METAPHOR: A STUDY 0F
VARIOUS THEORIES AND HE PLACE
OF METAPHOR IN PHILOSOPHY ;

Dissertation for the Degree of Ru. D. '
MECNGAN STATE UNEVERSiTY '
HAROLD JOEROO?
1974 '

 

5 I LLBRARY
“zihigan Strt‘

’ixrrlty I

Ill,llllllllll ‘ ---—-11

This is to certify that the

thesis entitled

METAPHOR: A srunr OF VARIOUS THEORIES AND
THE PLACE or METAPHOR IN PHILOSOPHY

presented by

Harold Joe Rood

has been accepted towards fulfillment
of the requirements for

Ph.D. Philosophy

degree in

 

 

Major Erofessor

Date M47234

0—7 639

HUAG & saws - '

800K BiNUERl INC

LIBRARY SINGERS .1.
mm Man... 1

 

ABSTRACT

METAPHOR: A STUDY OF VARIOUS THEORIES AND
THE PLACE OF METAPHOR IN PHILOSOPHY

By
Harold Joe Rood

This paper is a detailed examination and criticism of several
current theories of metaphor which purport to provide adequate
explanations of metaphor. The major views considered are those of
Philip Nheelwright, I. A. Richards. Max Black, Paul Henle, Monroe C.
Beardsley. Colin Turbayne. and Nelson Goodman. ‘The purpose of the
consideration of these views is to determine how well each view deals
with certain questions of philosophical importance. Those questions
concern: (1) the possibility of correctly ascribing truth values to
metaphor, (2) the relationship of metaphor to literal usage, and
(3) the suitability of metaphor as an instrument in philosophical
discourse. Since it is argued that no theory considered provided an
adequate answer to these questions it became the additional purpose of
this paper to formulate a discussion which does provide such answers.

To facilitate the characterization and examination of current
theories, five types of theory of metaphor are recognized. The dis-
tinction between types of theories is based upon which of several ‘
rather obvious features a theory treats as predominant. The five

types of theories are: (l) the Tension Theory, (2) The Comparison

 
 
 

«IAT

Harold Joe Rood

Theory, (3) The Relational Theory, (4) The Absurdity Theory, and

(5) The Veridicality Theory. One Chapter is devoted to each of these
types of theories. The nature of each type of theory is delineated
and one or more representative theories is considered in each chapter.
This consideration includes both an overall evaluation and a close
analysis of the details of the theory for correctness and internal
consistency.

This paper concludes that while each type of view offers sugges-
tions which are helpful in understanding metaphor, no view considered
is adequate as a general explanation of metaphor. They all fail to
provide a general account since each construes metaphor too narrowly.
This narrow construal is not the result of faulty definition since,
generally speaking, definitions are not offered. These views base their
explanations upon sets of closely related examples. As a result,
several of the views seem plausible, provided that only their examples
are considered. But none of these theories explain the full range of
metaphors or even the examples proposed in other well-known discussions.

The final chapter presents an explanation of metaphor which avoids
the difficulties found in previous explanations and which is based upon
the arguments contained in the earlier chapters. Metaphor, it is
argued, is the use of an expression, where 'use' is understood as 'the
occurrence of a token within a particular context'. This paper estab-
lishes that metaphors are both meaningful and ammenable to truth value

assignment. The meaning or sense associated with a particular

Harold Joe Rood

metaphoric use is in part, but only in part, dependent upon convention.
Its meaning is dependent as well upon a wide variety of relationships
between objects, uses of language, feelings and opinions. The repeated
metaphoric use of an expression may establish a new convention but the
metaphors sense is not dependent upon a set of new conventions. The
paper argues that since some metaphors are both meaningful and true,
metaphor is an appropriate tool for philosophical investigation.
Metaphor, however, suffers from a ubiquitous but nonsystematic ambiguity.

One is therefor, cautioned to avoid metaphor in doing philosophy.

METAPHOR: A STUDY 0F_VARIOUS THEORIES AND
THE PLACE OF METAPHOR IN PHILOSOPHY

By
Harold Joe Rood

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Philosophy

1974

ACKNOWLEDGMENTS

A special debt of gratitude is owed to Dr. Ronald Suter, advisor
and committee chairman. His direction and criticism of earlier work
was of great assistance in the preparation of this dissertation.

I appreciate also the suggestions and criticisms of other committee
members: Dr. William J. Callaghan, Dr. George C. Kerner, and Dr.
Herbert E. Hendry.

To Dr. Charles J. McCracken, I wish to express appreciation for

his interest and encouragement on occasions when they were needed most

during my graduate program.

 

ii

TABLE OF CONTENTS

CHAPTER Page

I. INTRODUCTION ....................... 1

Method of Investigation ................ 4

Types of Views of Metaphor .............. 6

A Note on These Classifications ............ 9

Notes to Chapter I .................. 10

II. TENSION THEORIES ..................... ll

Pazzaz Theory ..................... 12

Foss ......................... l4

Wheelwright ...................... 18

Supervenience ..................... 27

Eastman ........................ 34

Conclusions Regarding the Tension Theories ...... 38

Notes to Chapter II .................. 41

III. COMPARISON THEORIES ................... 45

I. A. Richards .................... 50

Max Reiser ...................... 65

Max Black ....................... 7O

Substitution Views ................. 76

Comparison Views .................. 82

Interaction View .................. 88

Notes to Chapter III ................. 101

IV. RELATIONAL THEORY .................... 106

Paul Henle ...................... 106

Iconic View .................... 107

Conclusions in Regard to Henle's View ....... 124
Conclusions with Regard to Comparison and Relav

tional Theories ................. 126

Notes to Chapter IV .................. 130

V. ABSURDITY VIEWS ..................... 132

Monroe C. Beardsley .................. 134

Emotive Theory ................... 135

iii

 

 

.
o

1.“. .
u

TABLE OF CONTENTS--continued

CHAPTER Page
Supervenience Theory ............... 140
Literalist Theory ................. 145
Controversion Theory ............... 147
Conclusions .................... 150

Turbayne ....................... 152
Conclusions Regarding the Absurdity Theories ..... 162

Notes to Chapter V ..... . . ........... 164

VI. VERIDICALITY THEORIES .................. 166
Goodman ....................... 167
Metaphoric Truth ................. 169

Realm and Schema . . . ............. 179
Understanding Metaphor .............. 185
Conclusions with Regard to Goodman's Theory. . . . 193

Elliott and New ................... l94

Myers ........................ 200
Conclusions ..................... 203

Notes to Chapter VI ................. 207

VII. CONCLUSION ....................... 210
Restriction of 'Metaphorical' to 'Use' ........ 211

No Examples as Paradigms ............... 213
Recognition of Metaphor . .............. 214
Recognition of Figurative Use ............ 215
Context. . . . ............... 216
Confidence in the Speaker .............. 219
Relevance of Conventional Use ............ 219

Truth of Metaphorical Statements ........... 223

Literal Equivalents and Truth ............ 224
Paraphrase and Understanding ............. 228
Metaphor vs. Ambiguity ................ 230

Final Note on Truth ................. 233
Metaphor in Philosophy ................ 234

Notes to Chapter VII ................. 236
SELECTED BIBLIOGRAPHY ..................... 237

iv

 

CHAPTER I

INTRODUCTION

Metaphor has often been characterized as applying the name of
one thing to something else. Aristotle says: "Metaphor consists in

"1 It has also

giving the thing a name that belongs to something else.
been characterized as a comparison without the use of 'as' or 'like'
and as a resemblance in difference. While each of these is a concise
statement concerning the nature of metaphor, to be significant such
statements must be supported by compelling argument. Whether such
support is available is not clear. Arnold Isenberg says:

We do not have and perhaps never shall have an explicit under-

standing of metaphor. For no one has come up with anything2

fa1ntly recogn1zable as a sufficient cond1t1on of the term.
The task of determining whether any one has ever come up with an ade-
quate theory of metaphor is a question which is too broad to deal with
here. However, several contemporary views are promising. Our investi-
gation is limited to works which are currently reputed by many to have
dealt satisfactorily with the various aspects of metaphor. Our prob-
lem is to determine whether any of these is a sound theory of metaphor.
That is, we wish to determine whether any currently accepted theories
are sound.

The first question arising in a discussion of metaphor is ‘what

is metaphor'? An answer to this at the outset would seem appropriate.

But the question is ambiguous and its ambiguity often goes unnoticed.
Two reasonable interpretations of this question are: (1) How is
'metaphor' to be used? and (2) What is the nature of metaphor? If one
is clear with regard to the answer to (2) then (1) is an easy question:
its answer is: 'Metaphor' is to be used to refer to metaphors. Such
an answer has little value if the nature of metaphors is undetermined
(inasmuch as an exhaustive list is not possible). Taking the questions
in reverse order is similarly of little help. If one does not know the
use of the term 'metaphor', what is to govern his investigation?

A common solution to this sort of problem (very common indeed in the
social sciences) is to stipulate a set of characteristics and use
'metaphor' to refer to that set of things which have all and only those
characteristics. We would have a clearly defined problem and could
commence at once with a clear understanding of our use of terms.

But here we are not much interested in that sort of agreement on
terms. We are interested in discovering the nature of those utterances
which are generally called metaphors. We want to know for example
whether sentences containing metaphors can be correctly labelled 'true'
or 'false' or whether it is an error to ascribe truth value to metaphors.
He would like to know whether metaphor can be eliminated through para-
phrase without loss of meaning and how metaphor accounts in part for
the expansion of language. We want to know how we understand metaphor.

A precise definition could hasten our work in answering these
questions. If that definition were sufficiently complete we might be

able to answer the above mentioned questions by deduction alone.

However, this sort of procedure will prove ultimately to be uninforma-
tive with regard to the nature of metaphor. An example will clearly
show the difficulty forseen. Metaphor has been defined as a figure of
speech having two arms which are to be compared. To this explanation,
the following example was proposed as an objection: 'The chairman
plowed through the meeting.‘ It was pointed out that this expression
did not have two arms to be compared. In answer to the objection came
this reply: 'your example is not metaphor since it does not have two
arms.‘ While a definition would provide a convention by which to use
the term, in adopting a convention we may fail utterly to understand
both ordinary uses of 'metaphor' and the nature of metaphor.

Any proposed definition may not capture the various senses of
'metaphor' in the ordinary use or philosophical use of that term. By
examining various discussions about metaphor and considering examples
presented by others we gain an understanding of the ordinary uses of
that term, even though we state no convention.

'Metaphor' as ordinarily used is not a precise term. Black offers
the following caution:

'Metaphor' is a loose word, at best, and we must beware of

attributing to it gtricter rules of usage than are actually

found 1n practice.

Any attempt at definition at the outset could well prejudice our outcome.
He would examine only those things which fell within the scope of our
definition and conclude that our definition correctly characterized
metaphor. Until we have concluded the investigation, the nature of

metaphor must remain an open question. If it is insisted that some

 

 

 

 

 

indication of the nature of metaphorb'e‘ provided, wewi‘l'l quote from the
Oxford English Dictionary whichdefinesmetaphor as:

The figure of speech in which a name or descriptive term is

transferred to some object different from, but anaflogous to,

that to which it is prOperly“applied.4
SO long as the key terms in this definition are understood in a vague
sense it is a reasonable definition. Some would quarrel with it since
it fails to discriminate among metaphor, metonomy, and synecdoche;
others may claim that it included other expressions which are not,
strictly speaking metaphors. He would argue that it includes too little.
It does, however, give an indication of the nature of that which we

intended to investigate.

Method of Investigation

He shall proceed by a detailed examination of currently held
theories of metaphor. Generally we will limit our consideration to
theories held by persons recognized as philosophers and the theories
upon which they comnent. He shall divide the theories of metaphor into
various types to allow some degree of efficiency in following out dis-
tinct themes of explanation. This classification will direct our atten-
tion to the conunon advantages and disadvantages of several theories at
once. Such an organization will also allow us to keep in mind various
types of claims and arguments.

In examining these theories we develop an understanding of the use
Of 'metaphor' by considering examples of metaphor which are presented in

the explanation of those theories. Our critical analysis of the various

theories indicates the problems to be avoided by a correct theory.
Equally valuable is the isolation of true claims with regard to metaphor
presented by these theories.

Both Black and Beardsley consider alternatives to their views,
either to show their view to be correct or to provide a contrast to
their own view. Beardsley says:

The theory I shall propose ... will appear in its clearest light
if we first contrast it with alternative theories of metaphor....

Both these motivations for the presentation of alternative views is
sound. But neither Black nor Beardsley discuss theories which are
seriously held by anyone. While the consideration of such theories as
they present is of interest, these theories are either invented or so
poorly characterized as to describe no particular theory. Since both
Black and Beardsley are critical of the alternative views, we would like
to examine the original work of persons whom these authors believe to
hold the views described. We should like to determine whether Black
and Beardsley understand the positions which they purport to criticize.
But except for Black's discussion of the substitution view, no authors
or works are cited as defenders of the specific views presented.
Beardsley says, in regard to his effort in this matter: "It is hard to
find clearcut examples of writers who defend one or the other of these
views ...“6 although he does claim that "The chief writers are sorted
out somewhat."7 The criticisms offered by Black and Beardsley are not
criticisms of seriously defended views but rather of bare outlines

which generally contain no defense. Their theories are not shown to be

superior to other currently held views since they do not consider such

views nor are their arguments addressed to such views.

Our investigation will proceed differently. The theories which
we discuss will not in every case fit neatly in the general description
of the types we propose. However, we will discuss the theories them-
selves rather than the types. This will allow us to take full
advantage of the contributions and arguments of those theories and to
address our objections to the genuine alternatives to the view we offer.
This will provide a more accurate picture of existing philosophical
positions on metaphor and the advantages of our position over competing

explanations.

Types of Views of Metaphor

Generally, the bases of the theories which we discuss are correct
in that such bases describe some aspect which a great many metaphors
share. Often these theories can be made on the basis of the particular
aspects emphasized by a theory. By discussing an example of metaphor we
can illuminate some of these aspects. Consider:
Richard is a lion.8
This expression apparently implies a comparison between Richard and a
lion. It would not be regarded as a literal statement since Richard is
human and no lion is human. In some sense, 'Richard is a lion' is false.
0n the other hand, this sentence conveys information about the nature
of Richard's character and seems therefore to be true in some sense.
One may notice a tension here associating the referent of 'Richard' with

the referent of ’lion' or between the obVious falsehood and truth of the

expression.

 
 
 

The different characteristics we find in this metaphor are tension,
falsehood, truth and comparison. The distinction between the types
theories is a matter of which of these features of metaphor is empha-
sized, that is, which of these aspects is considered to be the basis of
metaphor.

We call a theory which emphasizes tension a 'Tension Theory'. The
advocates of these theories with whom we deal are Foss, Nheelwright,
and Eastman. Nheelwright, who is taken as the chief spokesman for this
view states: “The essence of the metaphor consists in a semantic tension
which subsists among the heterogeneous elements brought together in some

9 No brief characterization of this view

striking image or expression."
will do it justice, so we delay further characterization until we can
fully present this view.

Two types of theory emphasize the comparative aspect of metaphor,
one we call the 'Comparison Theory'. the other the 'Relational Theory'.
The comparison view treats metaphors as stating comparisons. Such a
theory would claim that the above example is a comparison between
, Richards and lions. The comparison may be intended to point out the
likenesses and differences between the objects mentioned. It may also
be an attribution of certain applicable features of the one object to
the other. On this view, the metaphor is valuable because the audience
is able to draw out of the metaphor previously unobserved likenesses and
relationships between the compared objects.

The relational view does not claim that the metaphor is a compari«

son. Rather, it holds that resemblance or some other significant

relationship among objects is our basis for understanding the metaphor.
By being familiar with these relationships one can determine the mean«
ing of the metaphor. In the example above, it would be our knowledge

of the ways in which lions were like people that we would know what was
being said about Richard. To use the language employed by Henle, the
lion mentioned by 'lion' serves as an icon for a particular kind of
person. The author of 'Richard is a lion' is saying that Richard is
that sort of person; he is not comparing Richard to lions. As represen-
tatives of the comparison view, we treat Richards, Reiser and Black.
Henle's work exemplifies the relational view.

The Absurdity Theory of metaphor centers its discussion upon the
obvious inappropriateness of a literal interpretation of metaphor. We
shall consider two fairly distinct variations of this theory. 'The first
is that of Monroe Beardsley. On his view of metaphor, our understanding
of metaphor is based upon the recognition of some absurdity in a literal
interpretation. This recognition accounts for our attempt to find a
meaningful alternative interpretation on the grounds that we think that
people try to talk sense. Henle, also, provides a discussion of this
issue and we are tempted to lump Henle and Beardsley together; But Henle
considers the 'iconic' explanation as central to his discussion and this
explanation is not shared by Beardsley.

The second variation of the absurdity view is the view of Turbayne.
He:views metaphor as a matter ofzsort-crossing and the misuse of metaphor
as a matter of sort-trespassing. Turbayne's primary interest is explain«

ing how people are misled by metaphor. The inappropriateness of

 

 

p
i

 

- u
.4” n,
-

n I -

   

 

 

 

 

 

 

 

metaphor is used by Beardsley to explain our understanding and by
Turbayne to explain our misunderstanding.

The final type of theory which we consider focuses upon the truth
of metaphors. He shall call such views veridicality theories.' We shall
consider the work of Goodman, Elliott, Mew, and Myers. We shall pre—
sent only Goodman's view as a veridicality view. We consider the other
works as following out themes suggested in our consideration of Goodman.
At the outset, Goodman is most concerned with discussing the capacity
for metaphors to be true. But he presents no single theme as do the
other views. Instead, he treats several important topics which arise
concerning the truth of metaphor. His discussion provides the basis
for the development of some central considerations in our own view of

metaphor.

A Note on These Classifications

It will be the case that some of the theories treated do not fit
neatly into a single type. We do not consider this fact to be a problem.
We do not argue, as it seems that Black and Beardsley do, that a par-
ticular type of view is in error and therefore any view of that type
mistaken. Our criticisms are leveled against the specific claims of
the individual theories. Therefore, the soundness of our argument will
rurt depend on correct classifications. The classification is helpful in
'following out the various themes used to explain metaphor. But a dis-
agreement with that classification would not entail a disagreement with

our'evaluations of the views classified.

lO

NOTES TO CHAPTER I

. Aristotle, Poetics 145767, The Basic Works of Aristotle, Edit.
Richard McKeon (New YoFk: Random House, 1966), p. 1476.

 

. Arnold Isenberg, “0n Defining Metaphor", The Journal of Philosophy,
XXX (October 1963), p. 612.

 

. Max Black, "Metaphor", Proceedings of the Aristotelian Society, 55
(1954—1955). P. 276}

 

. Oxford English Dictionary.

. Monroe C. Beardsley, Aesthetics, Problemsin the Philosophy of
Criticism (New YOFk: Harcourt, Brace and World, Inc., 1959),
p. 134.

. Ibid.

. Ibid.

. He shall later criticize this as an example of a metaphor but it will
serve here to point out characteristics which have been
attributed to metaphor.

. Phillip Hheelwright, The Burnin ‘Fountain (Bloomington: Indiana
University Press, 1963), p. 152.

I
1
u, .'
I. '
'uv ‘-
‘ \
r . .

...

p0.

 

' ’1', .
x a
u
I
A
.
'I
‘
F
.1
H
V.

 

 

CHAPTER II

TENSION THEORIES

Although a great many theories of metaphor mention energy or ten-
sion or perhaps semantic clash, only relatively few theories claim that
this is the essential feature of metaphor. The theories which we call
'tension theories' take tension to be the essential feature of metaphor
which separates it from other linguistic forms. Tension plays such an
important role in these theories because, according to such theories, it
accounts for the aesthetic value and the expressive power of metaphor.

Two of the three theories to which we give serious consideration in
this chapter are classified by Beardsley as 'supervenience' theories.
Before proceeding we shall consider our rationale for providing an

alternative basis of classification.

Consider Beardsley's characterization of the supervenience theory.

The defense of this theory begins with the observation that

poetic language, and metaphor in particular, is capable of con-
veying meanings that literal language cannot convey. Why else,
indeed, would we have need of metaphor, unless it supplied a mode
of speech for which there is no substitute? Proponents of the
supervenience theory would generally, I think, brush aside examples
like "sharp wind" as not "true" metaphors, or not the important
sort: but in Plato's metaphor of the cave or Dostoyevsky's metaphor
of the Underground, they would say the natural became capable of
bearing a supernatural meaning. Only through metaphor ... is lan-
guage freed from the restrictions of literal speech and permitted
to range abroad at will. According to this theory, the meaning

of a metaphor does not grow out of the literal meanings of its
parts, but appears as something extraneous to, and independent of
them. The literal meanings are overridden and lost: the meta-
phorical meaning is inexplicable in terms of them.

ll

,'.. ... 4

 

 

 

....

 

 

o.

(.1
‘

 

   

_ I.A
v ..-
c

‘I
'u

'p
. .'

12

We shall argue that the theories of Foss and Wheelwright do generally

fit the description of a supervenience theory and we will be interested
in examining the claims of such a theory. We will want to examine

these claims, not because they are unique to these theories, but rather
because they are the basis of claims to be made by many other theories
and are therefore in part the basis for other theories. Our decision to
classify these as tension theOries is based on the fact that the empha-
sis given by these theories to semantic tension is much greater than

that found in other theories. Many other theories emphasize the 'new
meaning' of metaphor as much as do these theories. It is the place given
to tension which makes these theories unique, not the claim that metaphor

results in a new meaning.

Pazzaz Theory

Before we begin consideration of the serious theories, we shall con-
sider a theory which, no one (we hope) holds. It is a strawman and a
weak one at that. But its obvious weakness will alert us to possible
difficulties of the tension theories of metaphor. I

We shall call our theory the pazzaz theory. According to this
theory, the essentially significant element of true metaphor is pazzaz.
Linguistic expressions having pazzaz are metaphors; those lacking it are
ruyt. The greater the pazzaz of any given metaphor the greater will be
the aesthetic pleasure one derives from it. One could examine various
expressions and point out those which have pazzaz (These would, by co-

incidence turn out to be those pieces of literature already recognized

.0
.- 'l;‘

... III
.

 

..,~ ....
l

 

 

 

 

u.. _
a .'
'0: .-

r'"’ "
p
' b.
.;.;‘
.."II
I.- .
I. I A4
.t. . -1
HI
|
l;-
_ ,- 1’1
... i
.
u
:r:c
V
w .H
I'. II
. a
's

 

 

 

4:11:
D
II
.
. ‘r
V
q. .
. P \

 

"- A
’o 'V
\ if

13

as great works; those which have little or no pazzaz would not be so
recognized.). It would be obvious that those expressions having pazzaz
were more appealing thus establishing the aesthetic relevance of pazzaz.
Pazzaz would therefore be shown to be of central importance to poetry
and other literary forms.

This strawman has a major weakness and one which would make dis-
cussion of other aspects of the theory (if any were fully presented) a
waste of effort. That weakness is that a technical term has been intro-
duced without definition or explanation. An argument might be presented
that an understanding of the term was to be gleaned from the examples
which alternately contained or failed to contain the character indi-
cated. (To save space we omitted the actual presentation of these
examples.) But of course we did not know what we were to look for.

Since those said to contain or embody pazzaz were examples of the best
literature, we may have concluded that the pazzaz was the goodness of
the literature. If one cannot explain generally what is meant by a term,
examples will be of little help since one will not know what the examples
are examples of. One will have, in effect, introduced a primitive term.
‘To introduce a primitive term as the pasis of one's explanation of meta-
phor will leave metaphor unexplained. One may well introduce 'Metaphory'
as the name of the essential characteristic of metaphor; but perhaps
the circularity would too easily be perceived.

Suppose we could clearly perceive some before unnoticed character-
istric in the examples provided. An account such as that above would

fail to justify the claim that all and only expressions which have that

 

we:

.un

‘n

2:

 

l4

characteristic are metaphors. It would seem at first consideration a
wiser move to preserve the grammarians distinction and merely note that
some similes as well as other expressions embody pazzaz and that some
metaphors lack it.

The rather absurd pazzaz view of metaphor ought to alert us to
two considerations that must be dealt with by the tension theories of
metaphor. The first: some clear explanation of the meaning of 'tension'
must be provided; the second: arguments must be offered to support
tension as the distinguishing characteristic of metaphor and not just

a coincidence in the cases noted.

Foss
Although it is Wheelwright who makes the most of the concept of

tension, he refers to the view of Foss as presented in Symbol and
2

 

Metaphor. Tension does not play quite as significant a part in the
theory of Foss as in that of Wheelwright. Nonetheless Foss states:
"Metaphor is a process of tension and energy, manifested in the process

3 and "The characterization of the

of language, not in the single word,"
metaphor as energy is excellent.... It is energy tension."4 We agree
with Wheelwright's comment in the Swanee Review. "Foss's term "energy
tension" would help to explain the difference (between metaphor and
non-metaphor), but I do not find his actual use of it very clear."5
We shall attempt to understand Foss's use of 'tension' since his use
seems to have suggested a tension theory to Wheelwright. We shall also
examine his claim that metaphor destroys the conventional meaning of

its terms .

15

Even though Foss has suggested the tension theory to Wheelwright,
one finds a difference in the use of 'tension' in reading Foss and
Wheelwright. Wheelwright, as we indicate above, bases his explanation
of metaphor on tension. Foss makes no such attempt. 'Tension' is more
aptly a description of his explanation than a basis of that explanation.
He speaks of metaphor as "blasting the symbols and shattering their
customary meaning" and of "the dynamic process of the searching, striv-
ing, penetrating mind...." He says, "It is what Aristotle aims at when
he calls the metaphor energy".6 Given the terminology with which Foss
discusses metaphor we are not surprised that he goes on to label it a
"process of tension and energy."

Foss's discussion of metaphor is tied to an epistemological theory
in which he opposes symbolic process and metaphoric process. He says:

As the word symbol is generally used in connection with scien-

tific knowledge, we must find another word to characterize the

very different kind of knowledge we are concerned with in art,
myth and religion. To this end the word metaphor will best

serve our uses, for although it has heretofore been used in a

narrower sense than it will be here....7
Since our interest here is linguistic theory rather than epistemology,
we shall not pursue further Foss's epistemology. We do consider Foss's
theory as a forerunner of Wheelwright and consider his central claim
that somehow the conventional meanings of the terms of a metaphor are
lost. Foss states:

The metaphoric process, however, is different: here the known

symbols in their relation to each other are only material; they

undergo a complete change in losing their familiar meaning in

each other and give birth to an entirely new knowledge beyond

their fixed and addible multitude ... the living creation is

born out of a destruction of the old, it is not a summing ug of
the old parts, but their entire absorption in the creation.

.1

 
 
 

16

and
It is the unique generality of the intensional process to which
the terms are sacrificed, and it is their mutual destructign in
the process out of which a new and strange 1n51ght ar1ses.
The question we wish to answer is whether, in metaphor, terms retain
their conventional meaning or whether as Foss suggests the conventional
meanings are lost. Wheelwright objects to Foss's view as follows:
In the first place, while "sacrifice" is correctly introduced,
destruction is surely too strong. If the conventional meanings
of the terms drawn into the energy tension were really destroyed,
would not the tension cease to exist.10
If one assumes, with Wheelwright and Foss that tension is an essential
feature of metaphor and if one further assumes that tension is somehow
the result of conventional meanings, then Wheelwright's objection is
valid. The destruction of the conventional meanings would result in the
elimination of the tension and therefore of the metaphor. The weakness
of this objection is that it has premises which themselves are still
open to question.
We are tempted to criticize Foss's use of 'destroyed' since it
is not clear what would count as the destruction of a meaning.
'Destruction' is certainly used here metaphorically. We shall not pur-
sue this criticism since there appears a reasonable interpretation of
Foss's use of 'destruction of meaning-' We will assume that Foss means
that the conventional meaning of the metaphorically used expression is
no longer associated with nor considered in the understanding of the
metaphor.

We are, however, able to offer an objection to Foss' view which

does not depend on Wheelwright's premises. We will argue that if the

17

conventional meanings were destroyed, the metaphor would be meaning-
less. Suppose that Foss is correct, that the conventional meanings are
destroyed in metaphor or perhaps that in their metaphoric use these
words are divorced from their conventional interpretations. We now
want to consider how it is that such metaphoric expressions are meaning-
ful. We reject the possibility that the mataphors are iconic or
stipulatively defined as a general explanation since generally metaphors

fail to fall under either of these descriptions.]]

As a matter of fact,
we are able to understand metaphor by making use of the meanings conven-
tionally associated with its terms. Notice that given that metaphors
are not icons (that is, the linguistic expressions do not resemble their
referents) if they did not have conventional meaning, then every word
would serve as well as any other in forming metaphors: it would not
matter what words were combined to form the metaphor. But surely the
words used determine in part at least whether the expression is a meta-
phor and if so which one. If there were no conventional meanings for
the terms we would not be able to determine the sense of the metaphor.
While it may be the case that the meaning of a word used metaphorically
is new (i.e., novel and unconventional) and perhaps even goes beyond
(in some sense) any combination of conventional meanings, it is not the
case that conventional meanings are irrelevant, since it is through the
conventional meaning that we understand a term's meaning as a metaphor.
While we may want to claim that a metaphor has a single meaning

distinct from conventional meanings, we may not go so far as to deny

tflmt conventional meanings are still associated with its terms.

