This is the main point
of difference between the 'cases, and it is unimportant. The
disagreement was of a sort that would terminate only when both
disputants had the same kind of interest in the law. Beliefs
were relevant only to the extent that they redirected interests.
Which beliefs did so, and whether they were expressed in the
initial statements of both opponents, determined merely the
complexity of the argument, and not its fundamental character.
These remarks prepare us for a further case:
(4) Suppose, as before, that the first man uses " just " to refer
to A and B, and the second man (who denies the justice of the
law) uses " just " to refer to B and C. Suppose further that both
have fully established that the law does lead to A and B, and
that it does not lead to C. Conceptually speaking, of course,
they have as yet located no point of disagreement, nor is there
the possibility, as in (3), of one man's refuting the other " even
according to the opponent's faulty conception of justice ". Yet
they may still argue about the justice of the law. The laudatory
force of " just ", and the derogatory force of " unjust ", are still
indicative of a disagreement in interest.
With regard to methodology this case is of particular impor-
tance. It represents a disagreement which the empirical method
may be wholly incapable of resolving.
This will be clear if we again consider, at the expense of partial
repetition, why the empirical method was decisive in the first
three cases. In each of the earlier cases the initial judgment of
one disputant was false. This was guaranteed either by the law
of contradiction or by explicit hypothesis. Each disputant,
moreover, would have had a favourable interest in the law only
so long as he believed that " just ", in his sense, was truthfully
predicable of it; for otherwise he would have used the laudatory
term in a different conceptual sense. For these reasons the
disputants had only to look to the truth of their initial statements,
and this would lead them to have the same kind of interest in
the law. In short, the disagreement in interest, which was the
mainspring of the argument, was rooted in a disagreement in
belief-in some belief which at least one of the opponents had
falsely expressed in his initial statement. The empirical method,
by upsetting this belief, would likewise resolve the disagreement
in interest.
In case (4), however, the initial statements of the opponents
are both true. The men are disposed, as above, to favour or
disfavour the law in accordance with whether " just " and," un-
just ", in the disparate senses which they employ, are truthfully
predicable of it; but an empirical inquiry will serve ,to support
both of their statements. Hence the first man will continue to
call the law " just ", with favour; and the second " unjust ",
with disfavour. Their disagreement is not rooted in some
belief which either is expressing, and may be due solely to their
different temperaments. Since the empirical method alters
interests only through altering beliefs, how can it be used to
resolve this disagreement ?
It is immediately clear that the empirical method has not the
same direct application, in (4), that it had in the earlier cases.
Yet we shall conclude too hastily if we say that there is no room
for it here at all. Let us examine further.
If case (4) continues to be disputed, persuasive definitions,
which hitherto have been responsible only for the ambiguity of
" just ", will come to play a more overt and important role.
Each man, in order to influence the other's interests, will insist
upon his own definition. They will argue about whether the
law is just in the true sense of " just ". Until they agree upon
the sense of the word they will not agree upon their fundamental
issue, namely: whether the law is to be described by a name
that indicates their praise.
The empirical method, however unavailing it may be in alter-
ing the truth of the conceptual predications which the disputants
first made, may reappear as a means of supporting their persuasive
definitions. The second disputant, for instance, may be led to
discover that C, to which he refers by " just ", has the further
consequences, F, G, and H. If he has an unfavourable interest
in these consequences, he may no longer wish to define " just "
in terms of C. If he is led to discover that A has the further
consequences I, J, and K, in which he has a favourable interest,
he may decide to use " just " to refer to A. In other words, he
may accept the definition upon which his opponent has been
insisting. Both men will then come to agree that the law is
just in a mutually accepted sense of "just ". This sense will be
a product of their wider empirical knowledge, and it will terminate
their argument not merely because they both believe that it is
trulthfully predicable, of the law, but because their mutual ac-
ceptance of it indicates that they no longer disagree in interest,
but both favour the law.
The argument in case (4) may be resolved, then, in an empirical
fashion; but we must remember that it also may not. Even
if the disputants know all the relevant consequences of the law,
one of them may still wish to praise it, and the other to condemn
it. They will be led to no common conceptual sense of " just ",
and although neither man need be stating anything false about
the law, they will continue to disagree about its justice'. The
disagreement will be one in interest, not rooted in any sort of
disagreement in belief. If resolved at all it will be resolved only
by exhortation.
It is a general truth that the empirical method can resolve
ethical disagreement, or any other kind of disagreement in in-
terest, only when this is rooted in a disagreement in belief.
The present outline of methodology has become complicated
only with regard to which beliefs are at the root of the disagree-
ment in interest-whether there are any, and if so, to what
extent they are expressed in the initial judgments. Such con-
siderations are essential in clarifying the nature of the argument,
but they are of no additional importance. This is obvious
from the fact that arguments of this sort spring from the emotive
meaning of the initial judgments, more than from the conceptual
meaning. It is evident from a further consideration: In actual
practice " just " is used so vaguely that neither disputant will
be sure which consequences are included in the definition of
" just ", and which psychologically guide him to make this
definition.
The present pattern of analysis is conveniently applicable to all
of the more specific ethical terms, and likewise to "beautiful ".
The pattern of analysis exemplified elsewhere by "good" 1 is
conveniently applicable only to the more generic ethical terms.
(It does not provide any ready means of indicating differentiae.)
But which of these patterns of analysis we select for any ethical
term is largely a matter of technical convenience. "Just"
could perhaps be treated after the manner of " good ", and
distinguished from " good " by the kind of interest involved-
though present psychological terminology does not provide a
means of making the distinction accurately. "Good " could
doubtless be treated after the manner of " just ". Moritz Schlick
made a beginning of this,2 but his failure to stress disagreement in
interest, and all that it implies, largely vitiates his account. The
same may be said, although with several qualifications, of the
original account given by Ogden and Richards,3 and of the
account given by C. D. Broad.4
MIND, vol. xlvi, N.S., No. 181.
2 Fragen der Ethik (J. Springer, Vienna, 1930), ch. i.
3 The Meaning of Meaning (Kegan Paul, 1927), p. 149.
4 " Is Goodness the Name of a Simple, Non-natural Quality ? " Pro-
ceedings of the Aristotelitan Society, 1933-34.
350 CHARLES LESLIE STEVENSON: PERSUASIVE DEFINITIONS.
The ethical terms are used so vaguely that many different
patterns of analysis are relevant to the conventional usage.
It is idle to select some one of these as the pattern of analysis.
All that is required is that the analysis clarify, whether in one
way or another, the essential features of ethical arguments.
These are emotive meaning, dynamic usage, disagreement in
interest, and an important but not definitive role for the em-
pirical method.