18

Wheelwright

Wheelwright's theory of metaphor is found in two recently published

12 13

works, Metaphor and Reality, and The Burning Fountain. Both these

 

works preserve the tension theory and the supervenience view, essential-
ly as described by Beardsley. Although these themes overlap in
Wheelwright's discussion we shall first consider his view in terms of a
tension theory and then consider those aspects of his theory supporting
the supervenience view.

In Metaphor and Reality, Wheelwright proposes to deal with metaphor
by the introduction of two terms 'epiphor' and 'diaphor' as names of
distinct features of metaphor. These features amount to the consider-
ation of comparison and of unusual juxtaposition of linguistic expres-
sion, each of which is often found in metaphor. In a revised edition of'
The Burning-Fountain, Wheelwrights approach rather than contrasting
epiphor and diaphor, emphasizes a contrast between simile and plurisig—
nation.

Both Metaphor and Reality and the revised edition of The Burning

 

Fountain were published in 1968. It is reasonable to assume that
Wheelwright believed these views to be compatible. And while he empha-
sizes different aspects in the detail of his explanation they both
support the tension and supervenience views. Although bringing out
these differences in detail we shall attempt to present his work as a
single position.

Wheelwright initiates both his discussions by objecting to a

view which treats 'metaphor' as a grammatical concept.‘4

19

In Metaphor and Reality]5 he bases an argument to this effect upon the

 

following line of Robert Burns:
0, my love is like a red, red rose.
According to the classical grammarians' distinction, the presence of the
word 'like' would qualify this line as a simile. Wheelwright, however,
claims that Burns' line has nmore tensive life, more metaphoric
vitality" than the following similar line:16
Love is a red rose.
While it may be the case that the former line has more tensive life
(whatever that may be) we still have seen no argument for rejecting the
grammarian's distinction. However, if we adopt as a premise a statement
from The BurninggFountain we will produce an argument. That premise is
"The essence of metaphor consists in a semantic tension which subsists
among the heterogeneous elements brought together in some striking image

"17

or expression. If we accept this notion of semantic tension, then

the above discussion perhaps shows that the distinction based on gram-
mar alone is not sufficient to characterize metaphor.18 A similar argu-
ment is presented in The BurningiFountain:

"He fled from court like a man escaping naked from a fire."
Cicero's simile rests on a plain logical analogy, and the pleasure
which it gives, if any, is simply of intellectual recognition.

The same may be said of many tropes which have the grammatical form
of metaphor, as when Aeschylus calls a harbor the stepmother of
ships. This minor piece of wit is not metaphor in the essential
and semantic sense of the word, for it makes its connection by
analysis and labored comparison rather than by "sudden perceptions
of an objective relation." One might call it a tabloid simile.

It lacks what Martin Foss calls the "energy tension" proper to real
metaphor.l9

20

At this point Wheelwright's theory bears a striking resemblance to the
Pazzaz theory. We are as yet unclear about the meaning of 'metaphorical
tension' as we are about 'pazzaz.! Wheelwright provides no adequate
characterization of tension in either work. However, he does provide
analyses of the various examples of metaphor. Attention to these analy-
ses provides an indication of what he intends by 'tension.' We will
discover, however, that no single clear-cut meaning is indicated.
Rather his various examples indicate that the term is used in various
ways in the various examples.
We may begin to gain some feeling for his use of 'tension' by
examining his discussion of what he means by a live language.
Now what is it for language to be alive? In all organic life there
is a ceaseless but varying struggle between opposite forces, and
without such struggle the organism would go dead.... In man the
basic organic strife shows itself in various tensions, of which he
may be unconscious or at most only partly conscious--the tension
between self and other persons, between self and physical environ-
ment.... As man grapes to express his complex nature and his sense
of the complex world, he seeks On creates representational and
expressive forms ... the problem becomes that of finding suitable
word-combinations to represent some aspect or other of the per-
vasive laving tension. This, when conscious, is the basis of
poetry.2
These comments do provide some indication of Wheelwright's use of
'tension.' But they are far from a definition of that term.
Further characterization of tension is found in Metaphor and Reality
in Wheelwright's discussion of semantic motion.
What really matters in a metaphor is the psychic depth at which
the things of the world, whether actual or fancied, are transmuted

by the cool heat of the imagination. The transmutative process
that is involved may be described as semantic motion.21

21

A review of his discussion of semantic motion together with close at-
tention to his examples will provide possible interpretations of his

use of the term 'tension.‘ The semantic motion of metaphor is explained

 

in terms of two other motion words 'epiphor' and 'diaphor' "... the one
standing for the outreach and extension of meaning through comparison,
the other for the creation of new meaning by juxtaposition and synthe-
sis."22

Wheelwright attributes his use of 'epiphor' to Aristotle, recalling
that he described metaphor as being the application of the name of one
thing to another. Wheelwright's explanation of epiphor is:

Epiphoric metaphor starts by assuming a usual meaning for a word;

it then applies this word to something else on the basis 33’ and

1n order to 1nd1cate, a compar1son with what 15 familiar.
Wheelwright characterizes the movement as being from a more "graspable
image 'over on to' what is vaguer, more problematic, or more strange."23
The essential mark in epiphor, Wheelwright says, is a similarity between
objects, one of which is more well-known. Similarity alone, however, is
not sufficient to characterize metaphor. "An already obvious resemblance
would not provide any energy-tension; a steno-statement of comparison is

24

not an epiphor." The energy-tension is achieved only where the com-

parison is between dissimilars and "comes as a shock which is yet a

25 We may then consider this as our first explan-

shock of recognition."
ation of tension. It is a reaction on the part of the reader (a shock)
to the recognition of a ground of similarity between two dissimilar

objects. The Burning Fountain provides support for this analysis.

22

There is more energy-tension, I would say, in the idea of a pan-

cake moon than in the idea of a round moon, because there are

more disparate groups of qualities huddled together and thus an

element of paradox, of ambivalence, of tension between astronomi—

cal and‘breakfast-table associations, is introduced.
Though this example supports the 'shock-of-recognition'-interpretation,
it invites consideration of a further question. That question is
whether, as we suggested above, the tension is constituted by the shock
(the response of the audience) or whether it is constituted by the dif-
ference in qualities of the objects compared. Wheelwright's examples
suggest that tension is either a quality of the linguistic expression,
its referents, or a quality of the audience. He does not indicate which
of these he has in mind. But as we consider alternative characteriza-
tions of tension based on other examples we shall find that the audience
reaction to the characteristic mentioned remains a reasonable interpre-
tation of 'tension:! Wheelwright's discussion of diaphor does not
present an explanation of tension nor does it seem reasonable to attempt
to extract one. We shall postpone our analysis of that discussion
until we consider this theory as a supervenience theory.

By attending to the discussion of plurisignation and simile in
The Burnithountain we shall encounter further explanation of tension.
Wheelwright claims that energy-tension is achieved by a merging of these
aspects of metaphor. He states:

The hypothesis which I should like to put forward is that meta-

phor at its best tends to achieve fullness of semantic energy-

tension by a merging of two complementary elements--simi1e and

plurisignation. In simile, two verbal expressions each conveying

anITndividual image or idea, are joined; in plurisignation, a

single versgl expression carries two or more meanings simul-
taneously.

23

Wheelwright attempts to support this hypothesis by considering a set of
graded examples. Our interest is not so much whether he is successful
but rather in the comments which he makes about tension.29 In this
consideration he notes that those tropes which involve only simile

or only plurisignation lack metaphoric tension while the combination
tends to produce tension and therefore metaphor. He states:

A simple simile without a plurisignative depth is usually not

very interesting.... Such similes in each case depend on a single
point of reference. They offer us no fresh appercepti8n: only at
best amiable surprise of a likeness before unnoticed.3

It is evident from the explanation of his examples that at least a part
of what he calls energy tension is an openness to alternative interpre-
tation. Wheelwright considers two examples from Shakespere.

(A) To thy fair flower add the rank smell of weeds. (Sonnet 69)
(B) Times thievish progress to eternity. (Sonnet 77)

I would judge (A) to be a supressed or tabloid simile. Its expres-
siveness is limited by the implicit analogy: the beauty of youth
likened to a fair flower, and his treachery likened to the rank
smell of weeds. On reading (8) I think we feel at once the greater
expressiveness of the figure, and I do not believe that this comes
merely from the melodramatic evocations of the word "thievish" nor
from the quasi—mystical and hyperbolic lift of the word "eternity."
The greater interest of it depends more legitimately upon the way
in which the word "thievish" functions in the sentence. As con-
nected with "progress" it suggests the kind of progress a thief
might be supposed to make.... But "thievish" also connotes the
act of stealing.... The images are not suggested for their own
sake, but because they refer to, and evoke a sentiment about, the
concretely complex idea of the pang of the loss through time's
passage.3

It is clear in this example that the difference between the lines is .
primarily a difference in the number of interpretations to which it is
amenable. (A) is open to the single interpretation presented (or so his
discussion goes) while (B) is subject to a great variety of reasonable

interpretatiOns. Tension is related in part to the number of alternative

24

interpretations. His mention of what we feel ("the greater expressive-
ness") raises the question whether it is these feelings which constitute
the tension or whether 'feel' is here synonymous with 'perceive' in
which case the tension would be a matter of multiple interpretation.
Nonetheless as with the similarity among dissimilar explanations, it may
be audience response which constitutes tension.

In a somewhat confusing discussion Wheelwright suggests "... a last
approach to metaphoric tension from the side of simile ..." by consider-

ing the “... problem of poetic synaesthesia...."32

The Oxford English
Dictionary provides a definition of 'synaesthesia' somewhat related to
Wheelwright's use of this term it is: "Production, from a sense impres-
sion of one kind, of an associated mental image of a sense impression of

33 This sense of the term would label events such as

another kind."
having visual images of a particular color upon hearing certain sounds.
To avoid confusion it is important to notice that Wheelwright differs

in his use of the term. He explains 'synaesthesis' as "... the fusing
of imagery drawn from two distant sense-channels. Dante's description

of Hell as "there the sun is silent" (dove i1 sole tace) and the "Blind

 

mouths!" of Milton's Lycidas stand among the great familiar classical

34 Wheelwright does not

examples. Each depends on intersensory analogy."
explain what characteristic it is that these examples have which quali-
fies them as examples of tension.

One characteristic may be that of a fusion of dissimilars. This
would reaffirm that tension is the result of perceiving a similarity

among dissimilars. In a discussion of the two examples above,

 

h

 

 

25

Wheelwright suggests that he may view tension as an emotive matter.
He claims that a particular discussion of these examples is sufficient
since they “account for the impact of these two great cases of synaes-

"35 This discussion indi-

thesia upon a normally responsive reader....
cates that the greatness of these cases of Synaesthesia lies in their
impact upon the reader. Since these examples are presented as instruc—
tive with regard to tension, it is reasonable to assume that 'tension'
may after all refer to the emotive response of the reader. While it is
not clear which of these features, the resemblance in differences or
emotive response, Wheelwright's comments certainly warrant consideration
of 'tension' as an emotive prediction.
The view that tension refers to an emotive matter, according to
Wheelwright, is also supported by his discussion of dead metaphor.
The last example is a reminder of the fate that eventually over-
takes radical metaphors. They grow old and moribund losing the
vital tension of opposed meanings, dramatic antithesis, paradox
which was theirs at their inception. They become fossilized and
enter into everyday speech as stenosymbols which have lost their
one-time allusiveness and power to stir.
Here 'tension' as "power to stir" is an emotive term. But the discus-
sion of "opposed meanings," "dramatic antitheses," and "paradox“
indicates that perhaps the "power to stir" is a result of the tension
which is the opposition of meanings, etc. This latter interpretation
would indicate that tension is a semantic notion. It is similar to
Henle's discussion of 'clash of meanings.‘37
There is then good support for a claim that the semantic tension
of a work is constituted by the reaction of its audience (or perhaps

the way its audience generally reacts). Such an interpretation would

26

contradict Wheelwright's statement that the test of essential metaphor
is "the quality of semantic transformation that is brought about."38
Since the essence of metaphor is to be semantic tension,39 we ought to
look for semantics as the basis of tension. We could do so and consider
his reference to audience response as a discussion of the benefit of
employing a figure possessing tension. In attempting to consider ten-
sion as a semantic matter, rather than pragmatic matter, we find three
different uses of 'tension.' It is used as a label for (1) comparisons
of dissimilar objects, (2) openness to alternative interpretation, and
(3) opposition of literal meanings. All of these, which are labeled
‘tension' are features which other theorists comment on and find valu-
able. Nothing seems to be gained by introducing the term 'tension':

it only obscures the position. It obscures the position since it is
often unclear which sense of 'tension' is being used and therefore un-
clear which point is being made. If it is claimed that 'tension' refers
not just to those features but rather to all those special features of
metaphor then the position becomes circular. If tension is something
created by metaphor rather than a feature of metaphor, it is a matter of
pragmatics rather than semantics as is indicated above. If tension is
the response of the reader then Wheelwright is to be criticized for
ignoring this aspect of metaphor, since while he mentions it, he makes
no attempt to Specify what sbrt of reaCtion is 'tension.' In dis-
cussing Max Eastman's view, we shall find an attempt to carry out an
explanation of the audience reaction and a reaction which is one of ten-

sion in the conventional sense of ttension.‘

 

 

27

There is no basis in Wheelwright's work for adopting a notion of
tension. His work in fact makes us suspicious that other discussions
(H semantic tension may be conflating several aspect of metaphor and

producing a new term to handle.

Supervenience

In several arguments Wheelwright advocates the position that a new
meaning is created by the use of metaphor. He does not, however, go so
far as Foss in claiming that the conventional meaning of the terms of
the metaphor is destroyed. He claims that the new meanings allow the
poet to express that which had not before been expressed and provides
for possible expansion of language. He argues from etymological evidence
that language has in fact expanded through metaphor. Then by examining
plurisignation and diaphor he discusses the inability of literal speech
to capture the meaning of metaphor.

The importance of the expansion of language through metaphor is
expressed in terms of language adjusting so as to be able to express new
types of perceptions.

But language is also needed for dealing with the living flow of

experience which is not yet formalized into definite concepts

and which cannot be conceptionalized without suffering distortion

of character.... When we try by language to express presentative

immediacy, seeking to express by words the phenomenological object
rather than an object that has been largely fixed by conceptual
definition, it follows that the language employed must have more
flexibility in order to be relatively clgae to the fluctuating
character of the experience in question.

Though not clearly stated, this discussion seems to include the problem

of attempting to talk about a new perception; that is, a perception not

28

usually, or perhaps never before made. In such a case, no word in the
present vocabulary would have a conventional meaning which would com-
municate our experience. "Wheelwright is particularly concerned with
this matter in poetry. As an example he considers the word 'horse.‘
The general concept represented by the word "horse" results
from one kind of such grouping (of here-hows of the experienceable
world), based mainly upon publicly understood similarities....
But clearly other bases of association are possible. Any aesthetic
experience is such an aliodimensional grouping. An artist's char-
acteristic attempt, in its semantic aspect, is to express and
communicate an experience that involves a new grouping of experi-
ential moments--that is, of perceived and imagined here-nows--for
which there is no Rublicly acceptable word, formula, or other symbol
already available. 1
His particular concern with this aspect is expressed in his quote of
Ezra Pound "The function of the artist is precisely the formulation of
what has not found its way into language verbal, plastic or musical."42
It is clear from this that Wheelwright believes that through meta-
phor one is able to create expressions which have meanings which are not
expressed by conventional use of language. He states that an artist's
characteristic attempt
... in its semantic aspect, is to express and communicate an ex-
perience comprising some grouping of perceived and imagined here-
nows for which there is n3 publicly accepted word formula or other
symbol already available. 3
Our interest in this position of Wheelwright is not limited to metaphor's
effect in poetry. If he presents arguments to support his position,
he will explain how the metaphor expresses this new meaning. We may in
that case also have an explanation of metaphor in the non-artistic use
of language. We may have an explanation of the ability of language to

expand so as to express relationships which were previously unperceived.

29

Wheelwright argues that metaphor is capable of expanding language
on the basis of supposed etymological fact. He states:

It has often been remarked that essential, or functional,

or radical metaphor, is 2 chief contributor to the growth and

enrichment of language.4
In support of this claim about metaphor Wheelwright refers to the work
of Friedrich Max Miiller.45

The Sanskrit word arka, from a root meaning 'to shine,‘ comes

to signify both the sun and a hymm of praise. The splitting seems

to have occurred not by deliberate comparison and transference,

but through a mode of experience in which the visible shining and
the bursting forth of joy from the heart appeared as two manifes-

tations of one and the same effulgent reality. 5
Wheelwright suggests that shyscraper, bulldozer, arm of a chair, leaf
Of abook, are examples of English words which originated as metaphors.
He claims that these and

countless others have by now lost all trace of the semantic ten-

sion they must have had for their inventors and first users;

consequently they are no longer living metaphors, but merely ex—

metaphoric corpses, stenoeterms, units of literal language. 7
Conventional language is expanded, according to Wheelwright, when meta-
phors through repeated use lose semantic tension and become conventional.
The loss of semantic tension could be our growing accustomed to the new
use of the word so that we experience none of those qualities we earlier
contemplated as constituting tension.

If we accept this etymological argument, then we are convinced that
the number of types of uses of a particular term can be increased; old
terms can make reference where they failed to do so before. But this
argument fails to demonstrate that other words, employed conventionally

could not have been used synonymously. No evidence is provided that

30

there is available "no publicly accepted word formula or other symbol
already available." Further, the etymological discussion fails to
explain how the meaning of the metaphor is acquired or perceived in its
first use.

Wheelwright's discussion of plurisignation suggests a basis for
his claim that for at least some metaphors no literal use of language
would be synonymous.

... the poetic symbol tends characteristically to be plurisigna-

tive, in that its intellectual meanings are likely to be more or

less multiple, yet so fused as sometimes to defy any attempted
analysis into monosignitive components, and always to produce an
integral meaning that radically transcends the sum of the in-
gredient meanings.
This statement suggests that the particular combination of meanings of
a metaphor could not be produced by any non-metaphoric expression and
that the meaning of a metaphor would transcend any sum of meanings of
literal usage. He explains further:

Frequently only a single meaning is denotative and capable of

literal translation, while the remaining meanings are merely conno-

tative--felt rather than thought--and, although controlled to a

high degree by the context, do not have the kind of precision that

would enable them to be satisfactorily expressed in analytical
language.49
It is evident from this last remark concerning meanings which are felt
rather than thought that Wheelwright is claiming that no literal use of
language would have the same emotive force as the metaphor even though
it may have the same cognitive force. He provides an analysis of this
situation.

"Pray you, undo this button;" the request is plain, the denotation

single, but the connotative overtones generated by the context of

King Lear's final tragic predicament and vision, have a strange
power of suggesting more than can possibly be articulated--not

31

excluding, perhaps, a hushed arctypal reference to some undegined
ceremony of preparing for the soul's release from the body.5

Two issues are suggested. The first is that no literal expression could
have the same power of suggestion. But Wheelwright presents no argument
for such a position. The second issue suggested is a bit more subtle.
We noticed that in the above example Wheelwright has made a literal use
of speech to explain the meaning of the metaphor. Yet he claims that
metaphors have no literal equivalents. Given his attempt at literal
explanation of metaphor here and throughout both works, his position that
there are no literal equivalents has an air of contradiction. He is
either knowingly on a fool's errand when he attempts to explain metaphor
in literal terms or does not believe his claims about literal equiva—
lents.

In Metaphor and Reality, Wheelwright's discussion of diaphor
suggests the manner in which metaphor creates new meaning. He explains
his use of diaphor as follows:

The other and complementary kind of semantic motion that metaphor

engages may be called diaphor. Here the movement (phora) is

through (dia) certain particulars of experience (actual or imagined)

in a fresh way producing new meaning by juxtaposition alone.51
Given the repeated reference to new meaning this last claim is of great
importance. The words 'new meaning' suggest several different things and
we would like to understand Wheelwright's use of these terms. Perhaps
the consideration of his example and commentary will provide insight
into his use of 'new meaning.‘

My country 'tis of thee

Sweet land of liberty
Higgledy-piggledy my black hen.

32

Leaving aside questions of worth and taste, let us note that in
this combination of elements, and by their combination alone,
the writer manages to convey what is not expressed by either of
the parts. His intention is evidently to make an antipatriotic
utterance, but clearly there is nothing unpatriotic about either
the first pair of lines taken by itself nor about the third line
taken by itself. The antipatriotic sentiment is expressed solely
by their combination.52
This example is clear enough with regard to the meaning of 'diaphor.‘
It is not very helpful with regard to the notion of a new meaning.
Wheelwright's observation that the lines taken together mean something
more than any of the lines taken separately is an observation which
could be made with reSpect to most prose. Consider for example these
lines: "He ran. He ran. He ran.‘ The three sentences together say
something that none of them say alone. If this is all that Wheelwright
means by 'new meaning' then having a new meaning is not a special
characteristic of metaphor but is a feature of language in general.
The claim that metaphor results in a new meaning loses much of its import.
The new meanings were to provide the motivation for employing metaphor.
If the same result is attainable through literal usage, metaphor could
well be abandoned. The new meanings were, according to Wheelwright, of
great significance since they were meanings which were not expressable
in the literal use of speech. But if this is his claim, his conclusion
is premature. He says “The anti-patriotic sentiment is expressed solely
by their combination." His example may be in some ways a more interest-
ing expression of contempt for the views expressed in "My Country 'tis
of Thee," but there are many others: for example, 'I have great contempt
for the sentiments expressed in "My Country 'tis of Thee".' That the

diaphor would be more pleasing would be a ground for employing it, but

this in no way supports the claim that a new meaning is created.

33

To support his assertion that new meanings can arise from mere
juxtaposition he makes reference to "... the broad ontological fact that
new qualities and new meanings can emerge, simply come into being, out

"53 He invites the

of some hitherto ungrouped combination of elements.
reader to consider a time at which Hydrogen and Oxygen had never com-
bined later followed by the coming together of these elements to form
water. Wheelwright claims that "Analogous novelties occur in the sphere

of meanings as well."54

While his analogy makes a pleasing image, it
is hardly an argument which we find pursuasive. It fails to provide a
basis to suppose that linguistic expressions act as do chemical elements.
Neither his example of diaphor nor his analogy provide a basis for hold-
ing that new meanings are created by diaphor.

It is clear that Wheelwright in part holds a supervenience theory
as described by Beardsley. According to Wheelwright the aim of the
poet is the expression of that which could not be expressed literally.
He holds that metaphor allows us to express and communicate perceived
or imagined here-hows for which there are no conventional signs. These
new meanings, he claims, are the result in part of diaphor and plurisig-
nation. We find, however, that many of these broad claims have no
backing. In particular we find no support for his claim that the mean-
ings of metaphor cannot be accurately communicated by the literal use
of terms. The etymological facts whichhe represents only establishes
“that words may be used in ways which they had not been previously.
'This does not establish as a fact that no literal expression could have

55

been synonymous. Wheelwright's discussion of plurisignation made

34

reference to the emotive features of language; but still no argument
was presented that no literal expression has that same emotive force.
Wheelwright, we find, is convinced of two things: (1) Metaphor
creates new meanings and (2) tension is the most significant aspect of
metaphor. The first of these claims is generally unsupported if by
'new meaning' Wheelwright means a meaning which cannot be expressed by
literal expression. The second claim is unhelpful and the use of

'tension' tended to obscure rather than clear matters.

Eastman

We include the work of Max Eastman56

as a tension theory, not
because he employs 'tension' in his explanation but because his explana-
tions turn upon a psychological phenomenon which might reasonably be
called tension, and because it is significantly treated by Richards
whose work we consider. Eastman's discussion of metaphor is a part of
his discussion of and attempt to define poetry. He determines that
poetry is an attempt to convey the quality of an experience.
And to make us clearly and intensely, or richly, or vividly con—
scious of this quality is the whole of his [the poet's] effort.
It is an effort ... to heighten consciousness. Pure poetry is
the pure effort to heighten consciousness.
Metaphor is an aspect of poetry to which Eastman attributes this con-
sciousness-heightening ability. Although objecting in general to the
grammarians use of terms he indicates that he takes metaphor to mean
"... [calling an experience] by the name of some other experience that

58 He believes that metaphors could be "... described

59

is similar."

adequately as figures of comparison."

35

Given the characterizations of metaphor as a comparison and the
aim of poetry as the heightening of consciousness, Eastman is presented
with the question of the value of metaphor in poetry. What has compare
ing dissimilar objects to do with heightening consciousness? His answer
is based on a psychological theory that interference with a task height—
ens awareness. Metaphor accomplishes this since metaphor, according to
Eastman, tends to obscure meaning. He states "... those swift figures
of metonymy and metaphor ... do not help explain like maps or illustra-
tions, but rather obscure the meaning of the sentence in which they
occur...."60
Eastman argues that to have one's consciousness or awareness
heightened one needs only to be presented with a problem. He uses as
an example putting on a coat. Ordinarily he says this is done without
much thought and one is not conscious of the motions through which he
is going. However, should he be unable to get his arm into the sleeve
he becomes aware of what is happening and what he is doing. He main-
tains this awareness until the problem is solved and then goes on as
before not noticing what takes place.
I do not know how you could better define consciousness in that
case, than to say it was the process of resolving the experience
of a coat-with-an-impenetrable-sleeve into the experience of a
coat-with-a-hole-in-the-lining. It was identifying an unfamiliar
experience with a familiar one for the purposes of action. And
agesgggsgsofiggzsld$gglggcgfion was accomplished and action resumed,
From this example he generalizes that any situation which blocks
our progress heightens consciousness. He states:
It seems that consciousness is, arises out of, or depends upon,

two things--a blockage of action, and an identificatiog of one
experience with another so that action may be resumed. 2

36

According to Eastman's view, reading literature would be an action and
metaphor would be a blocking of that action. The action would be
blocked by the obscure meaning of metaphor which derives from unprac«
tical identifications.

What could a person do who desired to heighten consciousness or
intensify, or preserve or prolong, or in any other way cultivate
it for its own sake--what could he do-—that would be more funda-
mental than to suggest impractical identifications? Poetic meta-
phor is the employment of words to suggest impractical identifica—
tions. You may choose this or that identification for any one of
a thousand reasons, all of them very interesting but none general
enough to enter into the definition of poetry. Ahy_impractical
identification that you can induce somebody to listen to is
poetic, because it is the essence of an attentive consciousness.
It is mind suspended on the brink of action.63

 

It is this suspension at the brink of action which seems to us to fit
the meaning of 'tension.' The tension created by the inability to pro-
ceed creates a heightened consciousness.

Eastman's discussion, if sound, would strengthen Wheelwright's
claim that tension was the central factor in metaphor. And the situa-
tion Eastman describes fits well with Wheelwright's explanation of
tension which results because of "... desperate groups of qualities
huddled together.“64

Eastman's psychological theory seems plausible and some amount of
experimentation might provide sufficient confirmation to induce us to
adopt it. But his aesthetic theory is clearly inadequate. Consider
"Pure poetry is the pure effort to heighten consciousness....“ This is
clearly not an adequate statement upon which to base a theory of meta-

phor. Such an explanation standing alone would make the man who turns

the wheel on the rack a poet so long as his interest was restricted to

37

heightening consciousness. Bamboo shoots under the fingernails hardly
seem to constitute a poem whatever the motivation. It may be that the
value of poetry lies in heightening consciousness. But if that is so
it must heighten it in some special way. The placement of impediments
to one's action hardly seems to be the sort of heightening required.
Such impediments rather than heighten consciousness with regard to the
poem are often the reason poetry is not read. Some explanation is
needed in regard to why one would undertake to read the work in the
first place. One's consciousness is heightened only when there is
interest in completing the task. It would be strange to say that people
read poetry for the frustration they experience from it.

One point, which Eastman mentions but makes little of, may be
salvaged. That point is that the identification of the perplexing situ-
ation as’a familiar situation which allows one to go on often gives
positive satisfaction. An important contribution to the aesthetic value
of metaphor may indeed be in solving the riddle it poses (so long as
the riddle does not so inhibit the reading that one loses track of the
theme). This may be one aspect of metaphor which makes it pleasing,
but it is the aspect which makes the solution to any puzzle enjoyable.
While such solutions may enhance poetry they cannot be the entirenbasis
of metaphor in poetry.

Eastman is correct about metaphor presenting obscure meanings and
while this may cause the reader to experience a certain tension, it seems
that it is not a tension that would be enjoyed. It is not a tension
which would account for the importance which Wheelwright places of meta-

phor.

<
a

 

 

38

ConclusionsRegarding the Tension Theories

In reading the work on metaphor of those whose interest is in
literature one often finds reference to tension, energy, life, and simi-
lar qualities of literature, none of which appear to be the qualities
of linguistic expressions. Predications such as tension, life, etc.,
were, at least in their first such uses, metaphors and a student in an
introduction to literature class may very well experience the blockage
of action about which Eastman speaks. The blockage is the result of
attempting to understand literature and finding it discussed in terms
which seem non-applicable to literature-ascribing properties which seem
incommensurate with the other more obvious qualities of literature.
Such a student wants to know what qualities a work has which entitles
it to be called lively. The mere presentation of works which are lively
is of little help in discovering the relevant qualities. In examining
the tension theories we investigated the use made of 'tension' in an
effort to discover those features of a metaphor to which 'tension'
refers. In the work of Foss we found tension to be an apt description
of the terms in which he explained metaphor. He used such phrases as
"blasting the symbols and shattering their customary meaning." Such
events would seem tense enough, if 'blasting' or 'shattering' were not
also used metaphorically. But those words are used metaphorically.
Nothing is blasted or shattered in the literal sense of those terms.
His explanation would be clearer if he discussed symbols and meanings
using literal speech. Such an explanation would claim that in its meta-

phoric use an expression: (1) has a sense which it did not conventionally

 

_..n
-\V

 

 

, .
.1
I. ‘
... c
i
-
... -
uL .
ood-
”or“
-
DI I
..P
a
0‘
""P
....
t ...
-«
’l“.

n
...

.c-.'

51.,

 

 

‘l-
v
‘
...
n
'o
u
.
W
a .
'I
ul“
-
e
I
i
...
o.‘
I‘;
‘9.
‘1
PI‘
v
’n
v

39

have and (2) fails to have a sense with which it is conventionally
associated. This sort of explanation clearly applies to linguistic
expressions where blasting and shattering do not. While talk of blast-
ing meanings may suggest tension, such talk seems only to confuse the
issue and induce one to wonder whether the author's point has been
correctly interpreted.

In Wheelwright's work we encountered a similar situation. The
term 'tension.' still undefined, was used in connection with the
description of several aspects of metaphor, but generally left open the
question of whether these aspects were the tension or, whether these
aspects of linguistic expression caused the reader to be tense. In
either case, the use of 'tension' is in need either of further defini-
tion or, more probably, elimination since it is a metaphoric expression
making reference to several different aspects of literature all of
which are clearly expressible in more conventional English.

In Eastman's work we found an argument that causing tension (where
'tension' is roughly synonymous with 'frustration') may be aesthetically
valuable in terms of heightening the reader's consciousness. The psy-
chological theory that a blockage of action tends to heighten conscious-
ness may be correct. But we found that the mere heightening of
consciousness is not sufficient to constitute aesthetic value. We
noticed a more plausible explanation, that the satisfaction one got from
solving the blockage of action may account, in part, for people's
interest in poetry. It would, if this were the only value of metaphor,

put metaphor on a plane with solving puzzles of other sorts. But we

'I-Alfi' I
n

07"1 .
-

 

t"!-

(I)

\ n
l A
u

o...

,.
11:1

4O

recognize that the jigsaw and crossword are not poems and that a good
deal more is required of a good poem than the presentation of puzzles.
Generally our conclusions with regard to the tension theories are:

(a) The introduction of notions such as life, tension
without stipulative definition are not helpful in understanding
metaphor, even in its poetic use (although the mention of energy,
tension and life may have greater aesthetic value).

(b) The claim that in metaphoric use terms 1055 their con-
ventional.meahings Cannot-be defended, since it is only with an '
awareness of the conventional meanings (perhaps among other
things) that one is able to understand metaphor.

(c) The presentation of a riddle to be solved may be one
aspect of the aesthetic value to metaphor.

(d) Several important features indicated by these views which
warrant further attention in the consideration of other views are

(l) Metaphors are open to an indefinite variety of
interpretation.

(2) The various interpretations need not be consistent.

(3) Metaphors involve some conflict between the meanings of
their terms under conventional interpretation.

(4) Metaphor involves comparison.
While the tension theories bring out points which must be accounted for
by a theory of metaphor they fail to provide a satisfactory explanation

of metaphor.

41

NOTES TO CHAPTER II

 

 

 

1. Monroe C. Beardsley, Aesthetics (New York: Harcourt, Brace & World,
Inc., 1958), pp. 135-136.

2. Martin Foss, Symbol and Metaphor in Human Experience (Princeton:
Princeton University Press, 1949).

3. Ibid., 61.

4. Ibid.

5. Philip Wheelwright, "Symbol, Metaphor, and Myth,“ The Sewanee Review,

11.

12.

13.
14.

15.
16.
17.
18.

0‘0me

 

LVIII (1950), p. 687.

Foss, p. 60.

. Philip Wheelwright, Burning Fountain (Bloomington: Indiana University

Press, 1968), pp. 107-108.

Even Henle, who claims all metaphor to involve icons would agree
with us since he uses icon and iconic in an unusual way.

Philip Wheelwright, Metaphor and Reality (Bloomington: Indiana
University Press, 1968).

Wheelwright, BurningyFountain.

We will see that according to his new 'metaphor' that metaphor is a
matter of pragmatics, as opposed to a matter of grammar.

Wheelwright, Metaphor, p. 71.
Ibid.
Wheelwright, Burningyfountain, p. 102.

We say 'perhaps' since his argument is based on 'Love is a red rose'
rather than the more similar line '0, my love is a red, red rose.I
Whatever 'tensive life' refers to, it seems that the comparison
ought to have been made with the more similar line; our proposal
may have more tensive life than Burn's line.

19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.

30.
31.
32.
33.
34.
35.
36.
37.
38.

42

ijg,, 104.

Wheelwright, Metaphor, pp. 45—46.
ijp,, 71-72.

Ibid., 72.

 

bi

0—1
G.

, 72.

bi

0—!
O.

s 72-73.

 

bi

H
G.

, 74.

bi

0—!
O...

 

Wheelwright, Burning Fountain, pp. 121-122.
Ibid. ’ 109.

In considering the graded examples we are somewhat at a loss. We
were to consider them and to notice that those examples which
had a mixture of plurisignation and simile were also those
exhibiting tension. Our agreement that this were the case
would essentially constitute an agreement with Wheelwright's
hypothesis. But we are still unaware of what tension is and
are forced to use his examples as examples of tension in an
attempt to understand the term. Nothing in his examples
(Wheelwright, Burning Fountain, pp. 109-118) brought to our
attention anything resembling the "ceaseless but varying
struggle between opposite forces ..." mentioned earlier
(Wheelwright, Metaphor, p. 45).

Ibid , 109.

Ib'd., 112-113.

1
Ibid , 113.

O
["1

. D.

Wheelwright, Burning Fountain, pp. 1134114.
1919,, 114.

Ipig,, 120.

This is treated in chapter II.

Wheelwright, Metaphor, p. 71.

 

 

 

‘n

'-
‘1

'Iv

'I
‘1
'II

..‘

a
..‘1
(T)

39.
40.
41.
42.
43.
44.
45.

46.
47.
48.
49.
50.
51.
52.
53.
54.
55.

43

Wheelwright, Burning Fountain, p. 102.
Ibid., 79.
bi

H

O.

. 89-90.

H

bi

O.

, 90.
, 91.
bid., 119.

bi

0—.
O.

 

o—i

 

Friedrich Max MUller, "Metaphor," Lectures on the Science of
Language, seventh edition (London, 1873), vol. I, Lecture VIII.

 

 

 

Wheelwright, BurninggFountain, p. 119.
Ipig,, 120.

ijg,, 81.

l!§fl1., 117.

Ibid.

Wheelwright, Metaphor, p. 78.

Ibid., 78-79.
Ibid., 85.
Ibid., 86.

We wish here to forestall an argument based on the obvious growth
of language. Obviously there were once fewer words than there
are now and obviously too few to represent the complex percep-
tions of today's man. The fact that language has grown along
with man's increased number and type of perceptions does not
show that metaphor is responsible for that growth. The growth
of language has included a significant increase in the number
of words. Metaphor does not, however, create any new words.
The bare fact of the growth of language will not establish that
metaphor creates new meanings. A related point is that our
interest is with current language. References to past periods
when language was not sufficient to express certain perceptions
will not establish that current languages in literal use cannot
express every possible meaning. (Even though such expressions
may be so cumbersome as to warrant the definition of new terms.)

44

. Max Eastman, The Literary Mind (Charles Scribner's Sons, 1931).

We are not overlooking the fact that Eastman's work proceeded
Wheelwright's. It is strange that Wheelwright did not make use
of Eastman's arguments in support of his own position.

. Ibid., 170.

 

. Ibid., 184.

 

 

 

Ibid

Ibid , 183.
_I_b_ig.,187-188.
1919., 188.
.1311-

. Wheelwright, Burning Fountain, p. 122.

CHAPTER III

COMPARISON THEORIES

When one asks his grammer teacher or consults a standard book of
grammar, he is apt to find that metaphor is spoken of as implied com-
parison or comparison without the use of 'like' or 'as.' Such infor-
mation may be followed by an example, as follows:

(1) 'Richard is a lion' metaphor (implied comparison).

(2) 'Richard is like a lion' simile (actual comparison).

And some mention will be made that the objects to be compared must be,
to some significant degree, different, or at least it must be unusual
to compare such objects. Now, without much familiarity with metaphor,
this all seems safe enough. But it seems safe because we have con-
sidered only one sort of example. Between the sentences in the above
example there is essentially only a grammatical difference. (1) could
obviously be converted to (2) by inserting 'like' with no change in
meaning. But, consider an example of a different sort.

(3) 'The lion of England leaped upon the infidels.‘

It is not so easy to explain this example as a comparison. There is no
place to insert 'like' or 'as.' An argument could still be presented
that Richard is being compared to a lion. A suggested literal transla-
tion might be:

(4) 'Richard attacked the infidels as if he were a lion leaping
upon its prey.’

45

46

But (3), in itself, provides no basis for assuming that Richard is the
object to be compared to the lion. Richard is not mentioned. In (1)
there is obviously in part at least a comparison; both items to be
compared are mentioned and that mention is through the use of literal
speech. But in (3), the explanation in terms of comparison must be
more complex since only one of the compared items is mentioned. In
order to maintain the comparison view, one must now explain our ability
to recognize the other element of the comparison. A relatively simple
answer to this challenge is that 'lion of England' refers to Richard,
while at the same time containing the word 'lion' which refers to lion
(or perhaps a lion). Realizing this dual reference, we compare the two
objects. But this essentially pushes the problem back one step. Now we

1 Such reference

must explain how 'Lion of England' refers to Richard.
is often explained in terms of context. Were we to come upon (3) in
isolation, we might very well find not a comparison, but rather a pecu—
liar sentence. (Perhaps, in this case, peculiar only inasmuch as there
are no English lions. If the phrase contained 'lion of Africa,‘ we
would fully expect that a group of non-believers were about to be a meal
for a member of the species Eplj§_lgp,) But we do not find metaphor in
isolation. The context, whether literary, physical, or other, may indi-
cate the subject of the metaphor, or the metaphorical reference. The
'lion of England' will refer to Richard III, because he has previously
been our subject. (Or perhaps, since lions are the Kings of beasts and

people have been our subject, we know that it is a king who is the meta-

phorical subject.) We recognize 'lion"s conventional extension.

.l l"
U
'l 'U'
1
I
I. O
1 ' 1'

 

.
O
C :r;r

'1
'09.. .

.-
... 05/ ‘

1.3, r:
.1. 1.

 

'5’.

he.
- .rr-i
‘\I‘

A r; .
' ‘N:
X ;-
I. l){
"r .
I: E

O

I.

F' ‘1
"I

   

t.
d 5 .
V. .
‘I
J‘r.
'- r.
7v :
'l
O. o .
.. _.;_
.‘HI
.
‘n
.’
b q.
.'u I

47

The comparison is between the referent of the term employed metaphoric-
ally (the subject as determined by context) and the referent as deter-
mined by conventional usage.

But this explanation in terms of context fails to hold up in many
cases. While it is true that all metaphor will be in some context, it
is not always the case that context alone will determine a subject (or
referent of the metaphor). We may for example have found (3) in a
context which would determine only that the subject is human. One how-
ever recognizes that the human king is subject since the lion is king
of beasts. By recognizing the relationship involved one is able to
determine the subject. It is by comparing 'person' with 'lion' that we
recognize that the king is being spoken of. Notice here that the process
is reversed. The comparison is the basis for determining the referent.
It is only in understanding the comparison that we know what is compared,
where in the previous example the recognition of two objects invited
comparison. In the more recent example, it is recognizing the royalty
of the beast and of the particular man that we know it is this man rather
than that man to which 'lion' refers. Before we recognize this similar-
ity we are unable to determine the referent of the metaphorical use of
'lion.‘ By trying out comparisons between the referent of the term in
its literal use and various possibilities suggested by context, we
determine the referent of the metaphoric use. Here knowledge of what
is to be compared does not precede the comparison but is rather con-
temporaneous with the comparison. Understanding such a metaphor cannot
be explained as a matter of comparing two objects since recognition of

one of the objects to be compared depends on recognizing the comparison.

48

In considering the examples above we made use of two different
types of explanation both of which involved the notion of comparison.

In the first, the metaphor was construed to be a comparison. Under-
standing this metaphor amounted to understanding the comparison made,
i.e., understanding the relationship between objects compared. The
second example was not construed as stating a comparison. Rather it
was explained that to understand the metaphor, in particular to become
aware of its referent, one must make (as opposed to state) a comparison.
It is through the ground of a comparison that one determines the
referent. As one or the other of these functions of the metaphor is
emphasized one obtains a different theory of metaphor.

To simplify our discussion of comparison theories, we will intro—
duce the following notation: 'A' and 'B' will name the two sides of
the comparison. 'R' will name the relationship, similarity, or ground
of the comparison. Given this terminology one might conventionally
express a comparison as 'A is like 8 in respect of R' or 'A is related
to B by R.‘ We can distinguish two types of theories using this termin-
ology and considering the above examples. The first type describes

2 to the audience, of both A and B. In

metaphor as the presentation
considering both of these the audience determines one or more appropriate
R's. In making this determination an audience has interpreted and
understood the metaphor (though perhaps not the utterer of the metaphor).
We shall call this the comparison view. The second type of view describes
metaphor as the presentation of A and R. In considering these the

audience determines B. An object or event is mentioned by the expression

interpreted literally and by some means the audience is made aware of the

49

ground of comparison. By knowing the ground of the comparison and the
object mentioned under the literal interpretation, the audience is able
to determine the referent of the metaphor. We call this the 'relational
view.‘

The primary difference between these views is what is to be ex-
plained. 0n the first view the theorist is generally most anxious to
explain our comprehension of the new and various relationships which
metaphor is capable of expressing. On the relational view, theorists
are more interested in determining how we know what is being talked
about, that is, how we understand that to which the metaphorical use of
an expression refers.

We consider three examples of the comparison view in this chapter:
Richards, Reiser, and Black. In presenting Richards' we will include
treatment of his various discussions of important aspects of metaphor
and sufficient description to characterize his position. We shall not
cover every aspect of his work. Reiser will be considered as an inter-
esting alternative interpretation of the comparison view. But we do not
consider his work to be a major contribution. Black will be considered
in somewhat greater detail. One basis for this detailed consideration
is a wide reputation for having "taken care" of the subject. Our inter-
est is to determine whether his arguments are sound, whether he has in
fact successfully refuted the viable alternatives to his position and
whether his position is sound.

In the chapter after this one we consider the relational theory of

metaphor.

50

I. A. Richards

The Philosophy of Rhetoric contains Richards' most extensive dis-
cussion of metaphor. But it could be argued that he presents no theory
of metaphor. In reading the Philosophy of Rhetoric one encounters
several statements which could be construed as definitions. These all
seem to be more appropriately treated as statements about metaphor.
However, in considering the points he makes, it is clear that he holds
a version of the comparison view. His interest in this discussion is
directed toward explaining the R (similarity or relationship) involved
in metaphor. One senses from his discussion, though it is not stated,
that the B of the comparison is obvious from the context and that the A
is mentioned by the metaphoric expression in its literal use. That
Richards fails to explain our knowledge of B would be a serious criticism,
if Richards had carefully laid out a theory. But as we.mentioned above,
there was no such laying out and we shall surmise that Richards would
provide an argument that context would provide for our knowledge of B.

The central thrust of Richards' discussion is that metaphor is not
some sort of mysterious phenomenon. He claims that previous views of
metaphor have stifled creativity and he wants to show metaphor as part
of our normal activity. He argues that the metaphors of language are a
subset of a much larger set of metaphors, namely the set consisting of
our thoughts. He claims that thought itself is metaphoric and that the
metaphors of language are merely reports of metaphoric thought. His
discussion dwells at some length on this point. All reasoning is

described essentially as comparing. Throughout most of Philosophy of

 

Rhetoric little attention is paid to linguistic expression. In the last

51

section he comes around to discussing metaphors as linguistic expres-
sions. Since metaphor as it occurs in language is our prime interest,
our discussion centers around remarks made there. We shall however
attend to various other significant considerations which he presents
earlier on in his work. The specific claims, in addition to his
general position which we will examine are:

(1) That metaphor constitutes a compounding of uses.

(2) All language outside science is metaphoric.

(3) That a distinction between tenor and vehicle is helpful in
understanding metaphor.

We shall reject all three claims. This rejection is significant since
if these claims were true they would play a central role in explaining
metaphor.

The first of Richards claims is found in his explanation of his
use of 'metaphor.‘ He uses that word

. to cover all cases where a word, in Johnson's phrase, 'gives

us two ideas for one,‘ where we compound different uses of the

word into one, and speak of one thing as though it were another.

And I took it further still to include, as metaphoric, those

processes in which we perceive or think of or feel about one

thing in terms of another.... (Rhetoric 116)
We are concerned with so much of his commentary as applies to linguistic
use. Here Richards talks about uses, and in particular compounding uses
of a word. He suggests that a metaphor is the result of some union or
mixture of uses.- But metaphor generally involves only two uses: a
single metaphoric use and a single literal use. The metaphoric use is

derived from the literal use. There are not in general two or more uses

which are compounded to produce a metaphoric use. Richards' analysis is

52

untenable. Obviously, he claims two things are being referred to by
the metaphor: that which the phrase normally refers to, and that to
which, under a metaphoric interpretation it refers. This is clear from

his examples.4

But a metaphor cannot be the result of the compound of
two such uses since its use to refer to one of the objects is the meta-
phoric use itself.5 In order for metaphor to result from a compounding
of uses, those uses must be independent of the metaphor.

To speak of metaphor as the compounding of uses suggests that the
two uses of the expression were prior to its use as a metaphor. But
the metaphoric use of the expression is one of those uses and cannot be
prior to itself. In terms of reference, the metaphoric use provides at
most a second referent for the expression. There is no third referent
which results from a compounding of uses. Although Richards describes
metaphor as having ... 'two thoughts of different things active to-
gether'6 the thoughts of the different things cannot explain our under«
standing of metaphor since we understand one of the "things" to which
the metaphor refers only after we understand the metaphoric use of the
expression. Talk of compound uses seems only to mislead. This error
is reflected in Richards' explanation of the meaning of metaphors. He
says:

A modern theory would object, first, that in many of the most

important uses of metaphor, the co—presence of the vehicle and

tenor results in a meaning (to be clearly distinguished from the
tenor) which is not attainable without their interaction. That
the vehicle is not normally a mere embellishment of a tenor which
is otherwise unchanged by it but that vehicle and tenor in co—

operation give a meaning of more varied powers than can be P

ascribed to either.

In our discussion of 'vehicle' and 'tenor' we claim these words to be

53

used ambiguously. Sometimes they appear to refer to meanings, at others
to referents. If we consider 'tenor' and 'vehicle' first to refer to
meanings, the resulting problem is precisely similar to the one above.
The two meanings (tenor and vehicle) "give a meaning of more varied
powers than can be ascribed to either." But such an interpretation en-
tails three distinct meanings; the tenor and vehicle which cooperate
and the meaning given by that cooperation. It is clear that a literal
interpretation of the metaphoric expression will result in a meaning
distinct from the meaning resulting from a metaphoric interpretation.
But there is no third meaning which results. We will claim that the
second meaning is in part a result of the first and that they are there-
fore related. But that relationship does not produce a third meaning.

Treating tenor and Vehicle as referents fares no better since
the referents of the terms are dependent upon their meanings. To sup-
pose the referents cooperated to produce a new meaning is to suppose
the pre-existence of two meanings which result in those referents.

In general, talk of interaction between x and y to produce some
2 requires the antecedent existence of x and y. If the meaning of a
metaphor is some 2 which results from the interactions of two meanings
x and y, then three and not two meanings need to be accounted for.
They are the meaning of the expression under a literal interpretation,
the meaning of that expression as metaphorically interpreted, and the
third elusive meaning whose cooperation is required in producing the
meaning under a metaphoric interpretation. No account is given by

Richards of any such third meaning.

54

The second claim which we wish to consider is that "... most sen-
tences in free and fluid discourse turn out to be metaphoric. Literal

8

language is rare outside the central parts of the sciences." "The

view that metaphor is omnipresent in speech can be recommended theo-

9 This theoretical claim can be stated: the nature of

retically."
language is such that except for the central areas of science a non-
metaphoric language is not possible. We object to this claim on the
basis that Richards either contradicts facts or is using 'metaphor' in

a way other than that in which it is ordinarily understood. The fact
which we say Richards contradicts, is that we do make distinctions be-
tween what is called the literal use of language and the metaphoric use
of language when the nature of the speech is not at all scientific.
Metaphor is seen as a great boon to the poet and often as a hindrance to
the philosopher. We attempt to understand this distinction which is
made so as to be able to encourage or discourage these uses of language
as appropriate.

We find ourselves asking each other on occasion whether a sentence
is being used literally or metaphorically. We ask each other on occasion
to speak literally rather than metaphorically and are sometimes satisfied
that this has been done. A theory which holds that there can be no
literal speech except in science merely deprives us of the use, in the
normal (or ordinary) sense, of that word. If we adopted Richards' posi-
tion we would soon find ourselves making the same sorts of distinctions
while employing a different terminology. This would likely be done in

much the same way as Quine agrees to stop using the word 'exist' and uses

55

'is' in its place.10 Richards is not, as one might expect, arguing

Mfiller's position, that most of the terms Of our language originated
as metaphors. (We presented.this on page 29.) Rather, Richards has
noticed that the meanings of terms fluctuate with use and context.
While we agree with this observation, we still maintain our position
that there is a distinction to be made between literal and metaphoric
use throughout the normal range of language.

The third point we discuss is the distinction between tenor and
vehicle. This distinction is often referred to and employed in the work
of others and is generally positively commented upon. By introducing
the terms 'tenor' and 'vehicle' Richards proposes to clear up certain
confusions in regard to understanding the nature of metaphor.
Specifically, he proposes to distinguish metaphor from other use of
language based on this distinction.

Whether, therefore, a word is being used literally or meta-
phorically is not always, or indeed as a rule, an easy matter to
settle. We may provisionally settle it by deciding whether, in the
given instance the word gives us two ideas or one; whether, in the
terms I suggested last time, it presents both a tenor and a vehicle
which cooperate in an inclusive meaning. If we cannot distinguish
tenor from vehicle then we may provisionally take the word to be
literal; if we can distinguish at least two cooperating uses, then
we have metaphor.H

Ribhards speaks as though distinguishing the tenor from the vehicle were
a fairly straightforward procedure (or at least that in many clear cases
this can be done). But Richards is not clear in his explanation of the
distinction between tenor and vehicle. He says:

A first step is to introduce two technical terms to assist us
in distinguishing from one another what Dr. Johnson called the two
ideas that any metaphor, at its simplest, gives us. Let me call

them tenor and vehicle. One of the oddest of the many odd things
about the whole topic is that we have not agreed upon distinguishing

56

terms for these two halves of a metaphor -- in spite of the immense

convenience, almost the necessit , of such terms if we are to make

any analysis without confusion.
An even odder point here is that Richards provides two terms to make a
distinction without explaining what is being distinguished. He states
that at present we have only clumsy phrases such as 'the original idea'
and 'the borrowed one,‘ 'what is really being said or thought of' and
'what it is compared to,‘ etc. But these clumsy expressions are as
close as Richards comes to defining his 'technical' terms. His term
'vehicle' appears to refer to three distinct facets of the metaphor.
They are: the term used, the referent which such a term would have if
employed literally, and the sense which that term would have if employed
literally. By tenor, he refers to the sense and to the reference of the
term used metaphorically. Richards' explanation on page 96 in terms of
the original idea and the borrowed one appear to be using tenor and
vehicle as synonymous with the sense of the term used metaphorically and
literally respectively. On page 118 he speaks of tenor and vehicle as
though they were the referents. ''A very broad division can be made be-
tween metaphors that work through some direct resemblance between the
two things, the tenor and the vehicle and those which work through some
common attitude which we take toward both." It appears that Richards
simply overlooks the distinction between sense and reference.

But even if Richards had consistently used these terms to refer to
the sense or the referent of the expressions used literally or metaphor-
ically, these terms would be of little value in determining whether an

expression is metaphoric. The adoption of these terms for that purpose

57

will instead expose a fault of the intended distinction. If we rephrase
Richards' statement, substituting definitions for tenor and vehicle, we
have the following: 'it presents both a sense when used metaphorically
and a sense when used literally which cooperate in an inclusive meaning.‘
The fault exposed is circularity. Surely a sentence with a sense when
used metaphorically is a metaphor. The vehicle-tenor distinction is of
no help in recognizing metaphors. The problem with this distinction is
that one must know that he is dealing with metaphor before he can make
that distinction. That is, he must recognize the metaphoric sense in
order to know that there is a tenor. Notice that it would be of no help
to drop the words 'when used metaphorically' in order to avoid circu-
larity, since to say only that the sentence had two senses or two refer-
ents would fail to distinguish metaphor from ambiguity. The reference
to cooperation in an inclusive meaning may limit the sorts of uses of
ambiguity but seems to describe the pun rather than metaphor.

The distinction which Richards desires to make, while not success-
ful, does indicate his adoption of a comparison theory. At some points
he speaks as though we were comparing objects with others as though the
similarity between the objects accounted for our understanding of the
metaphor. This latter view is expressed in his discussion of the example
'leg of a table.’

Let me begin now with the simplest most familiar case of verbal

metaphor--the leggof a table for example. We call it dead but

it comes to life very readily. Now how does it differ from the

plain or literal use of the word, in the lggof a horse, say?

The obv1ous difference is that the leg of t e table has only some

of the characteristics of the leg of a horse. A table does not

walk with its legs: they only hold it up and so on. In such a
case we call the common characteriStics the ground of the

 

58

metaphor. Here we can easily find the ground, but very often we
cannot. A metaphor may work admirably without our being able

with any cgnfidence to say how it works or what is the ground of
the shift. 3

The common characteristics form the 'ground of the metaphor' and we see
Richards' commitment to this ground as an aspect of every metaphor. He
is not however committed to saying that the ground is always composed

of common characteristics of the objects.

A very broad division can be made between metaphors that work
through some direct resemblance between the two things, the tenor
and vehicle, and those which work through some common attitude
which we may (often through accidental and extraneous reasons)
take up towards them both. The division is not final or irreduci-
ble, of course. That we like them both is in one sense, a common
property that the two things share, though we may, at the same
time, be willing to admit that they are utterly different.14

On Richards' view, we may be able to recognize that the two objects are

of one sort but be unable to specify any particular characteristics which

15'

they share. While speaking of the ground of the shift which tends to

indicate that the similarity was the basis of our understanding, it is
the making of the comparison which remains for Richards the essential

function of the linguistic metaphor.
Richards proposes three ways of interpreting the word "comparison'

and notes that "As we mean by comparison these different things we get

..15

different conceptions of metaphor. While he does not indicate one

interpretation as resulting in a correct theory of metaphor, he does
reject two of the three interpretations. He says of a comparison

... it may be just a putting together of two things to let them
work together; it may be a study of them both to see how they
are alike and how unlike one another; or it may be a process of
calling attention to their likenesses or a method of drawing
attention.to certain aspects of the one through the co-presence
of the other.17

59

If comparison is interpreted as the 'calling attention to their like-
ness ..." then the explanation of metaphor as comparison fails,
according to Richards.
The more carefully and attentively we go over the senses and impli-
cations ... the less shall we find the resemblances between
vehicle and tenor counting and the more will the vehicle ... come
to seem an excuse for saying about [th? tenor] something which
could not be said about the [vehicle]. 8
Richards says that we should not "... suppose that the interactions of

."19 He observes

tenor and vehicle are confined to their resemblances...
that generally the differences between tenor and vehicle exceed their
similarities and it is through these differences that the vehicle brings
about the particular modification of the tenor.

The other form of the comparison theOry to which Richards objects
is one which he claims characterizes Eastman. It is the form which
Richards describes as 'just a putting together of two things to let them
work together.‘ Richards takes this sort of view to be an overreaction
to the comparison view just criticized. While applying the label 'crude
conception' to this view, Richards provides arguments which tend to
support it. Richards' discussion is sounder than Eastman's; but given
Richards' argument the view becomes less a tension theory and more akin
to Richards' comparison view.