Stevenson Persuasive Definitions

Uploaded by

Copyright:

Available Formats

Stevenson Persuasive Definitions

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Stevenson Persuasive Definitions

Uploaded by

Copyright:

Available Formats

What is a persuasive definition according to the author?

What is a persuasive definition according to the author?

What is an example of a persuasive definition provided in the text?

What is an example of a persuasive definition provided in the text?

Mind Association

BY CHARLES LESLIE STEVENSON.

A " PERSUASIVE" definition is one which gives a new conceptual

meaning to a familar word without substantially changing its

emotive meaning, and which is used with the conscious or un-

conscious purpose of changing, by this means, the direction of

The object of this paper is to show that persuasive definitions

are often used in philosophy, and that the widespread failure to

recognise them for what they are-the temptation to consider

them as definitions which merely abbreviate, or which analyse,

common concepts-has led to important philosophical confusions.

Before considering philosophical examples, however, it will be

helpful to consider some simpler ones, which will serve to make

clearer what persuasive definitions are.

As an initial example let us take a definition of the word

" culture ". It will be convenient to invent pure fictions about

the linguistic habits of the people to whom the definition is ad-

is free from complicating irrelevancies. Let us consider, then,

a hypothetical community in which " culture " began by having

an almost purely conceptual meaning. Let us sketch the de-

velopment of its emotive meaning, show why the emotive mean-

in which this redefinition achieved its purpose.

There was once a community in which "cultured " meant

widely read and acquainted with the arts.

In the course of time these qualities came into high favour.

If one man wanted to pay another a compliment, he would dwell

at length upon his culture. It became unnatural to use " cul-

used the word with self-satisfaction, or perhaps with careful

modesty. In this way the word acquired a strong emotive

332 CHARLES LESLIE STEVENSON:

meaning. It awakened feelings not only because of its conceptual

gestures, smiles, and tone of voice which so habitually accom-

panied it. A public speaker, for instance, was never introduced

described, rather, as " a man of culture ". The latter phrase

had no different conceptual meaning than the former, but was

more suitable for awakening in the audience a favourable attitude.

As the emotive meaning of the word grew more pronounced,

the conceptual meaning grew more vague. This was inevitable,

defining qualities of " culture ". At first people readily dis-

tinguished these metaphorical compliments from literal state-

ments; but as the metaphors grew more frequent, the distinction

became less clear. People weren't quite sure whether a person

must know about the arts in order to be literally cultured. Per-

haps some other kind of knowledge would serve as a substitute.

Let us now suppose that one member of the community had

no wholehearted regard for mere reading, or mere acquaintance

served to develop imaginative sensitivity. He felt that they

were not always a reliable means to that end, and on no account

mechanical procedures as reading, or visiting museums, should

win instant praise, and that sensitivity should scarcely be noticed.

is imaginative sensitivity." He persisted in this statement, in

exactly this sense.

It will now be obvious that this definition was no mere ab-

breviation; nor was it intended as an analysis of a common

concept. Its purpose, rather, was to redirectpeople's interests.