Richards argues that when two quite different things are put to-
gether ('an impractical identification' in Eastman's terminology) the
most important happenings besides a general confusion are the mind's
efforts to connect those quite different things. Richards says

The mind is a connecting organ, it works only by connecting and

it can connect any two things in an indefinitely large number of
ways. Which of these ways it chooses is settled by reference to

60

some larger whole or aim, though we may not discover its.aim, the
mind is never aimless. In all interpretation, we are filling in
connections and for poetry, of course, our freedom to fill in--the
absence of explicitly stated intermediate steps--is a main source
of its powers.
Given this view of the mind's operation, our understanding of metaphor
is in part explained. We attempt to make some sense of any linguistic
expression and attempt to find connections in those expressions offered
as comparisons. While this explanation puts Eastman's view in a better
light, Richards rejects the appeal to tension alone as an adequate
explanation of metaphor.
As the two things put together are more remote, the tension
created is, of course, greater. That tension is the spring of
the bow, the source of the energy of the shot, but we ought not
to mistake the strength of the bow for the excellence of the
shooting; or the strain for the aim.21
Richards continues his argument against Eastman by suggesting that in
emphasizing tension Eastman may be overstating a single aspect of meta-
phor. The identifications Eastman claims as impractical, may when
removed from context, be impractical, but given the proper context the
impracticability diminishes. In argument Richards offers as an example
the line from The Drummer Boy's Communion by Gerard Manley Hopkins.
The line is offered in answer to a challenge that 'house' and 'bread'
can hardly ever have the same reference (this we would construe as one
of Eastman's impractical identifications).
Low-latched in leaf-light housel his too high godhead.
Richards claims that recognizing from the context that "... Hopkins is
speaking of the wafer as the dwelling of the Divine Presence"22 one

finds no strain in speaking of bread as a little house. Richards says:

61

But it is the rest of the poem that makes the connection easy

and obvious, which witnesses to a general truth. The mind will

always try to find connections and will be guided in its search

by the rest of the utterance and its occasion.
Argument by example, unless constituting a counter-example is weak.
Nonetheless, the general explanation presented by Richards is more plausi-
ble than Eastman's discussion.24 Rather than being an impractical
identification which interferes with the reader's progress, the metaphor
is the use of an expression to be interpreted in context so that the
audience will draw out the various comparisons. Of truly impractical

identifications, Richards says:

And bafflement is an experience of which we soon tire, and
rightly so.25

While one might, as Richards' characterization suggests, just put
two things together to see how they work, the value of such a metaphor
would lie not in any confusion which it would cause but rather in its
potential comparisons. In the case of successful metaphor the author
provides by context a setting in which the comparisons can be made.

Richards' discussion of the Eastman position supports our conten-
tion that he would argue that in the case of metaphors in which only one
of the compared objects is mentioned by the metaphoric expression, the
other will be indicated by context. It is by context that one knows
that the 'house' of his example refers to the host at communion. In
arguing against construing metaphor as impractical identification he
argues that the ground or similarity may (and is in some cases) be
provided by the context. The result is that a metaphor is a comparison
in which one side of the comparison is mentioned, the other side and the

basis of the comparison is given by context.

62

Richards fails to mention the third interpretation and since he
rejected the other two, we suspect that this best characterizes his posi-
tion. The view would be characterized by interpreting comparison as
'a study of them both to see how they are alike and how unlike.‘ His
rejections of the other characterizations of comparison were alterna-
tively based on noting that the first failed to take account of differ—
ences and that the second failed to recognize the part played by
similarity. A study to see how things were alike and different would
avoid both of these sorts of criticism. Our confidence that this inter-
pretation of comparison is the basis of Richards' position also is sup-
ported by other comments which have induced us to classify his view as
a comparison view. In particular, the view that the two objects (A and
B) are presented and the audience is to determine their relationship
(R) suggest that one is equally interested in similarities and differ-
ences.

Our major objection to Richards' theory of metaphor was mentioned
in our initial paragraphs concerning Richards. It is that Richards has
failed to carefully lay out a theory of metaphor. As a result, his
discussion has several deficiencies. The first is that Richards explains
metaphor as being a matter of comparison, yet does not explain how one
gains his knowledge of what is to be compared. As we suggested,
Richards would probably argue that this information would be provided
by context. We should like to have seen such an argument sufficient to
deal with metaphor in general, although it is clear that such an argument

could be constructed to cover special cases. Second, Richards does not

63

present a sufficient characterization of metaphor. Although he charac-
terizes metaphor as comparison, he provides no basis for distinguishing
metaphor from other comparisons. There are many comparisons which are
not metaphors. The statements he offers, which would tend to further
define metaphor, are either circular, in the case of the tenor-vehicle
distinction, or more accurately describe the pun, as in the compound of
uses discussion. Third, in failing to precisely characterize metaphor,
Richards also failed to explain our ability to recognize and appro-
priately interpret metaphor. While he mentioned a power of the mind to
make connections, he failed to indicate a means of recognizing cases of
metaphor and therefore cases where such connecting would be appropriate.

Our second major objection is one which will apply generally to
comparison views. These views either assume or attempt to establish by
argument that all metaphors make reference to things to be compared.
Many metaphors however make no such reference. Black provides us with
such an example, though he didn't propose it as one:

The chairman ploughed through the discussion.
One may indeed have to think about plowing in order to determine the
chairman's actions at the meeting, but there is no interest in comparing
these events. Such a statement conveys the manner in which the meeting
was conducted; two things are not compared. One could produce a sentence
of equivalent meaning in which no mention is made of plowing. Such a
sentence is:

The chairman proceeded directly through the agenda, dealing
summarily with objections, ruthlessly suppressing irrelevance and

generally refusing to allow the giscussion to stray from the
central issue or be interrupted. 5

64

That this sentence says essentially the same thing as the metaphor is
evidence that the intent of the original statement was not to compare
the conduct of a discussion with plowing a field, but rather to specify
how the discussion was conducted. Another example is provided by the
following line from James Whitcomb Riley's l'Out to Old Aunt Mary's."

While the waters gurgled and laughed and wept.
While such a line describes the sounds made by the waters, perhaps even
how they appeared, no comparison is stated. No objects are mentioned
in the line or the rest of the poem which would be compared. In fact
in making a comparison between sounds one loses entirely the effect of
the waters being alternatively happy and sad.

Our point is not that there are no metaphors which are intended
as comparisons and which do compare objects. It is a claim that
Richards' discussion of metaphor as comparison will not serve as a
general account of metaphor.

The specific claims which we disCussed and rejected were that
(1) all language outside of science is metaphoric, (2) metaphor is a
compound of uses, and (3) the tenor-vehicle distinction is valuable in
understanding metaphor. Our objection to the tenor-vehicle distinction
is not that more careful and precise terminology is needed but rather
that Richards' terms were no better characterized than those which they
were to replace. We argued that a compound of uses explanation presumes
multiple uses prior to the metaphoric use: the metaphor however may be
only a second use and that is generally derivative from a single other

use. We rejected the notion that all language outside of science is in

65

theory metaphor since such a statement is of no help in understanding
the distinction we do make. Such a position would only prompt us to
seek a new word for our present use of 'metaphor.‘

Richards' positive contributions lie mainly in his criticism of
other views. He criticized two views of comparison. The first holding
that the comparison involved in metaphor is a matter of noting similari-
ties between objects. This view, which he claims to be the basis of
18th Century criticism, overlooked the fact that it is often the differ-
ences, more than the similarities, which account for the meaning of a
particular metaphor. In criticizing Eastman, Richards demonstrated that
understanding metaphor often involves attention to context. He showed
that what might appear out of context as an impractical identification,
may within context be completely sensible and provide no difficulty in
interpretation.

A further contribution of Richards is the indication that talk
about metaphor must employ a more precise terminology. In examining his
discussion of 'tenor' and 'vehicle' we noticed that not only must one
distinguish between literal and metaphoric interpretation, he must dis-
tinguish between a word, its referent, and its meaning as indicated by

those interpretations.

Max Reiser

 

Max Reiser proposes an alternative basis for the ground of compari-

son.27 While it could be questioned whether his view ought to be

considered an example of the comparison view, it provides an interesting

66

contrast to Richards'. While Reiser emphasizes similarity, on Reiser's
view, it is a similarity of emotive response to objects rather than a

similarity of those objects themselves. We notice that Richards allows
our feelings towards objects to be common properties.28

Reiser's position could therefore be seen as contained within
Richards' position. But Richards says little concerning the emotive
aspect of metaphor. Reiser, on the other hand, stresses this aspect of
metaphor in regard to poetry. Reiser is clear that he intends his
discussion to be confined to the poetic use of metaphor.

In quotations we will find Reiser using 'simile' rather than
'metaphor.’ Reiser considers the simile to be the basic figure of
speech, with metaphors being "... transformations of the simile, deriva-
tions or abbreviations stemming from a common type of analogy most

29 This claimed relationship between

clearly evolved in the simile."
simile and metaphor is often disputed. But even if Reiser is not cor-
rect in this respect, we need not let this interfere with our considera—
tion of his theory, since we may assume that his remarks were addressed
toward metaphor directly.

Reiser's theory is that metaphor is not based upon any similarity
between the literal and metaphorical referent but rather upon the emo-
tions aroused by those objects. He speaks in terms of a distinction be-
tween scientific and poetic thinking. But this distinction in the end
comes to the difference between comparing physical objects and comparing

emotional responses.

The simile does not imply that process of association we are accus-
tomed to in normal scientific thinking. Contrary to the ordinary

 

AIL.

. D.

‘Po

 

O\U
.0

CF-

h

67

association wherein we associate objects according to their real
similarity, we shall see that the similarity of objects in the
poetic simile is not real but fictitious when viewed from a
realistic (scientific) point of view. These objects of the simile
may seem similar if 5ooked upon emotionally but there is no real
resemblance at all.3

The poet compares not the referent of the word or words but rather the

impressions which these referents have left upon him.
Not the objects themselves but the impressions created in his
"feeling" ... are similar, and that is why he connects these two
objects. The link is not factual but emotional. Thus he does not
compare two real objects in the outside world but two impressions
within himself. In other words the gmotional value is similar or
identical not their real appearance.

The ground of the metaphor is emotive. This is a significantly differ-

ent claim than one would be making if he claimed that the value of the

metaphor might lie in the feeling transferred from object to object as

32 Reiser claims

is suggested in Henle's discussion of induced content.
that in poetic metaphor two objects producing a common emotive response
are compared.

Reiser cautions us against a mistake which he would probably claim
is made by both Henle and Richards insofar as their work is construed
to concern poetic metaphor. He says that one can always discover simi-
larities between objects and this is the case in regard to poetic meta-
phor. Such a discovery is apt to lead us to believe that the basis of
these metaphors is a similarity between objects. But Reiser maintains
that such similarities between objects are not sufficient to support a
figure of speech.

In arguing for his position Reiser presents the following example:

the blood of grapes.

68

He claims that the comparison is based on an emotional response
The emotional value of the blood illustrating the emotional value
of the wine creates a simile and appeals to everybody. Without
this emotional value the simile would be "dead," cold, "meaningless"
and this hidden meaning is the meaning of both objects for our
sensibility.33
Reiser argues that there must be an emotional relationship between ob-
jects if one is to become the symbol for the other.

For Reiser the understanding of a metaphor is a matter of intuition.

This intuition means simply that the emotional spheres surrounding

the symbolized and symbolizing objects within the sensibility of

the poet and his public are identical. Hence the "intuitioné‘4

It is a result of emotional conformity towards these objects.

It is clear from the title that Reiser is attempting an analysis of
poetic simile. But a more accurate description might be 'an analysis of
the aesthetic value of certain poetic similes.‘ It is obvious from his
examples that some metaphors have emotion as their base. Often authors
choose the terms of their comparisons in such a way that the emotional
value of one of the objects will be transferred to the other. But in
such cases where emotion does play a significant role, Reiser's explana-
tion is incorrect. According to his explanation the metaphor is under-
stood through a similarity in one's emotive responses to the two objects.
But if there were such an emotional similarity then the comparison would
lose its emotive value, i.e., the comparison would not be a transfer of
emotive response from one object to the other since we already respond
similarly. The emotive force of the figures which Reiser mentions,
namely:

Thy eyes are like the stars

Mother Earth

The blood of grapes

69

lies in their ability to impute a certain emotion where one was once
lacking. The relationship required for understanding the metaphor is
often what Reiser calls a 'scientific' or 'real similarity.‘ Such a
"real similarity“ is necessary for the poet and the audience to consider
one thing in terms of another so that the emotional transfer can take
place. In his example concerning "the blood of grapes" he explains
several similarities between human blood and wine. And he is correct
that "If redness alone were to link these two objects, blood would make
a very poor symbol indeed." The value of the figure may well be its
emotive force. But he is incorrect in believing that there is an ante-
cedent rather than induced emotional similarity between blood and wine.
What the poet has attempted here to do is get us to have a similar emo-
tional response to the wine that we would usually have towards blood by
preying upon their non-emotional similarities.

While Reiser is correct in thinking that emotion plays a part in
metaphor we find his account lacking in two respects. First, even for
those examples he provides, it is the similarity of the objects involved,
or the similarity of our conception of them, that provides the basis for
our understanding of the metaphor. The emotive element is important in
that the metaphor allows us to transfer our feelings from one object to
the other. This has the result of a similarity of feeling towards both.
That similarity is however a result of the metaphor rather than its
basis. Second, his theory at best covers only the poetic use of metaphor

and then only certain types of metaphor.

70

The main value of Reiser's view is the consideration of some basis
other than similarity as the ground of comparison and to suggest another

aspect of metaphor of which we must take account.

Max Black

Black, in his paper ”Metaphor"35 proposes what he calls an 'inter-
action' theory of metaphor. This theory, in which he claims to follow
Richards, seems reasonably well to fit the definition of comparison to
which Richards did not object. That definition was: "A study of them
both to see how they are alike and unlike." We therefore classify his
view as a comparison view. He presents for consideration and criticism
two other views. He calls one the 'substitution' view; the other he
calls the 'comparison view.‘ We shall notice that Black's interpreta-
tion of the comparison theory differs significantly from our description.
In fact we shall claim that Black's theory differs significantly from

theories which anyone else holds.36

His arguments against these views
appear as straw-man arguments while he fails to discuss positions simi-
lar to those which are held.

Black's own view is presented with some confusion, and is in many
ways unsatisfactory. Black, however, makes three points of significance,
which have been many times overlooked. They are (1) that metaphor in-
volves essentially at least a sentence rather than a single word,

(2) that a community's beliefs about things rather than facts govern the
function of metaphor, and (3) that metaphor is a matter of meaning.

Black's initial discussion is based upon the first point, that a

metaphor is a sentence rather than a single word. In support of this

71

point, he begins by distinguishing between that part of the sentence
which is usually thought of as metaphoric and that part thought of as
literal. Black invites consideration of the contrast between
'ploughed' and the rest of the sentence in the example:
'The chairman ploughed through the discussion.‘
This would be commonly expressed by saying that 'ploughed' has
here a metaphorical sense, while the other words have literal
senses.37 Although we point to the whole sentence as an instance
(a "clear case") of metaphor, our attention quickly narrows to a
single word, whose presence is the proximate reason for the attri-
bution.
That part of the sentence which Black says has a metaphorical sense is
called the 'focus' and the remaining part is called the frame. His argu—
ment that the whole sentence must be considered as the metaphor is that
a change in either the focus or the frame will result in different meta-
phor, or perhaps no metaphor. The focus he notes will be considered
metaphorical only within certain frames and that a change in the frame
will result in a change in the metaphor. If the chairman was said to
have ploughed his field there would be no metaphor; if a ship is said
to have ploughed the ocean, then the metaphor is different.
Suppose somebody says, "I like to plough my memories regularly."
Shall we say he is using the same metaphor as in the case already
discussed or not? Our answer will depend on the degree of simi-
larity we are prepared to affirm on comparing the two "frames"
(for we have the same "focus" each time). Differences in the two
frames will produce some differences in the interplay between focus
and frame in the two cases. Whether we regard the differences as
sufficiently striking to warrant calling the sentences two metaphors
is a matter of arbitrary decision.
If Black expects us to take this question seriously, then he either
thinks we have not attended to his previous discussion or has confused

the focus of the metaphor with the metaphor. It is obvious that the

72

two sentences which he presents are different and that they have differ-
ent senses: therefore, that the metaphors differ is obvious. What might
be seriously questioned is whether the focus is being used in the same
way or has the same sense. It would be reasonable to question whether
the relationship between focus and frame in the particular instances
was sufficiently different to claim that the sense of the focus differed.
But given his claim that a metaphor consists of both a focus and a frame,
a change in either will result in a change of metaphor.

So far we have not taken issue with Black's claim that metaphors
are sentences or other expressions. This claim he expresses on page 275.
However, he speaks concurrently of 'metaphorical use' of words which
occur in such sentences. It appears that Black has not settled on
whether he intends sentences (sign types) to be the metaphors or whether
he intends uses of sentences (utterances of sign tokens for specific pur-
poses) to be the metaphors. In discussing his examples it appears that
he treats the sentences as the metaphors. But it is difficult to under-

39 If Black is to

stand how uses of words could occur within sentences.
maintain the position that metaphors are sentences then he will indeed
have to distinguish metaphorical sense from literal sense and face the
accompanying difficulties. By discussing metaphorical uses of sentences
he could avoid such difficulties and preserve a single notion of sense.
But rather than adopt one view and its advantages or disadvantages,
Black wavers. On page 277, he states:
So far I have been treating "metaphor" as a predicate proper-
ly applicable to certain expressions, without attention to any
occasions on which the expressions are used or to the thoughts,

acts, feelings, and intentions of speakers ondsuch occasions. And
this is surely correct for some expressions.

73

But later on that same page he states:

When Churchill in a famous phrase called Mussolini "that
utensil," the tone of voice, the verbal setting, the historical
BacEground helped to make clear what metaphor was being used.

And on 279 he writes: "On this view the meaning of M, in its meta-
phorical occurrence, is just the literal meaning of L." In proceeding
with our discussion we shall insofar as possible overlook this wavering
and suppose the most charitable view in each context.

Black is concerned to indicate that "to call a sentence an instance
of metaphor is to say something about its meaning." While we agree that
the syntax of a sentence does not serve to establish its status as a
metaphor the argument which Black presents is less than satisfactory.
That argument is:

If the sentence about the chairman's behavior is translated word

for word into any foreign language for which this is possible,

we shall, of course, want to say that the translated sentence is

a case of the very same metaphor. So, to call a sentence an

instance of a metaphor is to say something about its meanin , not

about its othography, its phonetic pattern of its grammatical form.

(To use a well-known distinction, "metaphor“ must be classified as

a term belonging to "semantics" and not to "syntax'--or to any

physical inquiry about language.)41
It is by no means clear that a word for word translation of metaphor into
a foreign language results in metaphor. Consider first the question of
translating a metaphor into a foreign language. A word for word trans-
lation will often be regarded not as a metaphor but rather as confusion
or perhaps a statement which is out of place. Consider, for example, 'No
tiene pelos en la lengua.‘42 A word for word translation of which is:
'He has no hair on his tongue.‘ Such a statement is true enough-for

most individuals to whom 'he' would refer, but quite a curious thing to

 

 

“re--

a...

'lpzn

" n31

(7“)

E, of 1
.' ssr [A

74

say. This expression is not recognized as a metaphor in English.
A brief glance into a bilingual dictionary will show such expressions
to be idioms with special directions for translation. When these
directions are not provided the word for word translations of such
phrases are not recognized as metaphors.

Black does not provide any criteria of a correct word-for-word
translation. Several possible criteria come to mind, none of which
sustain his point. They are:

(1) that the result of the translations have the same meaning
as the original

(2) that individual words be replaced by synonyms of the other
language

(3) that individual words be replaced by synonyms and adjustments
be made to preserve grammatical form.

Given the point of Black's argument, it cannot be that the result
of the translation have the same meaning as the original. If this were
the criterion the non—metaphoric idiomatic translations of the bilingual
dictionary would count as correct. In other words the translation would
not be a metaphor but rather a literal expression with the same sense.
Given this criterion, Black's argument fails since the translation will
not be regarded as a metaphor, contrary to his claim.

We cannot believe that Black would envoke the criteria that each
word be replaced by a synonym, since to do so would result in nonsense.
Consider for example the sentence '11 y a un maison blanc ici.’ Word
for word replacement yields 'It there has a house white here.‘ The gram-
mer of language being different, mechanical substitution will not result

in sentences. Even if some criterion of adjustments for grammer is

75

added, Black has no guarantee that he will have the same metaphor and
further, no guarantee that the expression will not be literal. ('He
has no hair on his tongue' is a literal statement.)

A third consideration weighs against Black's claims that 'to call
a sentence an instance of metaphor is to say something about its mean-
ing ...“ and that "'metaphor' must be classified as a term belonging to
'semantics'...." It would seem that this position entails the further
claim that it is in virtue of its meaning that a metaphor is a metaphor.
However, every day, literary scholars discover the meaning of some

metaphors and provide literal equivalents.43

These literal equivalents
have at least on some occasions the same meaning as the original meta-
phor. It may appear that we contradict ourselves talking of literal
equivalents since we deny that every metaphor has a literal equivalent.
But it is important to notice that one can deny the statement "all meta-
phors have literal equivalents" without claiming that no metaphors have
literal equivalents. It is clear from our argument above that having
the meaning that a metaphor has is not in itself a sufficient criterion
of being a'metaphor. Black, himself, agrees that there are some
'substitution metaphors'44 which have literal equivalents which are not
metaphors.

If this is the case, then some literal statements mean the same
as some metaphoric statements. Therefore (since the metaphor and the
literal expression may have the same meaning), the criteria by which we

distinguish metaphor from literal speech must include other considera-

tions besides meaning.

1S 5'.

p»~r

‘ 1 I
a. P‘,
v ..1

‘Q
A! .

this
If:

.1

‘ lll

Id”

‘76

Black, even though he argued for the position that meaning alone
is sufficient to determine that an expression is metaphoric, appears
unconvinced. He states:
We recognize that to call a man a "cesspool" is to use a metaphor
without needing to know who uses the expression or on what occa-
sions, or with what intention. The rules of our language determine
that some expressions must count as metaphors; and a speaker can
no more change this than he can legislate that "cow" shall mean
the same as "sheep." But we must also realize that the established
rules of language leave wide latitude for individual variation,
initiative, and creation.45
One wonders what sort of rules these are to which Black refers. In view
of his earlier discussion they cannot be syntactical rules. He later
calls them 'rules of usage.‘ Perhaps he has in mind some set of semanti-
cal rules. But Black fails to tell us what these rules are. In short,
Black maintains that metaphor is a semantical rather than syntactical
matter and that there are some rules which determine that certain gram—
matical constructions are metaphoric. Although he provides us with some
clear examples of metaphor he does not provide us with the rules for

recognizing them.46

While we disagree with Black's arguments which tend
to show that the having of a particular meaning is the determining factor
in metaphor, we support the position that meaning is of central impor-
tance in explaining metaphor. In our discussion of Henle we criticize
his work for failing to take account of the importance of meaning for

metaphor.

Substitution Views
Black turns his attention to the consideration of three views of

metaphor. The first such view is called by Black the isubstitution view

of metaphor." He defines that View as follows:

Any view which holds that a metaphorical eXpression is used in

place of some equivalent literal expression,.I.shall call a subs

stitution view of metaphor. (I should like this label to cover

also any analysis which views the entire sentence that is the 47

locus of the metaphor as replacing some set of literal sentences.)
In spite of Black's examples of persons to whom Black attributes this
view, it is not clear that anyone does hold it. Consider his first
example.

... Whately defines a metaphor as "a word substituted for another

on accognt of the Resemblance or Analogy between their significa-

tions.“ 3 *
While it is clear that Whately employed the word "substituted“ it is not
clear that he held the substitution view. An important aspect of the
substitution view as described by Black is that the metaphor must replace
some literal equivalent expression. It is certainly not clear from the
passage presented that Whately held that for each metaphorical expres-
sion there is some literal equivalent. The substitution may very well
be for the purpose of creating an expression having no literal equiva-
lent.

Black's other example from the Oxford English Dictionary fares no
better.

"Metaphor: The figure of speech in which a name or descriptive

term is transferred to some object different from, but analogous

to, that which it is properly applicable; an instance of this, a

metaphorical expression.
Here again we find the name of one object being used for another and this
might reasonably be construed as a case of substitution. Nonetheless the

lexicographer is not committed by his definition to the further claim

that there exists a literal equivalent for each metaphor. Thus even if

78

Black is successful in showing the "substitution view of metaphor" to
be incorrect, it is not clear that anyone other than Black has seriously
held such a view. In further describing this view, Black states:
“According to a substitution view, the focus of a metaphor, the
word or expression having a distinctively metaphorical use within
a literal frame, is used to communicate a meaning that might have
been expressed literally. The author substitutes M for L:."5
And in his note on terminology Black describes this view as follows:
For metaphors that fit a substitution or comparison view, the
factors needing to be distinguished are: (i) some word or expres-
sion E, (ii) occurring in some verbal "frame" F, so that (iii) F(E)
is the metaphorical statement in question; (iv) the meaning M] (E)
which E has in F(E), (v) which is the same as the literal meaning,
M(x) of some literal synonym, x. A sufficient technical vocabulary
would be: "metaphorical expression" (for E) "metaphorical state-
ment" (for F(E)), "metaphorical meaning" (for Mf) and “literal
meaning" (for M).
Notice the very strong claim with regard to synonomy. Not only must the
entire metaphorical sentence have a literal equivalent but the focus of
the metaphor must have one as well. According to such a view, the
author could always avoid metaphor by using F(x) rather than F(E). The
only effect of speaking metaphorically seems to be the creation of a
puzzle for the audience.
In explaining why a "writer should set his reader the task of solv-

52

ing a puzzle" we find that Black describes two theories rather than

one. The first reason he suggests is "there may, in fact, be no literal

53 But this

equivalent, L, available in the language in question."
explanation is inconsistent with the substitution view which Black had

been describing. Black gives no indication that he recognizes this con-
tradiction, which would in itself be a sufficient rebuttal of such a view.

So that we may follow Black's discussion, we will divide these views.

79

The view originally stated will be substitution view 1; the view allow-
ing the possibility that the metaphor has no literal equivalent will be
substitution view 2. It is clear, however, that the second view is
either absurd or misnamed. Claiming metaphor to be a matter of substi-
tution when there is nothing for which the metaphoric expression is
substituted is absurd. It would be better for Black to recognize that
these are two distinct views. Since he redefines "catachresis' it would
lead to less confusion to call substitution view 2 'the catachresis
view' and not a substitution view at all. A strong additional reason
for separately naming and separately treating this view is that it is a
view which has been held, in part at least, by several literary theor-

54

ists. We shall therefore adopt the term 'catachresis view' as the

name of the second substitution view.

Black explains that "Metaphor plugs the gaps in the literal vocabu-

"55 He continues:

lary....
So viewed, metaphor is a species of catachresis, which I shall
define as the use of a word in some new sense in order to remedy
a gap in the vocabulary; catachresis is the putting of new senses
into old words. But if a catachresis serves a genuine need it
will quickly become part of the literal sense.

 

Black fails to present any analysis of this view. He raises no objections
against it and from his later discussion of substitution metaphors it
appears that he accepts this position as an explanation of at least some
metaphors. If indeed metaphors do function in this manner they may serve
as a vehicle for expansion of common understanding, of explaining how
communication is not limited by our various languages. But after sug-

gesting this view Black fails to consider arguments in support or

80

opposition and fails to explain how metaphor is able "to remedy a gap
in the vocabulary." Rather than recognizing the importance of this
theory, Black says only "... it is the fate of catachresis to disappear

when it is successful."55

Apparently the rapid change to literal or
conventional use is sufficient ground for dispensing with further con-
sideration of them. His treatment here is grossly incomplete and this
implication of silence is that since metaphors of this sort are short-
lived (as metaphors) they are not worthy of explanation. But it is
just such a theory of metaphor which would account for much of the
expansion and change in our language. If such a theory is correct, it
is eminently worthy of evaluation. Further, since Black presents no
objection to this theory, he presents no reason to prefer his own.
Substitution theory 1 is portrayed as holding that metaphor is a
stylistic device. The purpose according to this theory of using metaphor
is to increase the reader's enjoyment. Black says,
The principle behind these "explanations" seems to be: When in
doubt about some peculiarity of language, attribute its existence
321311113521?11321-231138235111121223?” 11.11 H... “‘3 .1111
Such a scornful answer hardly shows the view to be incorrect. Such a
remark would seem to indicate that a reader's pleasure is irrelevant in
evaluating a figure of speech. But insofar as one is concerned with the
occurrence of metaphor in works of art it is not at all clear that the
reader's pleasure is irrelevant. Santayana, for example, says: ... the
true test [of the real merit of a work of the imagination] is the degree

and kind of satisfaction it can give to him who appreciates it most."

Given this view of literature it would not be unusual to hold that a

81

poet's use of metaphor was designed to please. If indeed such theories
of literature are correct, then establishing that metaphor affects
feelings is a central issue; if it could be demonstrated that metaphor
did not have this effect, it would be a strong argument against such
theories of literature.

One can treat this discussion of Black's as an argument with a sup—
pressed premise. That suppressed premise is that metaphors are cogni—
tively valuable and play a significant role in philosophy. The inclusion

of this statement as a premise results in the following reductio—ad-

 

absurdum of the substitution 1 view:

1. According to the substitution view, the value of metaphor lies
solely in the pleasure it gives the reader. ("the purpose of
metaphor is to entertain and divert")

2. "If philosophers have something more important to do than give
pleasure to their reader, metaphor can have no serious place
in philosophical discussion." A

3. Philosophers have something more important to do. (obvious)

4. Metaphor has a serious place in philosophical discussion.
(suppressed premise)

5. Therefore, the substitution 1 view is not correct.

Adding 4. as a premise results in a correct argument. However 4. or its
negation is a statement we would like to see as the conclusion of an
argument. It is probably the central question for a philosopher to be

concerned with and not something to be assumed.

82

If substitution view 1 is correct, it may be that we ought to
avoid metaphor except in literature since it can mislead us. If the
catachresis view is correct then it is important to develop ability in
the use of metaphor as a means of overcoming the restrictions of
language in formulating new theories. But Black fails to satisfactorily

deal with these questions.

Comparison Views

Black initiates his discussion of the comparison view by discussing
figures of speech as functions. He uses the term function in much the
same way as would a logician or mathematician. Figures are viewed as
mappings of one meaning onto another. He uses irony as an example. Here
the task the author performs is to say the opposite of what is intended.
On this function model Black describes metaphor:

What, then, is the characteristic transforming function involved

in metaphor? To this the answer has been made: either analo

or similarity. M is either similar or analogous in meaning to its

literal equivalent L. Once the reader has detected the ground of

the intended analogy or simile (with the help of the frame, or

clues drawn from the wider content) he can retrace the author's 61
path and so reach the original literal meaning (the meaning of L).

 

We recall that 'M' is the name of a metaphorical expression, not its
meaning. 'L' is the name of a literal expression, not its meaning.

Here Black is in a muddle. As he presents the comparison theory, the
analogy or similarity which ought to play a central role turns out to be
without any function. He states that "M is either similar or analogous
in meaning to its literal equivalent L." This statement is completely
safe since he assumes that L is the literal equivalent of M, their

meanings therefore must be identical; and things which are identical are

83

certainly similar. Talk about detecting grounds of similarity of mean-
ing is superfluous since the meanings are identical they have every
characteristic in common. One would not be 'retracing the author's
steps' to reach an 'original literal meaning' since in being aware of
the meaning of M, he is also aware of the meaning of L. At most, a
reader would note that there was another expression which had the same
meaning.

Black, in presenting this view, has perhaps lost sight of his own
terminology. M is an expression whose meaning "... in its metaphorical

52 Given this definition,

occurrence, is just the literal meaning of L."
if one understands the meaning of N, there would be no point in at-
tempting to determine L and therefore no need of and discussion of
analogy or similarity. A more appropriate explanation using Black's

terms would claim that the meaning of M given a literal interpretation

is analogous to the meaning of L given a literal interpretation. Since
the meaning of M, interpreted literally, is for some reason inappropriate,
one adopts the meaning of L as the correct meaning of M. Thus L is the
literal equivalent of M. Here the analogy or similarity is between the
meaning of M under a literal interpretation and the meaning of M under

a metaphoric interpretation. The analogy or similarity serves to de-

63 Since one could

termine the correct meaning of M in the context.
determine the meaning of M, under a metaphoric interpretation, on the
basis of similarities other than those between meanings, the L could be
eliminated from the discussion. This is an important consideration since
it is Black's requirement of L which separates his comparison view from

views which are held by others.64

84

The significant question which Black fails to bring out and which
is thought by some to be explained by the comparison View is the rela«
tionship between the meaning and/or referent of the term interpreted
literally and its meaning and/or referent interpreted metaphorically.
This relationship is, according to such views, analogy or similarity.

But neither the comparison view as explained by Black or the aug«
mented view just discussed characterize Black's examples and explanations.
Both those comparison views, in keeping with his demand for emphasis on
meaning, described the analogy as being between meanings. Yet, in the
light of what he goes on to say his understanding of the comparison
view appears to be that the analogy is between the referents of the focus
and the frame and that a metaphor may be correctly described as an
elliptical simile. Consider Black's example:

When Schopenhauer called a geometrical proof a mousetrap, he was,

according to such a view, payjpg_(though not explicitly): "A geo-

metrical proof is like a mousetrap, since both offer a delusive

;::::ggeengigeuggei?771ctims by degrees, lead to disagreeable
Notice that the comparison is not between the statement (M) calling the
proof a mousetrap and the statement (L) saying that the proof was like
a mousetrap as would be indicated by his discussion. It is rather be-
tween geometrical proofs and mousetraps, neither of which are linguistic
expressions. The comparison in the comparison view as described in this
example is not between the metaphorical expression and some literal
counterpart but rather between objects or between characteristics of

those objects. Yet his definition above stated that the analogy was to

be between the meanings of M and L.

85

Black offers the following definition of the comparison view which
is divergent from his earlier discussion.

If a writer holds that a metaphor consists of the presentation of
the underlying analogy or similarity, he will be taking what I
shall call a comparison view of metaphor.... This is a view of
metaphor as a condensed or elliptical simile. It will be noticed
that a "comparison view" is a special case of the “substitution
view." For it holds that the metaphorical statement might be re-
placed by an equivalent literal comparison.

 

Given this definition of the comparison view, metaphors would be com-
parisons. Black's earlier discussion held that the essential feature of
metaphor is an analogy or similarity of meaning between M and L. While
these explanations are not strictly speaking inconsistent neither follows
from the other and once again, it appears that two views are to go by

one name. This second comparison view still retains the requirement that
the metaphor have a literal equivalent. His expression "It will be
noticed ..." gives the impression that the requirement of a literal
equivalent follows from other statements in that paragraph. This is not
the case. The claim that metaphor may be regarded as a condensed simile
does not entail the claim that the uncondensed simile will have an equiv-
alent meaning. We emphasize this point since again it is the insistence
upon a literal equivalent which distinguishes this view from widely held
views and makes of his discussion a straw man argument.

While Black claims the comparison view to be a special case of the
substitution view, he claims that they are, in one respect significantly
different. The respect of their difference is, according to Black, the
degree of the elaboration one would find in the paraphrase of metaphor.

Black maintains that the "Richard is a lion." example is paraphrased,

w
1%!

86

according to the substitution view of "Richard is brave." According to
the comparison view it would be paraphrased as "Richard is likp_a lion
(in being brave)." But Black fails to distinguish which of two possible
claims that he could here be making. The first is: the comparison
theorist and substitution theorist agree on how figures of speech are to
be correctly paraphrased but do not agree on which figures are metaphor.
The basis of their disagreement would be that the comparison theorist
would insist that only those figures whose paraphrase contains 'like'
or 'as' are metaphors while the substitution theorist would recognize
a greater variety of metaphors. This would be an unlikely position for
any comparison theorist to take since his theory would be by definition
correct and therefore vacuous. The second of the two claims is that
while the substitution theorist and comparison theorist agree with
regard to which expressions are metaphors, they differ with regard to
the correct paraphrase. That is, the comparison theorist is committed
to an approach which the substitution theorist is not. The comparison
theorist is committed to finding a comparison as the paraphrase of
every metaphor. The substitution theorist is not committed to disagree-
ing with this procedure. He would in fact say that if the literal mean-
ing of "Richard is a lion." is "Richard is like a lion (in being brave)“
then that is the correct paraphrase; he would not insist, as Black seems
to suggest, that the words 'like' or 'as' must be absent from the para-
phrase. There is on this explanation no reason to believe that the
comparison theorist would produce a more elaborate paraphrase.

The second point on which Black is incorrect is the assumption that

a paraphrase of a sentence which is an explicit comparison will be more

P\c

87

elaborate than other paraphrases. "Richard is a lion' could be para-
phrased "Richard is extremely brave and makes bold frontal attacks upon
his enemy causing them to flee and when not engaged in such an attack
is quite serene since he is confident that no one dares challenge him.‘
This paraphrase obviously demonstrating that status as a comparison does
not in itself guarantee the greatest elaboration.

In turning to Black's criticism of the comparison view, we find
him saying “The main objection against the comparison view is that it

67 We are not sur-

suffers from a vagueness that borders on vacuity."
prised at this remark on Black's part since it reflects Black's presenta-
tion of comparison views. His discussion would have been stronger had
he picked the views of one or two individuals and addressed his criti-

68 Black's view is

cisms and remarks against a well worked out system.
mistaken because be based his understanding of comparison views on a

bad picture. His picture of metaphor was that of a one place function
as was the case with irony. Irony can be considered a function which
maps meaning onto meaning, in particular it maps meanings onto their
opposites. Black looked therefore for a similar function in the case of
metaphor. Continuing to think of mapping meanings onto meanings, he
invented a comparison view where a meaning is mapped onto a similar
meaning. What he should have found, if he wanted to talk about func-
tions and the comparison theory, is that metaphor involves a much more
complicated mapping. Such a function must take as arguments at least

all of: the meaning of the metaphor under a literal interpretation, the

conventional referent of the expression, pre-existing relationships

88

between meanings and between objects. An explanation of irony is vastly
simpler than an explanation of metaphor. The result of using irony

as a model is a rather confused explanation.

Interaction View

The theory of metaphor which Black proposes is called the interacs
tion view of metaphor. He claims this view to be free of the main
defects of the other views. Black provides no cryptic definition of
this view, but begins with a statement borrowed from Richards:

In the simplest formulation, when we use a metaphor we have two

thoughts of different things active together and supported by

a single word or phrase, whose meaning is a resultant of their

interactions.67
To distinguish his view from the comparison and substitution views, Black
compares the interpretations which the three views would provide for the
following example: 'The poor are the negroes of Europe.‘ The substitu-
tion theory, he claims, would interpret this example as saying something
about the poor of Europe; the comparison theory would present some
comparison between the poor and negroes. Black has Richards continue to
speak for the interaction view: "In opposition to both, Richards says
that our 'thoughts' about European poor and American negroes are 'active
together' and 'interact' to produce a meaning that is a resultant of

that interaction."70

The focus of the metaphor is held to take on another
meaning and this new meaning is distinct from any meaning which a literal
substitution would have. At this juncture Black's own view greatly
resembles the catachresis view. His attempt at distinction fails; he

has presented another version of the catachresis view. But the adoption

A

89

in part at least of the catachresis view is not unique to Black.
Every theory we examine is a version of the catachresis view. The sig-
nificant differences between those theories lie in the various explana-
tions of the ability of some expression to acquire the new meaning.
Black's discussion of metaphor's ability to acquire new meaning
presents some genuine innovations. He makes use of the following meta-
phor as an example in his explanation: "Man is a wolf." We may, says
Black, say that there are two subjects, one principal (the man), the
other subsidiary (the wolf). The metaphor conveys its intended meaning
only to a reader with certain beliefs about these subjects.
What is needed is not so much that the reader shall know the
standard dictionary meaning of "wolf"--or be able to use that
word in literal senses--as that he shall know what I will call
the system of associated commonplaces. Imagine some layman re-
quired to say, without taking special thought, those things he
held to be true about wolves; the set of statements resulting
would approximate to what I am here calling the system of common-
places associated with the word wolf.71
The effect of the metaphor-is the result of associating with the primary
subject of a part of the system of commonplaces associated with the
secondary subject. That is, a suitable hearer will construct a set of
attributions to the principal subject based upon the commonplaces
associated with the secondary subjects. The members of the set of
commonplaces which are directly applicable to the primary subject are
taken over as is; other members are applied metaphorically. "Each of
these implied assertions has now to be made to fit the principle sub-

1:."72 For example, a man and wolf can both be seen as vicious, but

jec
these terms describe different sorts of behavior when applied to men

than when applied to wolves. The attribution of viciousness to a man by

90

'Man is a wolf' extends only to the vicious characteristics usually
found in men. Some commonplaces will not apply and are filtered out.
The metaphor, on the interaction view, not only attributes character«
istics of the secondary subject to the primary subject, it changes our
view of the primary subject's own (or perhaps normal) characteristics.
Those characteristics which are consistent with the make-up of subsi-
diary subject are accentuated while those which are inconsistent are
ignored. Recalling the man—wolf metaphor, Black states: "Any human
traits that can, without undue strain, be talked about in 'wolf—lan-
guage' will be rendered prominent, and any that cannot will be pushed
into the background. The wolf—metaphor suppresses some details,

"73 Black

emphasizes others-—in short, organizes our view of man.
olaims: "If to call a man a wolf is to put him in a special light, we
must not forget that the metaphor makes the wolf seem more human than
he otherwise would." Istvan Mészaros objects to this view. He says:
"Black seems to have forgotten that more of nothing (otherwise would)
is still nothing." He continues later, making reference to Black's
example: 'the chairman ploughed through the discussion,‘ that the meta—
phor does not "affect in any way the subject" (not unless one wants to
maintain that the chairman is put in some "special rural light," like
the wolf in a “human iight.")74

Black includes in his interaction view of metaphor consideration
not only of traits of subjects, but consideration of our attitudes
towards those subjects. He says to call a man a wolf is, in part, to

imply that he is hateful. Black therefore recognizes an important

emotive aspect of metaphor. But he fails to mention an equally

91

important emotive function of metaphor. That function is the evoking
of various attitudes and feelings of the reader. For example, in addi-
tion to expressing one's hatred by calling a man a wolf: one may arouse
anxiety or a feeling of alarm in the reader. Since Black says nothing
that would rule out this second emotive function, it seems that he
would accept it as an addition, perhaps as a friendly amendment.

While Black calls his view an interaction view, his discussion
indicates that his view is one species of the comparison view as we
have described it. That is, according to his discussion, the metaphor
presents the audience with two things and the audience considers their
relationship. Black's explanation reflects one of Richard's defini-
tions of comparison, namely: "a study of them both to see how they are

alike and unlike one another."75

Talk of "our ideas of two things active
together" suggests a comparison of those things. One sees how they are
alike and unlike as some traits are "rendered prominent" and others
"pushed into the background." Black does not claim to hold a comparison
view of metaphor, but his characterization of that view differs from
ours. His explanation falls clearly within our definition of a compari-
son theory in that his explanation presumes that the audience is pre-
sented with two objects and determines the relationships between them.
In criticizing Black's interaction view of metaphor we will start
by examining a criticism suggested by Black.
A fairly obvious objection to the foregoing sketch of "interaction
view" is that it has to hold that some of the "associated common-
places" themselves suffer metaphorical change of meaning in the
process of transfer from the subsidiary to the principle subject.

And these changes, if they occur, can hardly be explained by the
account given. 5

92

We remember Black's discussion of the man—wolf example. The ferocity
of a wolf is not the ferocity of a man. "Fierce" as used to describe
wolves applies to man metaphorically. To explain, then, how "fierce"
applies to a man will be to explain metaphor. According to Black,

"the primary metaphor, it might be said, has been analyzed into a set
of subordinate metaphors, so the account given is either circular or
leads to an infinite regress.“ To this "fairly obvious objection"
Black provides two answers, both of which are unsatisfactory. He sugv
gests that the objection "might be met by denying that all_changes of
meaning in the 'associated commonplaces' must be counted as metaphorical
shifts. Many of them are best described as extensions of meaning...."78
But this suggestion errs in two ways. First denying that all changes
are metaphoric will not explain those metaphors whose transfer of meaning
involves metaphor. If Black could show that for every metaphor, a point
is reached at which the change in meaning of the associated commonplaces
is not metaphorical, then he would avoid one aspect of his objection.

But he neither shows nor claims this and as his explanation stands he
fails to avoid the possibility of infinite regress of metaphors explain-
ing metaphors. The second error lies in Black's statement that the
changes of meaning can be explained in terms of extension of meaning.
While the extension of meaning has been associated with metaphor, the
hope has been that a theory of metaphor would explain such extensions.

If we, for example, consider the word 'green,' we can reconstruct the
extension of its meaning. Many fruits have the color green before they
are ripe. One naturally and literally uses 'green apples' as synonymous

with 'unripened apples.‘ Green then was applied (metaphorically) to

93

other objects which were not ready or finished in some respect. A new
farm hand might be said to be green. 'Green' is now used as a term
meaning not-ripe and such use is literal, conventional. 'Green' has
extended its meaning. And this extension is based upon metaphor. Since
some extensions of meaning result from metaphor, Black fails to avoid
his objection by appeal to change in meaning. Further, if Black were to
maintain that his argument is adequate he would foreclose the possibility
of explaining extension of meaning in terms of metaphor. Perhaps Black
realizes some difficulty here since he says at this point. "I have not
undertaken to explain how such extensions or shifts occur in general,
and I do not think any simple account will fit all cases."79

On another tack, Black presents an even more confusing explanation
to avoid his suggested difficulty.

Secondly, I would not deny that a metaphor may involve a

number of subordinate metaphors among its implications. But

these subordinate metaphors are, I think, usually intended to be

taken less "emphatically," i.e., with less stress upon their

implications.... In any case, primary and subordinate metaphors

will normally belong to the same field of discourse, so that they

mutually reinforce one and the same systems of implications.80
Apparently since these secondary metaphors are to be taken less empha-
tically, they need no explanation; or so Black would have us believe.
But recalling the objection, without an explanation of the secondary
metaphor one has no explanation of the primary metaphor. A simple wave
of the hand will not do. The fairly obvious objection has not then been
satisfactorily replied to by Black.

Two other objections come to light upon examination of the first

three of seven commitments Black makes on behalf of his interaction

theory. If his view is really committed as he indicates, then that view

94

is not adequate to characterize metaphor since it will not recognize
obvious cases of metaphor as metaphors. The first and third commitments
are:

(l) A metaphorical statement has two distinct subjects-~a "prin-
cipal" subject and a "subsidiary" one.

(3) The metaphor works by applying to the principal subject a
system of "associated implications" characteristic of the
subsidiary subject.

These two commitments are apparently based upon a single type of metaphor.
Being so based, the theory is not descriptive of the full range of meta-
phors. The example which seems to have captured Black's attention is
"Man is a wolf.‘I This clearly has the form 'x is an m.’ It is obvious
that two things are mentioned, perhaps even compared. But many meta-
phors are not of this form because they either fail to have two subjects
or because the subject or subjects which they have lie within the frame
of the metaphor and undergo no change in meaning via associated implica-
tions. Black's earlier example is just such a metaphor: "The chairman
ploughed through the discussion." This sentence has but one subject.

It could be claimed that the sentence was about two things, the chairman
and the discussion, but both these are contained in the frame of the
metaphor, not its focus. In any case such a metaphor has one subject and
a direct object rather than two subjects. Black's explanation will here
fail since the focus of the metaphor is not a subject. While it is the
case that there is a system of commonplaces which is associated with

the act of ploughing, there is no primary subject to which the system of
commonplaces is transferred. There are, in short, no two subjects which

interact.

95

The second commitment of Black's view is further evidence that he
has based his discussion on this single example. "(2) These subjects
are often best regarded as "systems of things," rather than "things"."82
Had he considered the "ploughed" example, he would not have expressed
his theory in terms of things, since the word 'thing' connotes a sub-
stantive (though not material) entity as opposed to an action or set of
actions. While one might very well want to compare ploughing with chair—
ing a meeting it would be at least misleading to speak of comparing
things. In general, explaining metaphor in terms of subjects and things
at least suggests that the focus of the metaphor will always be either
the subject or object of the expression as opposed to the verb or perhaps
modifiers such as adverbs and adjectives.

We notice in the conclusion of Black's article a change in tone.
This change amounts at least to a qualification of his view. There is
a shift from talking about types of theories of metaphor to talk about
types of metaphor. It had seemed that his discussion of the substitution
and comparison views were essentially rejections of such views as a-
basis for his own superior view. But in his concluding remarks he speaks
as though those theories correctly characterize some metaphors which his
own theory does not.

Now it is in just such trivial cases that "substitution" and

"comparison" views sometimes seem nearer the mark than the "inter-

action" views. The point might be met by classifying metaphors as

instances of substitution, comparison, or interaction. Only the
last kind are of importance in philosophy.83
Given this change in tone, the three views are recognized as each being

correct for some metaphors. Black's interaction view, rather than having

application to metaphor in general is a description of a particular type

96

of metaphor, namely, the interaction metaphor. This is a serious retreat
from what had appeared to be his aim, the explanation of metaphor in
general. This position is further weakened since he provides no argument
that the three views together characterize all types of metaphor. At
most, it seems that Black concludes that there are at least three types
of metaphor described by the substitution view, the comparison view and
the interaction view and perhaps that there are some metaphors of each
type.

Black offers advice regarding use of these various types of metaphor.
He advises that we avoid use of the substitution and comparison metaphors.
These sorts of metaphor can by definition be translated into literal
speech (except for catachresis) and could be eliminated with no loss of

ineaning. Interaction metaphors are to be retained since they have no

literal equivalents. They “are not expendable. The interaction meta-

]:hor cannot be reduced to a series of literal statements since ... "the
set of literal statements so obtained will not have the same power to

85

iiiform and enlighten as the original." The importance of the inter—

a<:tion metaphor finds its basis in this supposed untranslatability.
Black's argument for the untranslatability of interaction meta-
Phors fails to establish that such metaphors cannot be translated. His
argument is that any literal statement will be too precise.
For one thing, the implications previously left for a suitable
reader to educe for himself, with a nice feeling for their rela-

tive priorities and degrees of importance, are now presented
explicitly as having equal weight.35

This argument involves two issues. The first is that metaphor enjoys a

deQY‘ee of either vagueness or ambiguity which provides the audience a

111

-\v

.FI

97

variety of alternative interpretations in inferences. But Black's argu-
ment fails to establish that this same degree of ambiguity could not be
found in the literal use of some expression. It is the case that
generally, when one attempts to analyze metaphor, he attempts to resolve
ambiguity much the same as he would attempt to resolve ambiguity in
explaining a literal sentence. He is not seeking a literal equivalent.
But this in no way rules out the possibility that there is a literal
statement which has precisely the ambiguities contained in the metaphor.
When the metaphor in question is ambiguous, no unambiguous literal state-
ment will be equivalent to it. This is clear but Black's claim was much
stronger.

The second issue is that of priorities. Black claims that the
implications of any metaphor would, in their literal expressions, be
necessarily of equal Weight. But again there is no argument for this
claim. It would seem that a literal expression could contain phrases
:such as 'this is the most significant implication,‘ 'this is about as
:significant as the last,‘ etc. Such a procedure would preserve any
cognitive import of the relative priorities. Black has failed to show
either that the ambiguity or the relative weight of the implications
could not be captured by a literal statement.

In continuation of his argument for the untranslatability of inter-

action metaphors Black states:

The literal paraphrase inevitably says too much--and with the wrong
emphasis. One of the points I most wish to stress is that the

loss in such cases is a loss in cognitive content; the relevant
‘weakness of the literal paraphrase is not that it may be tiresomely
prolix or boringly explicit--or deficient in qualities of style:

it fails to be a translation because it fails to give the insight

that the metaphor did.87 .

98

This is a return to question begging. To claim that there is no lit-
eral equivalent on the basis that no literal expression says the same
thing, is to say that there is no literal equivalent because no literal
expression is equivalent. What is required to hold Black's position

is a showing that no literal expression gpplg_mean the same as the
metaphorical expression. We suggest that no proof is possible. In
discussing the work of Mfiller in the previous chapter we noticed some
metaphoric uses become literal uses without a change in meaning. Black
recognized this in his discussion of catachresis. In such cases a
literal expression would be synonymous with a previous metaphoric expres-
sion.

In short, Black has failed to show that the interaction metaphor
has no literal equivalent. If he retreats to the position that inter-
action metaphors are defined as having no literal equivalent, then he has
failed to show that there are any interaction metaphors.

But even if we allow that there are interaction metaphors and
substitution metaphors, Black has provided us with no means of distin-
gLJishing them. He advises us to avoid those metaphors which have a
l iteral equivalent. But he fails to answer the important question:
"hHiich metaphors have a literal equivalent?" It is recommended that we

employ only those metaphors which have no literal equivalent, but Black

fails to show that there are any, and he fails to explain a process for
dT'Stinguishing them. .

In examining "Metaphor" we found that Black argues against two sup-
POSed positions and presents his own as an alternative. Such a move has

I‘itifle value since the views which he presents are not held by anyone

99

and are not held for good reasons. The basis of his objections to

those views is a feature which similar views, which are defended, lack,
namely, the requirement that the metaphor have a literal equivalent.
Generally he ignores the treatment of what he calls the catachresis
metaphor. It is this metaphor which is of greatest importance to phil-
osophy. His charge that the comparison views suffer from vagueness
seems to be a characterization of his discussion of those views. All of
these criticisms of his work derive from the fact that he is jousting
with straw men rather than seriously held theories.

His own theory describes metaphor as 'two thoughts of different
things active together and supported by a single word or phrase, whose
meaning is a resultant of their interactions.‘88 While this line is
quite poetic it is hardly an explanation of metaphor. Black's view re-
lies heavily on Richards and on the examples 'The poor are the negroes
of Europe' and 'Man is a wolf.‘ Since Richards relies on that same type
of example we are not surprised to see both Black and Richards holding a
comparison view.

Two major objections to his view were at once evident to Black
liiinself. He is not able to escape from them. Moreover by basing his
theory on a single type of example, his explanation even if correct would

be applicable to only metaphors of that one type. Black's interaction
view proves to be no serious improvement upon Richards' effort.

Black makes two positive contributions in his discussions. The
f:l'Y‘Si: is that metaphor is not constituted by a single word but is made
“‘3 (If two parts which he calls 'focus' and 'frame.‘ His examples indi-

cate that the focus-frame compound generally includes a whole sentence.

100

His suggestion may be expanded to allow the frame to include a consider-
able amount of context. This is a point we further expand upon in
speaking primarily of metaphoric uses of sentences rather than metaphoric
sentences. Whether an utterance will be construed as metaphoric will
depend on the'situation of that utterance as well as the conventional
meanings of its terms.

Black's second major contribution is the recognition that it is
our opinion about objects rather than the truth about them which is sig-
nificant in understanding metaphor. A community's beliefs about objects
or events form a system of associated commonplaces. It is those common-
places which stand out in people's minds that determine how a metaphor
will be interpreted (or for that matter whether an utterance will be
interpreted as metaphoric). These points must be incorporated in any

correct theory of metaphor.

11.
12.
13.
14.
15.

O ‘0 00 \l 01
O O O O

101

NOTES TO CHAPTER III

. Since this is a dead metaphor, we recognize the dual reference as a

matter of convention. For the purpose of this discussion we
assume that this is not the case, that the metaphor is fresh.

. We use 'presentation' and 'presented with' as neutral terms since in

various formulations of these views, the method by which the
audience is said to be made aware of A, B, and R varies.

. I. A. Richards, The Philosophy of Rhetoric (New York: Oxford Univer«

sity Press, 1965), p. 116.

. The examples presented as 'leg of a table' and calling people pigs

and ducks. (ibid., 117). In addition he earlier states "...
when we use metaphors we have two thoughts of different things
active together and supported by a single word, or phrase, whose
meaning is a resultant of their interaction" (ibid., 93).

. Compounding two uses might describe a pun since a pun involves two
0

uses th of which are independent of the expression.

H

bid., 93.

bid., 100.

H

 

H

bid., 120.

H

bid., 93.

 

. W. V. Quine, "On What There Is," reprinted in From a Logical Point

of View (New York: Harper and Row Publishers Inc., 1953), p. 3.
Richards, The Philosophy of Rhetoric, 119.
_I_b_i_d_., 96.
Ibi

Q

., 117.
Ibid., 118.

 

The word 'sort' is for Richards a semi-technical term which he intro-
duces in a sketch of an epistemological theory. It is that dis-
cussion upon which Richards bases his claim that all thought is
metaphoric. In all thinking

"... we begin with the general abstract anything, split it, as
the world makes us, into sorts and then arrive at the concrete
particulars by the overlapping or common membership of these

16.
17.
18.
19.
20.

21.
22.
23.
24.

25.
26.
27.

28.
29.

102

sorts. This bit of paper here now in my hand is a concrete
particular to us so far as we think of it as paperish, herish,
nowish, and in my hand; it is more concrete as we take it as
of more sorts, and the more specific as the sorts are narrower
and more exclusive." (Rhetoric 31)

We choose not to examine Richards' epistemology since statements
made in its presentation conflict with things he later says in
regard to metaphor.

Ibid., 120.

 

 

Ipi_,

Ibid., 122.

Ibid., 127.

gplg,, 125. The statement that the mind works only by connecting
things appears to be in contradiction with Richards' character-
ization of thought and perception as a matter of sorting.

Ipig,

ijg,, 126.

gpjg,

We make use of Richards' discussion in our examination of Goodman.

We shall argue that every grammatically correct sentence is
meaningful and perhaps true if the proper context is provided.
Such a claim might seem obvious but Bertrand Russell denies
that 'Consanguinity drinks procrastination' is meaningful even
though he recognizes it as grammatically correct. Goodman's
position will result in the claim that no metaphoric interpre-
tation of a statement is true; we shall argue that an appro-
priate context will result in an interpretation under which
the sentence is true.

Ibid., 125.
This explanation is in part due to Black (Black 278).

Max Reiser, "Analysis of Poetic Simile," Journal of Philosophy,
XXXVII (1940), pp. 209-217.

Richards, The Philospphy of Rhetoric, p. 118.

Reiser, "Poetic Simile,‘ p. 210

30.
31.
32.W
33.
34.
35.

36.

37.

38.
39.

40.
41.
42.
43.

44.
45.

103

;_jg,
I_bi_d_., 210- 211.
ede a1 with this issue in Chapter III.
;p_g,, 213.
;p___., 214.
Max Black, "Metaphor," Proceedings_of the Aristotelian Society,

 

vol. 55 (1954-1955).

While Black cites two holders of the substitution view it will be
argued on page 76 of this paper that neither in fact hold the
view which Black described. We know of no one except Black
who holds such views.

Ibid., 275. Since in our discussion of Goodman's treatment of
metaphor we discuss at length the difficulties with distinguish—
ing literal senses from metaphorical senses, we shall not here
argue against the use of these terms here. Black in the second
succeeding paragraph speaks of metaphorical uses and literal
uses and they would serve as well for his present purposes.

Ibid., 276.

We argue this on page of this paper that it must indeed be uses
of sentences rather than sentences which are the metaphors.
The utterance of 'His heart is not in the right place' in an
operating room would hardly be treated as a metaphor.

 

I owe this example to Mrs. Carmen Suter.

For example: “Fascism is a cancer.... Literally, fascism is a form
of government; in the first sentence the writer asks the reader
to tolerate the fact that "fascism" is not organic (that is,
literally alive), in order that he, the writer, can attribute
to a political system the connotation of abhorrence that most
people have toward cancer." Owen Thomas, Metaphor and Related
Subjects (New York: Random House, 1969), p. 24.

bid., 292.

Ibi ., 277.

 

H

46.

47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.

59.

104

One rule which might be proposed is discussed in the following
chapter. That rule is: when an expression makes no sense
given a literal interpretation, try a figurative interpretation.

bid., 279.

0—.

 

H
U'
do
Q...

 

H
U-
do
a.

 

H
U'
..a

Q.

’ ., 280.
, 294.

 

0—4
0'
—l

O.

H
U
—.l

D.

' ., 280.

 

H

bid.

 

For example: Henle, Richards, and Wheelwright.

bi

H

H
G.

bi

D.

, 281.

0—.
CT
do
D.

 

George Santayanna, The Sense of Beauty (New York: Dover Publications,
Inc., 1955), p. 28.

60.

61.
62.
63.

Ibid., 279.

Essentially the same explanation can be made using the note on termin—
ology at the end of his article. Black's attention is focused
on the relationship between H] (E) and M (X). If we take seri-
ously Black's condition that L is the literal equivalent of M,
then the relationship;between M' (E) and M (X) is identity.
Given that this is the case, there is no need to invoke analogy
or similarity. A distinction which Black fails to invoke is
between the meaning of E in F (E) under a literal interpretation
(M(E)) and Mf(E). Employing this distinction one could charac-
terize a comparison theory as follows: M(E) is either similar
or analogous to M'(E). M (E) cannot be the forrect interpre-
tation of E in F(E) and 11Once the reader has detected the ground
of the intended analogy or simile...." Ibid., 294.

64.

65.
66.
67.
68.

69.
70.
71.
72.
73.
74.

75.
76.
77.
78.
79.
80,
81.
82.
83.
84.
85.
86.
87.
88.

105

His discussion of the comparison theory would be more significant
if he could demonstrate that serious writers in this area
actually hold the view he describes.

For example: Richards or Henle whose 'Metaphor' was available prior
to the publication of Models and Metaphor.

Ibid., 285.
Ibid., 286.
Ibid., 287.
Ibid., 288.

Istvau Mészaros, "Metaphor and Simile," Proceedings of the Arise
totelian Society, vol. N.S. 67 (1966—1967), p. 131.

Richards, The Philosophy of Rhetoric, p. l20.
Black, 289.

Ibi ., 290.
Ibi ., 291.

Ibi ., 292.
Ibi ., 293.

Ibi .. 285.

CHAPTER IV

RELATIONAL THEORY

As we pointed out both Black and Richards hold comparison views.
They both interpret metaphor as being a matter of comparing objects,
perhaps a matter of examining two objects to determine their likenesses
and differences. While Henle follows or parallels some of the points
of Black and Richards, it is obvious at the outset that he holds a
resemblance theory which is briefly characterized at the beginning of
the previous chapter. The distinction between the comparison and re-
semblance theories being that in the comparison theory the metaphor
states a comparison, while in the resemblance theory comparison or
resemblance is employed to determine the referent of the metaphor.
Henle's work fits this definition since he is most interested in explain-
ing our understanding the referent of the metaphor based upon our recog-

nition of similarity.

Paul Henle

Henle claims, in "Metaphor"1

, to follow Peirce. In so doing he
attempts to use Peirce's terminology, 'icon' and 'symbol' in such a way
as to establish a paradigm resemblance theory. We argue, however, that
Henle does not follow Peirce, nor is his conception of the resemblance

theory of much help in understanding metaphor. Rather, we argue that

106

107

his commitment to the resemblance view tends to interfere with his
genuinely instructive suggestion.

Throughout his article Henle speaks in terms of figurative sense
and literal sense. This suggests difficulties we find in Goodman's

discussion2

and our criticism of theories involving different types of
meaning. Careful attention to Henle's definitions will show that his
theory does not entail the sort of two-meaning theory to which we
object. It will show that 'figurative sense' and 'literal sense'
rather than names of radically different types of meaning are ways of
avoiding circumlocutions for the sense of an expression as used liter-
ally or figuratively.

By the literal sense of a word we may mean the sense which a word

has in other contexts and apart from such metaphoric uses. By

figurative senseAwe may mean that special sense on which the
metaphor hinges.5

 

'Literal sense' and 'figurative sense' both refer to meanings of
terms. We shall want some way of referring to the relationship
between a word and its various meanings. This may be accomplished
by saying that a word is an immediate Sign of its literal sense

and a mediate sign of its figurative sense. These terms are appro-
priate since it is only through the literal sense that one arrives
at the figurative.4

 

 

These definitions do not necessitate a dual meaning theory which entails
different types of meaning. They are consistent with maintaining a
single type of meaning and nothing Henle says hinges upon distinguishing
types of meaning. We do, however, note that his definitions do not

specifically rule out a two-meaning theory.

Iconic View
Henle attempts to base his explanation of metaphor upon Peirce's

discussion of language. "... it is possible to explain metaphor as a

108

type of symbolism and to assimilate it to a more general theory of

5. Such an assimilation would be valuable in terms of

symbolism"
describing metaphor as a regular feature of language rather than an
exception to the general explanation. We find this effort to lack that
value since he fails to complete the assimilation. He employs the terms
'symbol' and 'icon' of Peirce and on the surface, employs them much as
does Peirce. Further investigation shows that Henle's use varies sig-
nificantly from Peirce. Henle's definitions however square with those
of Peirce.

A sign is a symbol insofar as it signifies according to an arbi—

trary rule, insofar as it is a conventional sign. A sign is an

icon to the extent that it signifies in virtue of similarity.6
It is primarily upon Peirce's use of 'icon' that Henle's theory turns.
And it is Henle's attempt to follow Peirce which results in his theory
being a relational view. Peirce's discussion of icons is vital to our
consideration and we therefore summarize it below:

An icon is a sign which refers to the object which it denotes

merely by virtue of characters of its own and which it possesses

just the same, whether such an item actually exists or not....

Anything whatever, be it quality, existent individual, or law, is

an icon of anything insofar as it is like that thing and used as

a sign of it.
Henle makes use of Peirce's notion of symbol and icon in explaining
metaphor as a two part relationship. The metaphor acts first as a symbol
to refer to an object or event. The object or event, in turn, acts as
an icon referring to the object or event under discussion. Henle
describes this as follows:

Metaphor is analyzable into a double sort of semantic relationship.

First using symbols in Peirce's sense, directions are given for
finding an object or situation. Thus, use of language is quite

109

ordinary. Second it is implied that any object or situation fit-

ting the direction may serve as an icon of what one wishes to

describe.8
Henle introduces these two technical terms (symbol and icon), but to no
serious advantage. His explanation amounts to no more than saying that
the name of one thing is used in place of some other similar thing and
because of the similarity, we can determine the intended referent. Not
only do we not see an advantage to such an explanation, Henle's use of
Peirce's term is genuinely misleading. Consider his explanation of the
icon's role in metaphor.

We may say that we are given not the icon, but a description of

what would be an icon. Alternatively, one might say that not the

icon, but its essence is brought before the reader. Again to say

approximately the same thing more safely, one may claim that what

is presented is a formula for the construction of icons.9
Now consider the line from Keats' "To Hope" which Henle presents as an
example of metaphor:

"When by my solitary hearth I sit,

and hateful thoughts enwrap my soul in gloom"

not only do we not find an icon here, no event is described by the word
'enwrap' although an event is referred to; the essence of enwrapping is
not brought before the reader. (Consider having the appropriate essence
brought before you in regard to the metaphor 'Richard is a lion‘.) No
formula for the construction of anything is presented. None of the
events mentioned in Henles explanation take place. The problem which
Henle is trying to get around, by this talk of bringing essences and
giving directions, is the fact that there is no icon involved in metaphor.

And he recognizes this in saying "we may say that we are given not the

icon...."

110

The icon is never actually present; rather, through the rule,

one understands what it must be and, through this understanding,

what it signifies.10
On this view, one is referred to an object or an event by the terms in
the metaphor-—the mysterious rule--and understanding of what the icon
must be, is nothing more than understanding the conventional use of
language and to what terms customarily refer.

He mentioned above that we found no icon. Metaphor simply is not
iconic. The strongest claim that this view can successfully make is
that there are objects and events which would function as iconic signs
if they could in fact be presented. But, in fact, no such iconic signs
are presented and it is genuinely misleading to speak as if they were.
Further, it is a disservice to attribute such a theory to Peirce.

While we reject Henle's iconic theory, many of his observations
concerning metaphor are correct. Such observations might be considered
augmentations of his view, but talk of icons seems to obscure rather
than help his position. A clearer position is attained by ignoring
insofar as possible the notion of icon and concentrating on his observa-
tions. None of his positive points depends on the iconic theory and
insofar as he attempts to relate his arguments to that theory, his efforts
are confused. The first observation with which we deal is that similarity
alone will not account for our ability to determine the referent of a
metaphorical expression.11 Henle explains:

While it may be that any two things are similar in some respect,

this by itself is not sufficient to make one an icon of the

other unless someone is led to consider one by its similarity to

the other. The similarity must be noticed and used as a means of

signifying. This would require at the very least that the simi-

larity between an icon and what it signifies should not be shared
by a great many other things as well. 2

111

Other restrictions need to be placed on the similarity. Since even if
there was but a single object which shared a particular quality with

the icon, we may well not understand the metaphor since we may not con-
sider that particular quality of the icon. Not only, as Henle noticed,

is every object similar to every other object, there is a very large
number of similarities which hold between objects. And generally speak-
ing, the metaphor fails to state the nature of the similarity to be con-
sidered. In explaining how we are aware of the similarity or relationship
to be considered, Henle avoids these difficulties and presents a view
which is in line with our description of the relational view.

Henle explains that our knowledge of the ground of the comparison,
that is, the relevant relationship, is derived from some feature of the
literally named object or event. He agrees with Black's claim that a
knowledge of a society's beliefs and habits is necessary for understand-
ing metaphor. One must know '... minor facets of the general culture,
such as what characteristics are uppermost in people's mind'.13 In less
detail than Black, Henle expresses essentially the same position with
regard to commonplaces. He provides an interesting example:

A popular song of some years ago praised a young lady by saying

to her "You're the cream in my coffee.“ An entirely wrong impres- 14

Sion would be obtained in a community which drank its coffee black.
The minor facets and characteristics in people's minds indicate what
sorts of grounds of similarity one ought to search to discover the simi-
lar object. While the R is not provided by the expression itself it is
provided by context and those facets uppermost (hopefully) in the audi-

ence's mind. Henle, we notice, is careful not to explain too much.

112

Too much, in this case, would be our unfailing ability to understand
metaphor. He states: "This is not to claim of course that everyone
can understand every metaphor. Often of course metaphors are not
understood."15 As suggested by Henle's 'cream in my coffee' example,
if the facets foremost in the utterer's mind differ from those in the
minds of the audience, confusion and misunderstanding will result.
Comparisons will be made on grounds other than those intended or perhaps
none will be made at all since none may seem appropriate given a differ-
ent ground of comparison. Henle's explanation of metaphor is more
interesting than many since he not only accounts for an ability to com-
municate via metaphor, he also accounts for those misunderstandings
and frustrations in attempting to understand metaphor which actually
occur.

Further insight into the nature of the R involved in metaphor can
be gained from Henle's distinction between metaphor and other tropes.
He claims that the basis for this distinction is the amenability of
metaphor to multiple interpretation. We shall argue that a better basis
for his distinction is in terms of the relationship (or similarity) upon
which the trope is based. Henle observes that "The aptness of metaphor
depends on the capability of elaborating it--of extending the parallel

structure".16 He goes on to say: "It is this capability of extension

which justifies the differentiations of metaphor from other tropes."17
The other tropes which he discusses are synecdoche and metonomy. In the
case of synecdoche further elaboration is not, according to Henle,
possible. By definition, the relationship involved in this figure is

that of part to whole or whole to part. And according to Henle once

113

“... one apprehends the relationship, one gains a flavor of correctness,
(one perhaps attributes to the entire genus something of the feeling
toward the species and that is all. No further development of synecdoche
is possible."18 This lack of further development is of sufficient
import to separate metaphor from synecdoche. We suppose that the same
sort of reasoning is intended by Henle to apply to metonomy. But there
is a more significant difference between metonomy and synecdoche, and
metaphor than Henle's discussion emphasizes. It is more significant
since it is the basis of the distinction which Henle notices and it
lends further insight into the nature of metaphor. In synecdoche, a
definite relation is specified and even in the case of the whole stand-
ing for the part there is a fairly limited number of parts and in con-
sidering them, elaboration comes to an end. The relationship involved
in metonomy is not so precise. It is described as consisting of "...
looser connections by way of some relation felt to be important...."19
Metonomy has a looser connection in two ways: (l) no definite relation-
ship is specified and (2) there may be no definite number of objects to
which the named object is related. However, the first sense of loose—
ness is somewhat restricted. It appears that Henle intends to make
reference by 'metonomy' to figures which will inhibit any tending to
consider more remote relationships. Henle's example seems to illustrate
just that sort of relationship: "... when one speaks of reading an

"20 In the context of reading, it may be

author instead of his works....
that the relation authorship is foremost and we consider no other. This

suggests a basis for distinguishing between metonomy and metaphor.

114

That basis is the existence in the case of metonomy of a single obvious
relationship. That relationship provides an interpretation so clearly
correct that it effectively prohibits our consideration of alternative
interpretations based on different relationships. Metaphor, which
does not involve such a relationship, is open to alternative interpre-
tations many of which are reasonably correct. It is the number of
relationships which constitute reasonable alternatives that accounts
for the number of alternative interpretations. Henle's talk of elabora-
tion and extending parallel structure points to this basic difference
between metaphor and other tropes. The distinction between metaphor and
these other tropes is based on types of relationships which generate the
figurative interpretation. We shall return to a consideration of these
relationships as a part of Henle's explanation of our ability to under-
stand metaphor.

In explaining our understanding of metaphor, Henle relies, in part,

2] He claims

upon "... the sort of shock which it (metaphor) produces“.
that "ordinarily one takes words in their literal sense and this is
impossible in metaphor. This impossibility is in fact what drives one

to seek a figurative sense."22

Henle points out two respects in which
taking words in their literal sense may be impossible. The first respect
is that the sentence itself may be absurd or meaningless when interpreted
literally. The second respect is that although the sentence alone makes
sense, in context it is absurd. "Whether taking all of the terms in '
their literal sense produces an absurdity or merely something incongruous,

23

the clash of literal meanings must be felt". It appears, though Henle

does not mention it, that the clash of literal meanings is operative in

115

other figures as well as metaphor. As with metaphor, the clash provides
the impetus for discovering a figurative interpretation. This effort

to discover a figurative interpretation is based on "... the general
principle that people try to talk sense and to make statements which are

at least possibly true...."24

Combining the recognition that a literal
interpretation is not appropriate with the principle that people try to
make sense, one attempts a reasonable interpretation of tropes. In this
attempt, one tries various interpretations based upon the various rela-
tionships between objects and events mentioned under a literal interpre-
tation and other objects or situations which if considered as the refer-
ents would make sense of the trope. In the case of synecdoche, the part-
whole relationship is so fundamental or obvious that we gain a flavor of
correctness and do not try out other possible interpretations. In meto-
nomy some other single relationship is outstanding and we do not consider
other interpretations; a single interpretation seems obviously the cor-
rect one. In the case of metaphor no single interpretation is uniquely
correct. No single relationship stands out as a single parallel between
the object mentioned in literal use and the object mentioned in figura-
tive use.

We notice that in Henle's explanation the clash is not treated as
merely an additional feature to heighten our enjoyment. It is required
to perceivemetaphor at all. If the clash is not felt

... one of two situations must obtain--either the passage is taken

literally without encountering any difficulty and no suspicion of

a metaphor arises, or else the figurative meaning is so usual that

the reader goes to it immediately. In this case one has an ggiom
or a "dead metaphor“ which, properly, is no metaphor at all.

116

In the latter case one will not be misunderstood although he is not
likely to impress anyone with his authorship. The former case suggests
that in employing metaphor one must insure that a conventional interpre-
tation will be sufficiently out of place to insure that a figurative
interpretation is attempted.

While we argue vigorously in opposition to Henle's presentation of
an iconic theory, we find many of his other points to be insightful.
By combining those other points with the iconic view, a reasonable
theory might be produced. However, none of Henle's positive contributions
depend on the iconic view. They are matters which are not usually
associated with iconic language. We recall those points are: (l) the
clash of literal meanings indicates that a figurative interpretation
may be appropriate; (2) the principle that people usually attempt to
talk sense encourages the figurative interpertation; (3) the character-
istics uppermost in people's minds guide the attempted interpretation.
The appeal to icons was to be of value in suggesting that similarity
between objects or situations was to be the basis of the interpretation.
This suggestion, however, is as misleading as it is helpful. While we
agree that there must be some relationship uponwhich an understanding of
the metaphor is based, that relationship.need not be similarity,
particularly not the sort of similarity which governs iconic reference.26
Consider Henle's own example:

You're the cream in my coffee.
While a reasonable interpretation of this metaphor may be that the effect
of the young lady on the life of the author was that of enrichment,

there is no similarity except in the most strained sense between the

117

young lady and cream or between her actions and that of cream. But
according to Henle, the cream in the man's coffee is an icon either of
her or of her actions. While we admit that there may be a similarity
between the effect of the cream on the coffee and the effect of the
young lady on the man's life, the cream is not similar to anything
relevant and is not an icon. For a further and more obvious example,
consider H. G. Adams' line:

Hell is truth seen too late.
Here there is no similarity upon which this line turns. There are
several relationships between truth and hell, depending on what kind of
truth or hell one was discussing. Two interpretations based on such
relationships are:

(1) Hell (damnation) is the consequent of recognizing the truth
about certain religious matters too late.

(2) It is an extremely excruciating experience to learn the
truth when it is too late to act upon it.

One could argue as to which of these is better or whether either is
the best possible interpretation. Both seem reasonable interpretations,
yet neither is based upon a similarity between two objects. (Hell may
be an excruciating experience. But if this were claimed as the ground
of similarity, then one would not talk of icons since icons are not
icons of themselves.)

While it is important to maintain that there is some relationship
between the literal sense of a term and the figurative sense of that
term, it is not the case that, for every metaphor, the referent (if any)

determined by the literal sense is similar to the referent (if any)

118

determined by the figurative sense. 'To this extent it seems that the
discussion of icons tends to mislead one. While the similarity among
objects may be the relationship in the case of some metaphors it can not
serve as a general explanation of metaphor.
In discussing the significance of metaphor, Henle provides answers
for a question proposed by Black.
If we now ask why, on this view, the writer should set his readers
the task of solving a puzzle, we shall be offered two types of
answer. The first is that there may, in fact, be no literal
equivalent ... (and the second is) ... the pleasure it gives the
reader.
These answers which Black finds unacceptable are just those given by
Henle in his discussion of the cognitive and emotive function of metaphor.
With regard to the cognitive function of metaphor Henle states:

The function of metaphor in general is to extend language, to
say what cannot be said in terms of literal meanings alone.

Black recognized this view of metaphor as the catachresis view and held
the view to be correct in that it explained one type ofmetaphor. But

as we observed, Black misjudged the importance of this view and dismissed
serious consideration of it on the ground that such metaphors would be
short-lived and readily adopted as a part of literal speech. Henle, by
comparison, provides a more extensive discussion of the ability of meta—
phor to extend the usefulness of language. We find, however, that
Henle's discussion is misdirected in that he attempts to explain this
feature of metaphor wholly in terms of reference while failing to consider
meaning. His focusing upon reference induces him to distinguish two
"uses“ of metaphor on the basis of the existence or nonexistence of an

objective reference.

119

Henle describes the first use of metaphor, that involving objec-
tive reference as follows:

Metaphor may be used whenever something new is invented requiring

a name or whenever it seems desirable to call attention to an

undesignated aspect of something already known.29
In this use of metaphor, according to Henle, the icon is sufficiently
similar to the new object that one is able to recognize that to which
reference is figuratively made. It can be seen that the use of metaphor
here described could be eliminated from language or at least avoided and
is only a matter of conveneince (or perhaps a device to please the audi-
ence). One could use ostensive definitions and make reference with his
newly defined term. Consider Henle's two examples: "Thus the lower
shell of a turtle is called a plastron, which originally meant a breast-
plate, as in a suit of armor" and "hood of a car". In both, ostensive
definitions could be made and the metaphoric use of terms could be
avoided. In any event there are (or were) such metaphors in language
and these metaphors will (or have) become part of the literal use of
language.

Once the required part was recognized, however, the clash may be

forgotten and the phrase might be considered a single unit, desig-

nating something whose appearance is known. Thus the metaphor

would disappear and a new literal sense be born.3

Henle finds more difficulty in discussing the other 'use' of meta-
phor. That use does not involve an objective relationship. Specifically,
he finds difficulty in explaining the transition from metaphor to literal
speech because there is no objective referent with which one becomes

familiar. He says of the example 'cosmos':

. the term did not remain metaphorical, but engendered a literal
meaning. The process used to explain earlier examples will not

120

serve here, however, since there can be no recognition or identi-
fication of a cosmos in the same sense that a hood of a car can
be recognized. We know of no satisfactory theory as to how such
terms became literal...."31
It seems to be Henle's focus on reference which leads him to conclude
that he knows of no satisfactory theory.... But in examining his earlier
remarks we find that the clash of meanings separated the literal from
the metaphoric and not any considerations of reference. We recall that
in discussing the clash of literal meaning, Henle explained that if no
clash is felt the metaphor may be dead. He said that in such a case
“. . the figurative meaning is so usual that the reader goes to it immer
diately. In this case one has an idiom or a "dead metaphor" which,

Pr0P9r1y. is no metaphor at all."32

The basis for distinguishing meta-
phor from non-metaphor is whether or not the figurative meaning is
usual. An explanation of the expansion of literal use through metaphor
is suggested by these remarks. That suggestion is: as a linguistic
community becomes accustomed to associating an expression with a mean-
ing, that meaning is thought of without the special provocation of a
clash of meaning. The becoming accustomed to meaning is essentially
establishing a customary use of the word. Thus when a word's use is so
usual that one thinks of it immediately it no longer causes a clash,
that use will then be a conventional use and part of literal speech. If
the further question is asked 'How do certain metaphoric uses become
usual?‘ the answer is simple: by repeated use. What is significant is
that a type of use, through repetition becomes one of several ambiguous

literal uses. In such a case language has expanded. It is the immediate

recognition of the sense in question which separates metaphor from literal

121

use. Objective reference needs to play no part in this explanation.

We see that the use of metaphors which do not involve an objective
reference are of greater concern, since there is no easy device, such
as ostensive definition, by which to avoid the metaphor or to fix so
much of its meaning as is contained in reference. If we are to express
the same meaning it may not be possible to do so by the literal use of
any expression.

In cases of this sort, however, metaphor is even more important

for extended language than in those considered before. After all,

the underside of a turtle is simply there and can be pointed at if
necessary to call attention to it. A world order may equally be
there, but it certainly can not be pointed at in the same way.
31232.1’1'5‘222512fi3222:1233.§i2“§pl§t’“$¥ofiem§$§p’£§§?5’3"' Perhaps
It is therefore of great importance to abandon the explanation in terms
of reference and recognize that words are assimilated into literal
language on the basis of repeated use and common understanding.

The final aspect of Henle's work which we now discuss is his notion
of induced content. This induced content is particularly important
when the metaphor is not concerned with objective reference. Consider
the following metaphor:

The mind is the place where thinking occurs.

One might be tempted to search for the location of the mind since all'
places have location. Or perhaps the following may seem more familiar:

The mind is spiritual substance.

Here the word substance may induce one to attribute other qualities to

the mind, such being subject to physical laws, which may not be intended

by the expression. The induced content is particularly dangerous as the

122

metaphoric use of the expression becomes customary. Once the literal
clash, to use Henle's expression, is no longer perceived, one is apt to
forget metaphoric origin of the expression and attribute characteristics
appropriate only to its former literal sense. One would not make such
attribution while the metaphoric use of the term was new. People are
careful about metaphor. To refer to our example, the mind becomes a
place and a substance as that use becomes conventional. One may then
forget that this is a different use of 'place' and 'substance'. Recog-
nizing the difficulties which may be caused by induced content we look
to Henle's discussion for help in avoiding such difficulties.

Our main interest is in the cognitive aspect of induced content.
Henle's discussion centers around the emotive aspect. Nonetheless, he is
careful in his explanation to allow for induced cognitive content. His
remarks are addressed to poetry but the same effects result in non-poetic
use.

The literary or poetic use of metaphor is also creative, but in a

different way, bringing out a new aspect, or showing a new way of

feeling concerning something already describable in the language.34
Henle here notices a feature which Black overlooks. That feature is the
conveying a new feeling (and not necessarily a pleasing feeling) concern-
ing the topic of the metaphor. We agree with the position that philosophy
has something more to do than give pleasure to its readers. Nonetheless,
giving pleasure to the reader is often our intent when using language in
other situations and is therefore often a reason to use metaphor. This
feature of metaphor deserves a more detailed treatment than was provided

by Black. According to Henle:

123

This nuance of feeling may well be the reason for the employment

of the figure, for in many cases there does not seem to be any

other way of conveying just the same impression.35

Henle speaks of an antecedent similarity and an induced similarity.
The antecedent similarity is that which constitutes the iconic relation«
ship. It is the basis of the metaphor. Given such an antecedent
similarity one can evoke by the use of metaphor an induced similarity.
". . an induced similarity, such as the similarity of feeling which, as
we have just noticed is transferred from symbolizing situation to situ—
ation symbolized."36 It is unfortunate that Henle speaks of the
similarity being transferred; it is not of course. ‘The induced similare
ity is a similarity between our feeling (in the case just mentioned)
towards one object and our feeling towards another object. What is in«

duced is the feeling towards 'the situation symbolized', that feeling

being similar to our feeling toward the 'symbolizing situation.‘ When

 

 

the feeling has been induced the similarityi§4established. Henle
states:

... the feeling accompanying the figurative sense is modified by
and in fact caused by the feeling accompanying the literal sense.
This modification of the way of thinking of what the metaphor
symbolizes indirectly may be called the induCed content of the
metaphor.37

 

Henle does not, however, discuss reasons for this induced content nor
does he explain how the content is induced. Essentially his discussion
is a notation that metaphor may involve induced content and that such
content is of significant poetic value.

Henle's discussion involved only examples of induced emotive content.

He points out that his definition of induced content does not restrict

124

such content to feeling. But due to Henle's view of poetry and his con-
centration on poetry he gives little consideration to cognitive content.
He maintains: "such contents, if they exist, are of relatively minor
importance in poetic use of metaphor and so have not been discussed
here".38 Henle does however indicate a grasp of the importance of in-
duced content in regard to metaphor's use to extend language.
In some cases, we noticed that metaphor acquired literal meaning
by the development of the ability to recognize what it designated;
in other cases this explanation could not be given, Many philo-
sophical terms are of this latter sort and it is difficult if not
impossible to develop means of recognition for such terms as
'substance' or 'ego'. It may be in such cases that part of the
meaning is given by an induced content.39
Henle, however, does not pursue this explanation of the explanation fur-
ther. His rather vague notion of induced content would make speculation

tenuous at best.

Conclusions in Regard to Henle's View

Henle's reference to icons and the ensuing explanation provide a
prime example of the relational view; that is he characterizes metaphor
as referring to an object and providing the relationship between that
object and the object which the author intended to discuss. Under a
literal interpretation the sentence is construed as mentioning some
object; the relationship is understood to be a matter of similarity: the
audience recognizes the true referent of the metaphor. Two serious
problems face the iconic theory. First, as it stands, it fails to explain
our recognition that something is a metaphor and without that recognition
we would not invoke the similarity as discussed. Second, it fails to

explain what similarities we are to consider.

125

In correcting these deficiencies, Henle suggests the outline of a
theory of metaphor which does not depend on the resemblance of objects.
Neither does it depend on his explanation in terms of icons. It is
rather an independent theory based upon the following criteria. Metaphor
involves a clash when interpreted literally; this is our signal to
attempt a figurative interpretation. We assume that there is a meaninge
ful interpretation on the principle that people try to make sense. We
attempt to find a relationship either between the words used in the meta-
phor and other words, between meanings associated with literal use and
other meanings, or the referent associated with the literal use and some
other referent. Our guide in attempting to discover such a relationship
will be the characteristics uppermost in people's mind in regard to those
words, meanings, or referents. When a relationship is determined, we
then consider the referent associated with the related word, related
meaning, or the related referent itself as the subject of the metaphor.
When a meaningful interpretation is established, one understands the
metaphor.

Notice that such an explanation does not depend upon finding a
similarity or even a relationship between compared objects and events.

We indicated above that Henle suggests such a theory since his concern
with reference did not allow him to escape a dependence on comparisons
of objects. Had he made good such an escape he could have provided a

valuable theory.

In his discussion of metaphors' function in the expansion of lan-

guage, Henle is once again hampered by his concern with reference.

126

This concern caused difficulty in explaining how metaphors without oba
jective reference become literai uses; 'His difficulty is surprising
since earlier in his article he explains this as being a matter of
immediate recognition of meaning due to repeated use. But he over-
looked the fact that it is the recognition of meaning which provides the
general explanation, not familiarity with a particular referent.

While Henle's view, as he expressed it is a relational view, those
aspects which make it so tend only to cause difficulty in explaining his
positive points. By centering his attention on a sense rather than
reference he could avoid these difficulties. And as we noted, by elimi-
nating his dependence on reference and the iconic model one can establish

a reasonable explanation of several aspects of metaphor.

Conclusions with Regard to Comparison and
Relational Theories

The comparison, as a basis of explaining metaphor is particularly
inviting. The words in their literal use do not mean the same as those
same words in their metaphorical use. But there is a relationship between
these meanings and often this relationship is a similarity between the
referents which are associated with those meanings. There is no doubt
that many metaphors are at least vague comparisons. But it is not clear
that all metaphors are comparisons and not all metaphors involve the
similarity between referents just mentioned. And even for those meta—
phors which are comparisons, the noting of the comparison does not explain
our understanding of the metaphor. In fact we must often understand that

the expression is a metaphor prior to recognizing that a comparison is

127

being made.

One is often tempted to show that every metaphor is after all a
comparison by constructing as reasonable a paraphrase of each as he is
able and then claiming that paraphrase to express the meaning of the
metaphor. But as we discussed the line of James Whitcomb Riley from
"Out to Old Aunt Mary's",

While the waters gurgled and laughed and wept."
we noticed that one would miss the point if be interpreted this line as
a comparison (perhaps of sounds). The line makes sense in the poem only
when one recognizes that the waters were weeping and laughing from
happiness. A report of how the water sounded would be completely out of
place. Isenberg sums up our point here.

And by introducing such words as 'like' into certain metaphoric

expressions, as if misquoting: "All the world is like a stage,"

we obtain—«whatever the impairment of the general effect-»a simile.

It is, no doubt, the possibility of such conversions that has led

many rhetoricians to say that a metaphor is a condensed simile and

a simile an expanded metaphor. Now this point I think is to be

rejected; for there are metaphors, even simple ones, that cannot

be thus converted without circumlocutions so devious as to make 8he

result very distant in meaning and effect from‘theiroriginals.4
One might, if he were willing to sacrifice meaning and effect, construe
all metaphor as comparison. But such a move would not be instructive with
regard to metaphor since it could as well be made with any literal expres—
sion (given that one was willing to accept some sacrifice of meaning and
effect). One could, for example rephrase

This house is a white house

to read

This house is like a white house.

128

Or perhaps one might similarly rephrase

The man moved swiftly
to read

The man moved as a swift man moves.

We agree that these restatements involve some change in meaning, but

argue that converting the metaphor to simile involves a similar change.
Therefore showing it is possible to produce reasonably correct paraphrases
which are explicit comparisons is not to distinguish literal from meta-
phoric use. Further, since as Isenberg and our own example indicate that
the result of paraphrase may be "very_distant in meaning and effect from
their originals", the claim that metaphors are comparisons is certainly
misleading.

Comparisons and relational views tend to be misleading in another
way as well. They argue that, or at least lead one to believe that,
metaphors involve the comparison or relationship of objects. Discussions
of the comparison and relational views are generally built around examples
in which either the subject or the direct object of the sentence is being
compared or in which the subject is compared to the direct object. For
example, Richards compares the leg in 'leg of a table' with the leg of a
horse. Black compares man and wolf in 'man is a wolf'. Not only the
examples from Richards and Black lead to comparison and relational views,
they concentrate attention on those grammatical elements which are in
the nominative or objective case and the referents of those elements.
This concentration results in an inattention to verbal and adjectival

elements of the sentence as well as inattention to relationships other

129

than those holding between objects. Inattention to verbal and adjectival
elements results in the failure to explain metaphors which have as their
focus the predicate (a verb form). Such metaphors are not explained
since the focus is not ordinarily construed as referring to an object or
even to a type of object. An example of such a metaphor is provided by
Black,

The chairman ploughed through the discussion.
In such cases there is no object which is related to another.

Concentration upon the relationships between objects results in
difficulty in explaining those metaphors whose basis is a relationship
between meanings and not a matter of comparisons at all. In particular,
we here think of those metaphors with which we found difficulty in explan-
ing their literalization.

The views of Black, Henle, and Richards all suffer to some degree
from this difficulty. We do find that the promising suggestions of

these theories can be stated without reference to comparisons.

12.
13.
14.
15.
16.
17.
18.
19.
20.

Noam-boom

130

NOTES TO CHAPTER IV

. Paul Henle, 'Metaphor', Language, Thought, and Culture, ed. Paul Henle,

University of Michigan Press, 1958.

This discussion takes place in Chapter VI of this paper.

. Ibid., 174.

Ibid., 175.

Ibid., 177.

bi

H

O
Q

. S. Peirce, Collected Papers, Hartshorne and Weiss, Eds., Cambridge,
Massachusetts, 1932, Vol. II, p. 247.

. Henle, 178.
. ijg,, 179.
10.
11.

Ibid., 178.

This problem is discussed on page . Henle provides an interesting
example of the difficulty by noting that "... the number seven-
teen is similar to an elephant in that they share the character-
istics of being different from the moon." Ibid., 179.

H
U.
...:

D.

H
U'
do
O.
U

186.

H
U.
....

O.

0—!
U"
—I

O.

H
U’
...
O.
U

180.

H
U
..I

G.

H
on
do
Q.

, 180-181.

0—.
0'
—lo
Q

U

175.

H
U'
—l

O.

., 175-176.

21.
22.
23.
24.
25.
26.

27.
28.
29.
30.

31.
32.
33.
34.
35.
36.
37.
38.
39.
40.

131

H
U-
do
a.
U

182.

O

H
CT
—I

O.

H
U'
—l

D.

, 183.
, 183-184.

H
O.

bi
bi

H
CL

., 183.

Richards also presents a criticism against too great a reliance on
similarity. We consider this in Chapter II.

Black, 280-281.
Henle, l86.
Ibid., l86—187.

Ibid., l88. We recognize that talk of disappearing meta hors and
birth of literal senses will raise questions as to w at disappears
and what is it for a sense to be born. We shall forego the track—
ing down of these loose ends since they could have been avoided
by a more careful statement“at the beginning.

Ib d., 188-189.

a—lo

H
U'
—l

D.

., 183.

H
U'
_l

D.

., 189.

H
U'
.a.
O.

H
U'
A.
D.

u

191.

L

H
U'
._n

O.

H
U"
—I0
0..
.

192.

lb

bi
bi

H

D.

D—i
0.

Arnold Isenberg, "0n Defining Metaphor", The Journal ofPhilosophyXXI
(October 1963), pp. 609-622.

 

CHAPTER V

ABSURDITY VIEWS

One of the features of metaphor which we noticed in the introduction
was the obvious absurdity which results from the literal interpretation
of metaphor. That absurdity resulted either through some conflict of
meanings within the expression itself or through a conflict between the
expression and its context. We shall call those views whose explanation
centers around the absurdity of literal interpretation 'absurdity views.‘
Both Beardsley1 and Turbayne2 hold such views. Both hold that recogni-
tion of a metaphor depends, in part at least, upon recognizing the inap-
propriateness of the literal interpretation. While the explanations of
Beardsley and Turbayne are not in direct contradiction they do constitute
rather divergent explanations. That divergence apparently stems from a
difference in the aspect of metaphor which seems most significant.
Beardsley is most interested in explaining how we understand metaphor
while Turbayne is most concerned with the results of overlooking the
absurdity of literal interpretation.

Beardsley's controversion (as he calls it) theory of metaphor is
based upon the contention that one's ability to recognize and understand
metaphor is based upon recognition of the absurdity of the literal expres-

tion. We recognize in the previous chapter that Henle considered this

feature of metaphor. However, it did not constitute a central aspect of

l32

133

the theory presented by Henle. We have chosen Henle's work as representa-
tive of the absurdity view, since the absurdity of literal interpretation
is central to his view. In addition, Beardsley's discussion provides a
basis for the discussion of various other significant questions. His
criticism of types of views which he calls the Emotive, Supervenience,

and Literalist theories treat in particular the claims:

(1) Metaphors have no meaning.

(2) The meaning of no metaphor is paraphraseable in literal terms.

(3) Metaphors are literal statements.

While we find that Beardsley's controversion theory makes little improve-
ment on certain observations of Henle and Black, his treatment of the
above claims makes his effort well worth considering.

Turbayne appears less interested in our understanding of metaphors
than in a particular sort of misunderstanding that arises from metaphor.
His concern is with the result of failing to be aware of the absurdity of
the literal interpretation after the expression has been accepted (as
true) on the basis of a metaphorical interpretation. Turbayne uses
terminology which leads one to suspect that he finds a moral significance
to a blunder of this nature. He says:

Metaphors furnish a clear illustration of the "imposture" or the

"delusion" or of the "cheat" of words. It is part of the nature

of metaphor to appear wearing a disguise. This disguise is often

so effective that not even automatic brains can penetrate it.
Turbayne explains that metaphor is always in some sense inappropriate
and that the use of metaphor involves make—believe. He says that one is
cheated by or used by the metaphor when he fails to recognize the make-
believe. It seems Turbayne's purpose is to warn us of these pitfalls

which he explains.

l34

While we find that Turbayne's warning is in order (the history of
philosophy is replete with such errors), we find that his explanation
is incorrect. It is incorrect since he claims that the use of metaphor
involves pretending and make-believe and it is clear that using metaphor
requires neither.

Both these examples of the absurdity are important since they deal
in some depth with aspects ignored elsewhere. We shall first consider

Beardsley's controversion theory.

Monroe C. Beardsley

Beardsley presents a discussion of what he terms a controversion
theory of metaphor. The view is mainly an investigation of the point made
by Henle: that the clash of literal meanings provides the basis for our
search for a figurative sense. Beardsley goes into considerably greater
detail in his attempt to deal with this aspect of metaphor. Prior to the
introduction of his own view Beardsley proposes to discuss three other
views, which he calls the emotive theory, the literalist theory and the
supervenience theory. Beardsley indicates that it is hard to find any
clear-cut defenders of these views.

These theories suffer somewhat from the same difficulty as those
views commented upon by Black. That difficulty, which Beardsley recog-
nizes, as mentioned above, is that these views are not associated speci-
fically with any advocate. They suffer from skeleton presentation and
skeleton argument. Further, there is no original source against which
to check Beardsley's presentation and arguments. Each of the theories

presented by Beardsley is based upon some aspect of metaphor. In each

l35

case there are examples which tend to support the claims of that view.

He discusses central issues of metaphor any one of which might be focused
upon to establish a fully defended view of metaphor. We are interested,

however, in the points discussed whether or not anyone desires to defend

those theories and will consider Beardsley's analysis of each.

Emotive Theory

The central feature stressed in Beardsley's discussion of the emo-
tive theory is the meaningfulness of metaphor. According to Beardsley,
the central claim of the emotive theory is that metaphors have no meaning.
Beardsley characterizes this theory as follows:

. according to the Emotive Theory, a word has meaning only if
there is some way of confirming its applicability to a given situa-
tion--roughly, only if it has a clear designation. For example,
the sharpness of a knife can be tested by various means, so that
the phrase "sharp knife" is meaningful. We may suppose that sharp
has some negative emotive import, deriving from our experience with
sharp things. Now when we speak of a "sharp razor' or a "sharp
drill" the emotive import is not active, because these phrases are
meaningful. But when we speak of a "sharp wind," a "sharp dealer,"
or a “sharp tongue," the tests for sharpness cannot be applied, and
therefore, though the individual words are meaningful, the combina-
tions of them are not. In this way the emotive import of the
adjective is released and intensified.4

Beardsley claims that such emotive theories are clearly wrong on two
counts. They are wrong because they: (l) construe testability and
therefore meaning too narrowly and (2) claim that all metaphors are emo—
tive. In considering Beardsley's argument we become aware of the sig-
nificance of the difficulty in his method of presentation which was dis-
cussed on the previous page. We should like to see an argument presented
which would settle the question of the correctness of the emotive theory.

Beardsley's argument is not, however, against any principle entailed by

136

the emotive theory but is rather against the example which he chooses
to assist in the presentation of the theory. Even if his argument is
conclusive against the example it does not defeat the emotive theory
but only the choice of examples. These circumstances suggest that a
person who chose to defend this view could have selected a more appro-
priate example. After considering Beardsley's argument we shall present
an example which his line of reasoning will not defeat.
Beardsley's argument is:
. it is not correct to say that there is no test for sharp
tongues. Anyone who often scolds and finds fault sarcastically
can be said to have a sharp tongue; these are the tests. Of course,
the tests for sharp tongues are not the same as the tests for sharp
razors, but the tests for sharg razors are not the same as the
tests for sharp drills either.
In considering this argument we notice that Beardsley has chosen to
attack the view presented on its home ground. That is, for the purposes
of his argument he has accepted testability as the criterion of meaning.
He might have rejected this view by insisting on and arguing for his own
position with regard to meaning. He says, "For what the sentence means
depends ... upon public conventions of usage that are tied up with habit

6 This move would result in

patterns in the whole speaking community."
talk about conventions rather than talk about tests. For the extent of
his discussion this will make no difference. The same arguments would
require only minor restatement. We argue that the talk in terms of tests
must reduce talk about conventions if the emotive argument is to succeed.
In returning to Beardsley's argument, we find a good point as well

as an error common to many theories of metaphor. The good point lies in

noticing that there are several different tests one might perform in

137

checking the applicability of a word. The test for sharp butcher knives
is not the same as the test for sharp masonry drills. The fact that a
particular test fails to apply does not show that a phrase is without
meaning. 'Sharp masonry drills' is not meaningless because its tests for
applicability differs from the test for the applicability of 'sharp
knife.‘ On a testability theory of meaning the question of importance

is whether agy_test is applicable.

The common error found in Beardsley's discussion is the use of a
dead metaphor as the basis of the discussion. The expression 'sharp
tongue' is recognized as employing 'sharp' in a literal manner; 'sharp'
out of context is ambiguous. 'Sharp tongue' will have a test because it
is a part of our literal vocabulary. Any reasonable defender of the
emotive theory presented would claim that whatever its origin (perhaps it
once was a metaphor), 'sharp tongue' is not metaphor. Perhaps the emo-
tive theorist would reply, "Beardsley ought to consider 'sedulous quarter-
staff;' here is a better example of metaphor; we have no tests for the
sedulousness of quarter-staffs." But before a reply is made to this, we
find that we must consider what is meant by 'test' or by 'confirming its
applicability.I

The tests proposed by the emotivist must be tests 'in principle.‘

If one argues that the means to carry out the tests must be at hand, then
a great many literal statements become meaningless which had never been
considered meaningless before. A more important question is whether the
relevant test must be known, generally or to anyone. If the test must
be recognized generally within a linguistic community then the criterion

of testability entails a conventional meaning. Given this explanation of

138

testability it is clear that some metaphors lack meaning since they are
a departure from customary use and are not therefore associated with a

generally recognized test. But to conclude from this explanation that

such metaphors are without meaning is to conclude that the only meaning
is conventional meaning. In dealing with metaphor such a conclusion is
clearly mistaken. We offer two arguments to show that not all meaning

is a matter of convention.

Our first argument is based on the consideration of the dead meta-
phor. As we noticed earlier, such metaphors have meaning. 0n the
assumption that the metaphor was without meaning in its first utterance
one is left with the awkward task of explaining when metaphors acquire
their meaning. Various sorts of answers to this question arise. Among
them are: 'after the phrase is used in a particular way for, say, 1000

times,‘ after that use is entered in the current edition of the Oxford

 

English Dictionary. Any answer to the first sort is arbitrary. Answers
of the second sort overlook the fact that lexicographers record uses, not
prescribe them. Difficulties in explaining when a metaphor acquired its
meaning can be avoided if one allows that at its first utterance within
an appropriate context a metaphor has meaning. As such meanings are
more widely recognized they are said to be more customary or customary
among a larger group. As such meanings are recorded in standard lexicon
they are said to be conventional. Therefore, conventional meaning is to
be explained in terms of expanding use of a term with an associated
meaning. If the term had no meaning there would be nothing to become

conventional.' Further if an expression were meaningless it would not be

139

repeated a sufficient number of times to become customary. We conclude
therefore that while metaphor goes beyond customary meaning it is not
meaningless.

Our second argument, in support of the view that metaphors are mean-
ingful, has a more firm base. One fact about metaphor has been over-
looked by Beardsley and our discussion to this point. Often the members
within a linguistic community generally understand a metaphor in the same
way. That is, they independently arrive upon a sense of the metaphor
in which they are in agreement with one another. Further, this sense is
not solely a matter of convention and the use of the expression in the
given sort of context is unique. In these situations there is common
understanding and communication. These are sufficient grounds for claim-
ing that the metaphor has meaning, even though it lacks conventional mean-
ing. Given that the metaphor has meaning, it can, as argued above, come
to have a conventional meaning as its use becomes customary. The discus-
sion of relevant tests is easily assimilated to this discussion by noting
that as a linguistic community apprehends a common sense, its members
are able to determine appropriate tests of the terms applicability.

We are now in position to deal with the proposed answer of the emo-
tivist concerning 'sedulous quarter-staff.' As we argue in our discus-
sion of Black and Henle, to constitute a metaphor this phrase must be
accompanied by some context. Depending on context, it may indeed be
meaningless. In those contexts in which the expression is meaningful to
someone relevant tests could be described by that person. In those con-

texts where the expression is meaningful to no one, that use of the

140

expression is not metaphoric. The emotivist will be faced with either
admitting that the phrase has appropriate tests or explaining how some
nonsense expressions evoke emotive responses and others do not. It would
seem that a meaningless expression would evoke only puzzlement.

Returning to Beardsley, we recall that his second objection to the
emotive theory is that not all metaphors are emotive. This argument is
supported by several theories which claim that at least on some occasions
the value of metaphor lies in its use to communicate information where
literal expression was lacking. We have no quarrel with this position.

We are in agreement with Beardsley‘s claim that'the“emotive position
is incorrect due to a too narrOW'construal"of'testability."Dur'objection
to Beardsley's argument was mainly that it only challenged an example
and not the theory. The argument which defeats the emotive view is not
all meanings or tests are conventional meanings or conventional tests.

To claim otherwise is to claim that communication is possible, using mean-

ingless expressions.

Supervenience Theory
The emotive theory as described by Beardsley is based on the claim
that metaphors have no meaning. The supervenience theory admits that
metaphors have meaning but claim that this meaning isnot expressible in
literal speech. Beardsley provides the following characterization:
The defense of this theory begins with the observation that poetic
language, and metaphor in particular, is capable*of‘conveying meane
ings that literal language cannot convey. Why else, indeed, would
we have need of metaphor, unless it supplied a mode of speech for

which there is no substitute? ... According to this theory, the mean-
ing of a metaphor does not grow out of the literal meanings of its

141

parts, but appears as something extraneous to, and independent of
them.... The Supervenience Theory regards a metaphor as a species
of idiom ... it denies that metaphors can be explicated.7

Beardsley attacks this position by arguing against the claim that no
literal paraphrase can convey the meaning of a metaphor. His first
argument is that the theory conflicts with fact, namely that words
which once were metaphors are now conventional and yet express the same
meanings which they did as metaphors. He states:

. the Intuitionist [a person holding the supervenience view] also
has some more specific reasons against the possibility of para-
phrase. First, the resort to metaphor in ordinary language as well
as poetry shows that we need it to express meanings that we have no
literal language for, and therefore any metaphor outruns the re-
sources of literal language. Now, as a generic account of metaphor,
and also as a practical justification of it, this objection is un-
doubtedly sound. But it does not show that the cognitive content
of the metaphor is such that no literal expression can ever desig-
nate it. And in fact we see all the time that, when metaphors are
introduced to supplement existing literal language, especially in
the history of science ("Nature abhors a vacuum," ”natural selection,"
the Freudian "censor") where an important characteristic is singled
out by the metaphor and for purposes of knowledge we want to refer
to it again, literal language takes over. Either the metaphor is
killed, or a technical neologism is invented and assigned that
designation.

We agree with Beardsley that the fact that some metaphoric expressions
have become literal expressions without change in meaning shows that it
is not impossible for a literal expression to have the same meaning as a
metaphoric expression has had. Nonetheless, a reasonable rebuttal to
this argument is available to the supervenience theorist. He might claim
that his theory was to apply to the current situation of language. He
could recognize that as languages change different syntactical groupings
will be paraphrased differently and some not paraphrasable may become so.

Similarly, some expressions which were once metaphors will cease to be

142

metaphors. But, he could maintain, it will never be the case that a
metaphor can be translated accurately into literal speech even though
some former metaphors are now a part of literal speech. This rebutting
of Beardsley's argument would be successful.

The remainder of Beardsley's discussion of the supervenience theory
consists of criticisms of arguments which tend to support that view.
This first such criticism is directed against the argument that any
accurate paraphrase of a metaphor must include a metaphor. The strategy
of Beardsley is to introduce a metaphor and an accurate paraphrase which
does not make use of metaphor. That is, he argues by counter example.

"Napoleon is a wolf" means, among other things, "Napoleon is

cruel, and the cruelty of Napoleon is like the cruelty of a wolf

..."--here the word "wolf" is used literally, not metaphorically.
This form of criticism suffers from an inherent difficulty; that is,
showing that ones counter example is in fact a counter example. Beardsley
has offered a purported counter example but fails to prove that it is one.
Consider the following reply from a supposed intuitionist: "Beardsley
fails to produce a complete paraphrase (as is indicated by the ellipses);
further, should he attempt to complete his paraphrase he will still have
either an incomplete paraphrase or will say too much." 'We of course see
little value in this reply to Beardsley, but since Beardsley provides no
argument for the completeness of his paraphrase he is reduced to begging
the question, "Don't you agree that this is a counter example?".

Even if Beardsley's counter example argument holds, it at best de-
feats only the stronger of two supervenience views. The stronger of the

two views is: No metaphor can be paraphrased without the use of further

143

metaphor. The weaker view: A great many metaphors cannot be paraphrased
without the use of further metaphorL‘ ThiS‘weaker view is not rebuttable
by a counter example of its opponent‘s choosing (in this case Beardsley's
choosing). The only argument by counter example which would be signifi-
cant would use an example which was claimed to be untranslatable in the
manner prescribed. (Even with such examples, one's argument would be
valid against the weaker claim only if every such example was translatable
without metaphor.) The weaker version seems indeed to avoid Beardsley's
argument and we find no reason to reject it. A second argument against
paraphrase considered by Beardsley is:
. it must be admitted that there is an important difference be-

tween a predication that is merely suggested and one that is

overtly stated, a difference that explains why we sometimes prefer

to suggest rather than state, and it is lost in paraphrase. It is

a difference in what I have called predicative weight, the differ-

ence between insisting and hinting, saying with force or just

barely calling to someone's attention. An overt statement is bound

to have more predicative weight than a suggestion, however obvious,
and this may be misleading in the paraphrase.

 

Beardsley responds to this argument by noting that one can include in the
paraphrase, perhaps as parenthetical remarks, indications of the predica-
tive weight to be understood while it is the case that paraphrases gener-
ally do not carry such indications nothing prohibits them from doing so.

The final argument for the supervenience view which Beardsley criti-
cizes is the argument that the paraphrase of a poem would not be a poem.
Beardsley correctly indicates that this argument fails to establish the
point at issue, namely, that there is no synonymous literal paraphrase.
It may be suggested that somehow the discovery of such a paraphrase in

some way diminishes the value of the original poem. While it may be the

144

case that the paraphrase may lack some aesthetic values it may well help
understand the sense of the original and increase one's appreciation of
the original poem.

Beardsley offers the following summation with regard to the question
of paraphrase and literal equivalents.

It does seem that there is no conclusive argument against the pos«

sibility, in principle, of paraphrase; but there are excellent

reasons for supposing it to be usually impossible in practice.

The most that we can even give is a partial paraphrase of a poem,

and no matter how elaborate the paraphrase, we shall nearly always

feel that something is left out, if the poem is complex. But we
note that the reason for this feeling may often be that we are not
sure that we have understood all the meaning of the poem, not that
we have understood some meaning which we cannot make explicit....

But even if we can seldom in practice, paraphrase all_of a poem

we can nevertheless paraphrase agy_of it we wish to paraphrase.1
While this conclusion is comforting, if one accepts it, it is mistaken
in two ways. First, it suggests that when we have doubts about a para—
phrase, those doubts stem from a lack of understanding of the metaphor
and not from an inability to make explicit some meaning which is under-
stood. (While Beardsley says only this "may often" be the case, context
indicates that he believes this to generally be the case.) Second, the
passage states that we can paraphrase any of“a poem that‘we wish to
paraphrase.

The first mistake appears based upon a belief that one can parav
phrase any expression which he understands. We should like to see an
argument for this belief. 'In literal speech we often understand an
expression but are unable to find other”wordswhichexpress the same
meaning. There is no reason to assume that one can provide a paraphrase,

in every case, when one cannot do so with literal speech even when he

145

understands the expression. Unless Beardsley resorts to circular reason-
ing by defining 'understand' so as to entail ability to paraphrase it

is often the case that we understand linguistic expressions but are
unable to explicitly paraphrase them. This is not a peculiarity of
metaphor, but a situation of language in general. Beardsley's second
error lies in the assumption that we only wish to paraphrase that which
we can understand. But what we often wish to paraphrase is that which we
do not understand. It is our lack of understanding which motivates our
desire for the paraphrase. We want to eliminate the feeling that some-
thing is left out. Often it is the part of the metaphor's meaning which
we fail to understand that we hope to elucidate by paraphrase.

What Beardsley ought to conclude from his arguments is that no
aspect of metaphor as far as we can tell makes it unparaphrasable. It
can be paraphrased as well as many vague and ambiguous terms in a literal
use. None of the arguments of the supervenience theorist sustain the
claim that every metaphor is not paraphrasable or even the weaker claim
that some metaphors cannot be paraphrased.

A point at which to criticize the supervenience theory would be
that no argument is presented to the effect that every metaphor is

unparaphrasable.

Literalist Theory
The literalist theory as described by Beardsley is based on a claim
nearly the opposite of the supervenience theory. The literalist theory

claims that metaphors are literal expressions. They are eliptical

146

similes. While this claim is reminiscent of Black's characterization

of the comparison view in that eliptical simile is mentioned, it differs
significantly in that (1) it makes no claim that every metaphor has a
literal equivalent and (2) it claims that metaphors are literal expres-
sions. Metaphor is similar to other literal eliptical expression in
that it is in need of no special interpretation.

. if nowadays people nearly always say, "She is the one I adore,"
it seems artificial to regard it as eliptical for I'She is the one
whom I adore"--they are not conscious of leaving out anything.

But there are sentences where we could say with some confidence that
certain words are implicit, though not uttered, because they are
needed to complete the sense by ruling out other possibilities.

"If he's going, I'm not" requires another "going" rather than
"staying," to be understood after "not." The Literalist Theory,
then holds that metaphors are eliptical similes. Since there is

no fundamental difference they can be understood the way similes
are, by the ordinary rgles of language, and present no special prob-
lems for explication.1

This theory is dependent upon the assumptions of the comparison theory,
namely, that all metaphors are comparisons or can be reduced to compari-
sons. Beardsley's objection to this theory makes use of a distinction
between two kinds of similes:

(l) An open simile simply states that x is like y--the minutes

are like the waves. (2) A closed simile states that x is like
y in such and such a respect.' In the lines

 

 

And custom lie upon thee with a weight,
Heavy as frost, and deep almost as life

(Wordsworth, "Ode: Intimations of Immortality from Recollections
of Early Childhood") custom and frost are compared in "heavyness,"
custom and life in "depth."1

Based upon this explanation Beardsley states his criticism:

Closed similes work like metaphors, and can be restated fairly

satisfactorily as metaphors. "Custom is frost heavy"--but not

all metaphors, I think can be satisfactorily restated as closed
similes--for example, in "The moon lies upon the straits," what
is it that the moon looks like?

147

This is the conclusion of our argument against the comparison view,
namely that not all metaphors are comparisons. But then Beardsley con—
tinues with an unusual claim.

But a metaphor cannot be reduced to an open simile, either, for
they work very differently in poetic contexts. The open simile is
empty and uncontrolled without a context: A is like B, but in
what relevant respects the context has to inform us. The metaphor
is full and rich, apart from any context; indeed, the function of
the context is rather to eliminate possible meanings rather than to
supply them. A metaphor is not an implied comparison.

His statement15

that context eliminates meanings is correct but the
statement that context fails to supply meanings is at least misleading.
We argue in the final chapter that it is only through the context that
some metaphors are meaningful. But here it will suffice to show that
his remark is in conflict with his position five pages later.

It must be admitted that we cannot know with certainty that a given

attribution is nonsensical, because someone may find a meaning in

it that we have overlooked, perhaps by inventing a line of poetry
in which the attribution works.

Thus for example, I once made up the expression "laminated
government" in class, on the spur of the moment, as an example of
nonsense. A student later’pointed out to me a very similar
expression in a book on government that gives it in context, a
definite meaning.

Beardsley is arguing that metaphor differs from nonsense in that it
makes sense and often it is an appropriate context that produces that
sense. This contradicts his position on l46 but is the better of the
two positions. Beardsley's best argument against the literalist position

is showing as he does that not all metaphors are comparisons.

Controversion Theory

Beardsley calls his own theory the controversion theory.

148

Its essential principle is that a speaker or writer utters a
statement explicitly bUt in such a way as to show that he does
not believe what he states, or is not primarily interested in
what he states, and thereby calls attention to something else
that he has not explicitly stated--“If he wins, I'll eat my hat."
It is discourse that says more than it states, by cancelling out
the primary meaning to make room for secondary meaning.

The reader, according to Beardsley, sees that the statement made is

absurd, yet something is presumably being asserted. "... he looks for

a second level of meaning on which something is being said."18
A bare self-contradiction is just that; but when the modifier con-
notes some characteristic that can be meaningfully attributed to
the subject, the reader jumps over the evident self-contradiction
and construes it indirectly, on the principle that the writer knows
that he is contradicting himself and wouldn't utter anything at all
unless he had something sensible in mind. Then the expression
becomes a significant self-contradiction.19

Beardsley's explanation is much like Henle's "general principle that

20 Beardsley's model for this explanation is

people try to talk sense."
what he calls an 'attribution.’ This he defines as:
. a linguistic expression containing at least two words, one of
which denotes a class and also characterizes it in some way, a9?
the other of which qualifies or modifies the characterization.
Examples of metaphor on this model are

The man is a fox
and

Metaphysical streets
In these examples 'man' and 'streets' are the subjects and 'fox' and
'metaphysiCal' are the modifiers. If these are metaphors, on Beardsley's
view, then there must be connotations of the modifiers which apply to
the subjects.

With this terminology Beardsley restates the controversion theory

as follows:

'149

a metaphor is a significant attribution that is either indirectly
self-contradictory or obviously false in its context, and in which
the modifier connotes characteristics that can be attributed truly
or falsely, to the subject.22
Beardsley characterizes the indirect self-contradiction as involving the
attribution of a characteristic to an object when that object's defini-
tion excludes that characteristic. Given his examples and talk about
connotations it is fair to assume that he would allow 'common knowledge'
to replace 'definition' in such a discussion.

As the connotations of the modifiers are more difficult to work out,
Beardsley claims, the metaphor is more obscure. But as long as there
are connotations the expression is a metaphor. Our understanding of
metaphor, on this theory, lies in working out the connotations or,
better, discovering which connotations may be predicated meaningfully of
the subject. The determination that an expression is a metaphor rather
than nonsense depends upon our success in finding meanings:

. even if we put all English adjectives in one hat, and all nouns
in the other, and drew them out at random we would find that the
strangest combinations yield possible meanings upon reflection....
Moreover it must be admitted that we cannot know with certainty
that a given attribution is nonsensical, because someone may find
a meaning in it that we have overlooked, perhaps by inventing a
line of poetry in which that attribution works.

As we discussed earlier, one's success in finding meanings will be de-
pendent, in part at least on the context of the expression.

Our treatment of the contraversion theory is brief. We see very
little improvement on the position extracted from Henle's work.24
The position is essentially correct. Some restatement would be required
to apply it to metaphors in which a subject term and attributive term

are not as obvious. For example the line of Coleridge in "The Rime of

150

the Ancient Mariner":

And every tongue, through utter drought,
Was withered at the root;

Here it appears there is no single subject or modifier as is suggested

by his discussion. If, however, the modifier is allowed to be more than

a single word the theory will be sufficiently adaptable. Since Beardsley
uses 'connotations' which seem to rule out nothing (except perhaps denota-
tion) he avoids the difficulty of talking about features of objects, or
features of meaning which have limited the application of other views

to a limited type of metaphor.

Conclusions

The controversion theory stresses a feature of metaphor which had
been discussed much before, namely that a literal interpretation leads
to contradiction or absurdity. In terms of positive suggestion it offers
no great improvement over the considerations of Black and Henle. However,
Beardsley's discussion of opposing views illuminates significant ques-
tions about metaphor. Those questions are whether: (1) metaphors are
meaningful, (2) whether their meaning is paraphrasable in literal terms,
and (3) whether the meaning of metaphor is literal. In discussing the
emotive theory of metaphor we concluded that metaphors have meaning (or
are meaningful) but that such meaning is not customary or conventional.
We noted that a metaphor is often meaningful in similar ways within a
speaking community and that within a community a metaphor can be used to
communicate. Based on this we claimed that metaphors are meaningful.

We would agree with Beardsley that if an expression could not be

151

meaningfully interpreted it is not a metaphor but rather nonsense. We
agree with Beardsley's claim that the fact that one test [or‘interpre«
tation] fails to apply does not render an expression meaningless.

In discussing the supervenience theory we considered in some detail
the question of literal equivalents. We noticed that the fact that some
metaphoric uses become conventional does not serve as an argument for the
possibility of literal equivalent. The claim that no literal use of
language is synonymous with a metaphor may mean only that no literal
equivalent is available at the time. However, the argument succeeds if
the claim was that the meaning of a metaphor had some specialcharactere
istic which prohibits its expression by the literal use of Speech.
Combining these considerations we may claim that the meaning'involved
in metaphor is not of a different type than that of‘a literal expression
even though the literal expressionis meaning is determined by convention.
That is, conventions alone may determine the sense of a literal expresv
sion.

We noticed that the proposal of examples for testing whether meta-
phor is translatable will settle nothing since in the end one must beg
the question with regard to his literal translation. We agree with
Beardsley that predicative weight can be assigned by the literal use of
language. Further, we agree that the fact that a literal paraphrase may
lack certain poetic qualities is not relevant to the question of whether
such paraphrase exists. Indeed it seems that Beardsley is correct in
saying that such paraphrase aids in the appreciation of the original.

We noted two important aspects of paraphrase concerning which we

found Beardsley in error. We noted that it is not the case that we can

152

paraphrase every phrase which we understand, whether or not the phrase
is a metaphor. Also we noted that it is the aspect of a metaphor's

meaning which we fail to understand that often prompts our desire for
paraphrase. We hope that the paraphrase will make that meaning clear.
Such a paraphrase can, if we discover it, be offered as proof that we

understand the metaphor.

Turbayne
Turbayne's discussion of metaphor is primarily concerned with
describing how one is used by or taken in by a metaphor. In laying the
foundation for such a discussion Turbayne presents a version of the
absurdity view. While he more often speaks of the metaphor as inappro-
priate rather than absurd, he does indicate that a literal interpreta-
tion would be absurd and further his work is directed to the considera-
tion of such absurd interpretations.
Turbayne begins his description of metaphor by introducing the
notion of sort-crossing. He says:
However appropriate in one sense a good metaphor may be, in another
sense there is something inappropriate about it. This inappro-
priateness results from the use of a sign in a sense different from
the usual, which use I shall call "sort-crossing." Such sggt-
crOSSing is the first and defining feature of metaphor....
In producing what he considers a more adequate definition of metaphor,
Turbayne makes use of Ryle's explanation of "category mistake." Turbayne

presents what he calls the defining feature of metaphor as follows:

Metaphor consists in "the presentation of facts of one category
in the idioms appropriate to another."26

153

Though Turbayne does not claim that this definition of Ryle is to be
adopted as the definition of sort—crossing, it is clear from his later
remarks that he has made such an adoption. We notice that this will
constitute a more narrow construal of 'sort—crossing' than is indicated
by the earlier explanation; there, any unusual use of‘a sign would cone
stitute metaphor. Bat it is clear that the unusual use must be of the
particular kind described by Ryle. But, even with Ryle's help the
notion of sort—crossing is none too clear (but we are none too clear
about category mistakes either). While the examples given by both Ryle
and Turbayne may suggest the sort of thing they have in mind, an example
is hardly a definition. And, we would be hard pressed to give a precise
explanation of either 'sort' or"category.' The following discussion

by Turbayne may indicate what he intends by 'sort.‘

. in a sense some things come with their names on them. We "find"
words already standing for things. We also "find" recurrent char»
acteristics, connotations, things already numbered and sorted out,
and necessary connections already threaded for us. They are out
there a riori, but not a priori to all experience, merely to ours.
These categories are as it were, built into our minds through a
long tract of time from the moment we appear. They are part of our
cultural, not natural, heritage.27

Still we hesitate to attempt any definition of 'sort.' Were there no
other ground to challenge Turbayne‘s view we would reject at least this
presentation since the nature of sorts remains unclear. ‘There is however
a more solid and obvious criticism to be made of Turbayne's view.' That
more obvious criticism opposes Turbayne‘s claim thatmetaphor‘involves
pretense.

For Turbayne, sortacrossing alone is not a sufficient criterion

of metaphor. He holds that, in addition to sort-crossing, metaphor

154

involves pretense or make-believe. He states:

The use of metaphor involves the pretense that something is the
case when it is not.28

In elucidation of the pretense involved in metaphor Turbayne provides an
analysis of the metaphor 'man is a wolf.‘
... When I say that man is a wolf (metaphorically speaking I am
actually giving him a name shared by all other wolves just as if
I believe that he is another sort of wolf like the timber-wolf or
the Tasmanian wolf, sharing with them all the defining properties
of "wolf," or sharing with them inclusion in the denotation "wolf."
But though I give him the same name I do not believe he is another
sort of wolf. I only make believe he is. My words are not taken
literally but only metaphorically. That is, I pretend that some-
thing is the casg when it is not, and I implicitly ask my audience
to do the same.2
In further clarification of the pretense involved Turbayne indicates that
there is no pretense about transferring names; that is, "man actually

shares the name 'wolf.'"30

One pretends, however, according to Turbayne,
that man is a wolf. (We suppose that since Turbayne is a man, when he
uses this metaphor, he gets down on all fours and howls at the moon.
This is certainly how we would pretend to be wolves.) But according to
Turbayne, more is involved here than pretense. He says:
"I do not merely pretend that man shares the properties of wolves;
I intend.... Thus when I say that man is a wolf (metaphorically
speaking) I intend that he shares some of the properties of wolves
but not enough of them to be claSSified as an actual wolf....
We wonder at Turbayne's use of 'pretend.' It does not seem to be a con-
ventional use of a term nor does it seem necessary to use the term at all.
We notice a distinct difference (which Turbayne, bygthe way, alludes to)
between man being a wolf and man sharing properties with a wolf. If, as
is indicated by Turbayne's statement, the point of the metaphor is that

man shares properties of wolves, no pretense is necessary. It is clear

155

that man shares many properties with wolves. To speak here of pretending
and intending is unnecessary. One could more easily maintain that the
sentence 'man is a wolf' indicates that Man has many of those properties
which are shared by other creatures called wolves. When we say, to use
another of Turbayne's examples, 'The timber wolf is a wolf,‘ we surely

do not mean that the timber wolf shares all the characteristics of other
wolves; if we did, we should be unable to distinguish it from the Tasman-
ian wolf. What would be crucial to explain is how one fails to mean that
man shares a sufficient number of properties to be a wolf. What Turbayne
needs to explain is how he manages to "intend that he shares some of the
properties of wolves but not enough of them to be classified as an actual

wolf...."32

This, he fails to do.

Talk of pretending is somehow out of place. Pretense indicates one
makes a show of belief or perhaps attempts to deceive someone. But when
one says, 'Man is a wolf' he certainly does not want people to think that
he believes that men are wolves; neither does he attempt to deceive his
audience. While he does speak as if he would if man were a wolf, neither
he nor his audience acts as if man were a wolf. No one pretends.
Turbayne might reply to these remarks by observing that we had been used
by his metaphor. (Being used by a metaphor is a notion of Turbayne's
which we consider later in this section.) But we have looked for the
metaphor and found none. If he takes such a position then he must admit
to being unclear and failing to present his view in an understandable
manner.

Turbayne attempts to shed further light upon the nature of pretense

involved by discussing various uses of terms. Returning again to the

156

'man is a wolf' example, he says:
But when I speak of the timber wolf and the man-wolf it is not a
case of referring to two different sorts of wolves. This is so,
because the word "wolf“ is being used in two different senses even
though it is as if it were being used in only one. The two differ-
ent senses are called the literal and the metaphorical uses Of
'wolf.'33
Here it is unclear as to what would count as using a word as if it had
only one sense as opposed to using it as if it were being used in two
senses. In his example comparing the timber wolf with the man-wolf, each
case was a case of using 'wolf' in one way. And the way 'wolf' was used
differed with each case. And it was clear that the word was in each case
being used differently from the other. But neither of his examples
gives any clue to the feature of use which constitute being used as if
there were only one sense (since he clearly does not intend that they be
understood as synonymous). It is not at all clear for example, how these
examples differ from simple ambiguity (in so far as 'as if' is concerned).
This question is of particular import since it is the uas if" feature by
which Tyrbayne distinguishes metaphor from ambiguity. In explaining that
duality of meaning is not a sufficient criteria of metaphor, he states:
Awareness of duality of meaning, as we have seen is not enough to
do it. Neither my awareness of my ability to see the point of a
needle at the same time as I say that I can see the point of a
joke.... In all these cases I am using words in one or another of
these literal senses. I am representing the facts of one sort in
words that may be equally appropriate to facts of another. What
more is needed to make these expressions metaphorical? The answer

ties in the as if or make believe feature already sketched and
illustrated.

 

Still, Turbayne presents no adequate explanation of the "as if" feature
nor any reason for us to agree that metaphor involves pretense or make

believe. His talk of words ”being used in two different senses even

157

though it is as if it were being used in only one" is of no help since
Turbayne fails to provide any criteria for such use.

In his explanation, Tutbayne indicates that whether an expression
is a metaphor is a relative matter; relative that is, to the person
uttering or receiving the metaphor.

In considering, among others, the dead metaphors 'machine of nature'
and 'kinetic energy,‘ Turbayne notices that there is usually no pretense
since most people are not aware of the original literal meanings. He
says:

Placed in their customary contexts these present to him nothing

but the fact of literal truth. To the initiated, however, who are

aware of the "gross original" senses as well as the now literal 35

senses, they may become metaphors. There are no metaphors peruse,
Thus, according to Turbayne, an expression is a metaphor "for an indi—
vidual" if he recognizes both meanings and engages in make-believe.

The talk of makeebelieve or pretense becomes even more significant. We
have above brought strong criticism against that feature as it was
sketched and illustrated. 'Using two senses of a word as though there
were only one' becomes no clearer with repetition; neither does 'pretend-
ing that two senses are one.’

While we find little value in Turbayne's explanation of metaphor,
we are interested in his discussion of how people (especially philoso—
phers) are misled by metaphor. Turbayne attempts to explain how the
literal interpretation of a metaphor which is obviously false or absurd
may be taken as true. As a background, Turbayne uses a discussion taken

from Black. Turbayne states:

Black's meta-metaphors, however, are more illuminating. They
accomodate the feature of attitude shift, and are, therefore, '

158

harbingers of things to come in this book. An effective metaphor,
he says, acts as a screen through which we look at the world; or it
filters the facts, suppressing some and emphasizing others. It
Brings forward aspects that might not have been.seen at all through
another medium...." A good metaphor produces thereby "shifts in
attitude."36

Turbayne likens the repeated metaphor to advertising or propaganda in
that through repetition it "comes to be believed more seriously.”37
As the phrase becomes conventional, the awareness of the duality of mean-
ing is lost while the assumption of truth remains. Turbayne says:

The human characteristics that Aesop pretended were owned by

animals have become literally part of their equipment. We no

longer make believe that foxes are cunning and lambs gentle.

They are. It.will take another Aesop to make bulls cruel and

lions murderous. When the pretense is dropped either by the orig-

inal pretenders or their followers, what was before called a screen

or filter is now more appropriately called a dis uise or mask.

There is a difference between using a metaphor and Being used by it,

between using a model and mistaking the model for the thing modeled.

The one is to make believe that something is the case; the other is

to believe it.38
Turbayne's point is, it seems, that what begins as a means of explaining
one thing in terms of another with an awareness that those other terms
are not appropriate and therefore, require careful consideration, becomes
an acceptance of that explanation without such awareness. The mind has,
for example, become for some a place in which certain events occur.

To provide an explanation of this feature of metaphor, Turbayne in-
troduces a term which parallels 'sort-crossing.' That term is 'sort-
trespassing.’ While giving the appearance of tying the explanations of
use and misuse of metaphor together, the term 'sort-trespassing' suffers
all the problems of 'sort-crossing.‘ The chief among those problems is
a lack of explanation of what constitutes a sort. Turbayne explains

sort-trespassing by distinguishing it from sort-crossing. Both involve

159

the presentation of facts of one sort in the idioms of another. This
error provided that sort-crossing involves no error provided that one
is aware of it. "On the other hand it is a mistake to present the facts
of one sort in the idioms of another without awareness. For, to do this

is not just to fuse39

40

two different senses of a sign; it is to confuse
them.
Not every case of sortatrespassing involves metaphor, according to
Turbayne. One could be confused with regard to different literal mean-
ings of a word. (As he would be if he wondered whether general content-
ment were a major general or a lieutenant general.) A case of sort-
trespassing becomes a case of taking a metaphor literally "only when one
of the two different senses confused is metaphorical and this is taken

for literal.41

Returning to the man-wolf example Turbayne claims that
one who recognizes the duality of meaning is using the metaphor and makv
ing believe. Someone who does not recognize the duality of meaning and
believes (as opposed to make believe) that man is a wolf is "taken in"
by the metaphor.

For him it is not a case merely of different sorts of values.

In which case for him the assertion that timber wolves and men

are wolves contains no absurdity.42
At this point, the explanation of someone being taken in by a“metaphor
requires careful explanation. We recall that a metaphor is not a meta-
phor pgr_§§_but only for someone. According to that discussion, an expres-
sion would not be a metaphor for someone who recognized only a single

sense or who was not pretending. If this is the case, then a metaphor

which has taken in someone turns out not to be'a metaphor. In attempting

160

to avoid this problem, Turbayne says:
... it is better to say that sometimes the metaphor is not noticed;
it is hidden. That is, if X is aware of the metaphor while Y is
not, X says that Y is taken in by the metaphor, or being used by it,
or taking it literally. But for Y it is not a case of taking the
metaphor literally at all, because for him there is no metaphor.
He is speaking literally or taking it literally.43
But this discussion seems only to pile complication on complication.
A discussion of 'is aware of' and 'being taken in by' will bring this to
light. Given Turbayne's other statements, X's being aware of a metaphor
must be a matter of X's being aware that there is a duality of meaning
and further that he or someone else is pretending that the expression
had but a single meaning. 'Y, in Turbayne's example, would not be aware
of both the duality and the pretense (though he might be aware of one of
them). Now, according to the example, Y is said to have been taken in or
to have been used by the metaphor. But ordinarily the phrases 'being
used' and 'being taken in' indicate that someone was cheated or deceived.
But nothing in the examples suggests cheating or deception. That Y was
unaware of a duality of meaning or that someone else was pretending,
entails no deception or cheating. Such a situation could certainly
result in Y's failing to understand the person uttering the expression in
question. When one fails to recognize that another is using language
metaphorically he would think that the person was confused or uttering
nonsense, or perhaps that a sign had been used inappropriately. He might
go on to suggest that the speaker needs instruction with regard to certain
facts or with regard to language usage. Y might be somewhat perplexed

by what he has heard; but he won't be deceived or cheated. To misunder-

stand another's utterances is not to be deceived. Where communication is

161

important a user who understands the metaphor is no better off than the
listener who does not understand; communication fails to take place.
While we disagree with the details of Turbayne's explanation, we
find that it indicates and attempts to explain an important feature of
some metaphors. There are cases where it appears that people have made
errors through failing to recognize a metaphor. Nevertheless, we are
not claiming that anyone has failed to recognize cases of pretense or
make believe in regard to metaphor. Rather we claim that they have
failed to recognize the fallacy of equivocation, that is, they have
failed to notice a change in meaning. As we shall discuss in the next
chapter, the metaphorical use of a sentence can express significant
truth. When this occurs and the statement is recognized as true, one
attempts, among other things, to determine what follows from that state-
ment. One attempts to determine what is implied by this new truth (or
perhaps old truth). One falls into error if in the process of his
reasoning he attributes to this statement some meaning which is appro-
priate to its literal, not its metaphorical, use. One could indeed be
misled if he accepts an expression as true given one interpretation and
then bases his inferences from that expression some other interpretation.
This is perhaps what Turbayne attempts to point out when he explains
how we are taken in by a metaphor. He says "The long continual associ-
ation of two ideas especially if the association has theoretical and
practical benefit, tends to result in our confusing them."44
While it is important to consider the errors which one can fall into
when considering a metaphoric explanation, we find that Turbayne's

explanation is generally unhelpful. 'His nation of sort is never eXplained:

162

therefore, sort-crossing and sort-trespassing remain a mystery. It is
obvious that one is able to use metaphor without pretending or making
believe. 'Being used by' or 'taken in by' metaphor similarly contain
connotations which are at best misleading. As we have demonstrated above,
it is possible to discuss these problems of metaphor without resorting to

such misleading terminology.

Conclusions Regardingithe Absurdity Theories

Both versions of the absurdity view which we considered indicate
important aspects of metaphor. Unfortunately neither of these versions
is particularly helpful in understanding metaphor. Beardsley adds little
to the work done by Henle. From this, one suspects that Beardsley's
basic contention is correct; namely, that the falseness or absurdity of
the literal interpretation is our indication that a figurative interpre-
tation is in order. But one also suspects that this observation in itself
is not sufficient to explain metaphor. We conclude that Henle's and
Beardsley's observations will play an important part in the correct view
of metaphor, but more is required for a complete explanation.

Turbayne's explanation considers a problem that has tripped up
philosophers in the past. But his introduction of terms such as 'sort-
crossing,‘ 'sort-trespassing' fail to provide either a clear explanation
of metaphor or of being misled by metaphor. We noticed that what
Turbayne explained as being used by metaphor can be explained without
resorting to Turbayne's special terminology. While his discussion does

suggest our explanation in terms of the fallacy of equivocation, we find

163

his discussion raises more questions than it answers.
Both these versions of the absurdity indicate features of metaphor

to be explained. Neither constitutes a satisfactory account of metaphor.

10.
ll.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.

O m \l 01 U1 4:- (A)
I O O C O C

164

NOTES TO CHAPTER V

. Monroe C. Beardsley, Aesthetics Problems in the Philosophy of

Criticism (New York: Harcourt, Brace and World, Inc., 1958).

. Colin Murray Turbayne, The Myth of Meta hor, revised edition

(Columbia: University of South aro ina Press,'1962).

. Ibid., 95.

Beardsley, l34-l35.
Ibi
bi

O.

, 135.

H
O.

., 25.
136.
434.

H
U'
a—lo
Q.
.

H
U”
do
0..
U

H
U-
do
0.
.

435.

H
U'
..a

Q-

0—!
U
—lo
0..

U

436.
137.

H
U"
do
O.
'

D—l
U"
do
0.

137—138.

H
U-
do
Q.
o

H
U'
do
0.
.

138.
143.
138.

H
U'
.1.
o.
o

H
U"
a-lo
D.
.

H
0'
—lo
0.

H

bid., l40-l4l.
Henle, 183.

 

Beardsley, 138—139.

22.
23.
24.

25.
26.

27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.

40.
41.
42.
43.
44.

Ibid

bid

H

165

., 142.
., 143.

We treated Henle's work separately since his emphasis was on an
explanation of a relational view rather than a statement of a
controversion view.

Turbayne, ll.

Ibid

., 26. Turbayne cites: Gilbert Ryle, The Concept of Mind

 

(London: Hutchinson University Library, 1949), p. 8.

bi

H
O.

H
O-
-I

Q

H
U"
—I

Q

0—:
U"
—I

Q

H
0'
_|

Q

0—!
U'
--I

Q

H
U'
-I

Q

0—!
U'
-I

Q

H
U'
..a

Q

0—.
U
..I

Q

H
U"
...:

Q

bi

0—:
Q

., 79.
., 13.
o, 140

.. 14-15.

0,170

0’ 18.
., 21. Turbayne cites: Black, 287-288.

., 22.

Turbayne's previous mention of fusing meanings is "Any trope can
achieve full metaphorshood only but only for that user who
fuses the two senses by making believe there is only one sense."
Turbayne, 18.

bi

H

H
Q

bi

Q

H
U'
.4.
Q

H
U'
..a

Q

H
U'
..I

Q

., 23.

,, 23-24.
, 26.

CHAPTER VI

VERIDICALITY THEORIES

The last feature of metaphor to which we made reference in our
introduction was the truth of metaphor. From the point of view of phil-
osophy in general, the most significant question is whether metaphors
can be true in the same sense as other statements can be true. The
possibility of using metaphor depends upon one's ability to speak truly
when doing so. In this chapter we consider the work of philosophers who

1 2 3 and

have treated this question, specifically Goodman, Elliott, Mew,
Myers.4 The result of our consideration is that some metaphors are
indeed true.

In presenting Goodman's position a question may arise concerning
his nominalist position. We do not, here, portray him as a nominalist

since in The Languages of Art he does not speak as a nominalist. We take

 

him at his word, rather than reinterpreting in nominalist terms because
we believe that a nominalistic interpretation would be inconsistent with
positions that he takes.

Goodman proceeds by arguing that some metaphors are true and then
discussing the nature of what he calls 'metaphoric truth'. Elliott
offers an argument that all metaphors, at least in poetry, are true.
Mew's article is written in response to Elliott. While Mew's article is

not a complete refutation of Elliott, it offers arguments which should

166

167

cause Elliott to weaken his position. Myer's position is that metaphors
do not have truth value. Metaphors, he says, may be considered as
useful or not useful, but should not be considered as either true or
false. We will find Mew's argument unconvincing and maintain our posi-

tion that at least some metaphors are true.

Goodman

Goodman's presentation treats three major issues: (l) the question
of truth in metaphor, (2) an explanation of the nature of metaphor in
terms of realm and schema, and (3) the understanding of metaphor.

Goodman's discussions of truth in metaphor relies heavily upon the
introduction of the terms 'metaphoric possession' and 'metaphoric truth'.
We argue that these terms do not help us understand the nature of truth
in regard to metaphor. They only create new problems and tend to hide
the central issues.

Goodman's proposal concerning metaphorical truth, we shall argue,
is based upon an example which fails to represent many aspects of metaphor.
It so fails because it is a dead metaphor. His example is 'This picture
is sad'. Such a metaphor, if taken as an example of metaphor in general,
gives the appearance that there is some rather obvious interpretation
for each metaphor. It is this assumption which leads Goodman to believe
that answer to the question of truth is quite straightforward. In fail-
ing to recognize the unlimited number of metaphorical uses of an expres-
sion, Goodman fails to provide criteria of truth for live metaphor.

Goodman's second concern is the explanation of metaphor in terms of

realm and schema. A metaphor, he proposes, results when an old word is

168

applied to a new set of objects to make distinctions among them which
bear some similarity to those made by the conventional“nse'of'the*term.
While we agree that this explanation does characterize some metaphors,
we argue that it fails to characterize all metaphor. We'find that even
for that class of metaphorS'which is characterized above, the introduc-
tion of 'schema' and 'realm' does not facilitate the explanation of a
change in meaning.

Goodman's third discussion treats the understanding of metaphor.

His explanation is a return to the comparison theory in which he collapses
the distinction between metaphor and simile and treats metaphor as ellip-
tical simile. This analysis forms, for him, a basis for rejecting ques-
tions regarding the relationship between metaphoric and literal speech
(the very question to which he addressed himself). The question is
reduced to the question of why some objects have the properties they do,

a question which Goodman leaves for the cosmologist. We shall argue that
his reduction of this question, if it does indeed follow from his explan-
ation of metaphor as elliptical simile, constitutes a reductio-ad-
absurdum of his explanation.

We shall end our discussion of Goodman by noticing that it is for
the most part based upon the analysis of his example 'This picture is
sad'. Given the general theme of the book, it may be that his is the
only metaphor (or only type of metaphor) which needed explaining. The
presentation does however appear to be intended for all forms of meta-
phor. We shall note also that the technical terms introduced by Goodman

are generally unenlightening and that insofar as the theory in

169

Languages of Art is correct, it merely reflects the work of Goodman's
predecessors in this field, particularly Black and Richards, both of whom

he credits.

Metaphoric Truth

Goodman is interested in showing that metaphor is not an alternative
to truth; that is, to say that a statement is metaphoric is not to rule
out the possibility of that statement's truth. People sometimes say, for
example, "'x is p' is false but you can see what I meant." In so doing,
they indicate an underlying assumption that a statement which is meta-
phoric is not true. Goodman argues that there is a sense in which meta-
phors may be true as well as false.

Goodman initiates his attack on the theory that metaphors cannot be
true by considering the death of metaphor. As the novelty of a metaphor
wanes, he says, the metaphor becomes more like literal truth. What is
lost is the liveliness of metaphor, not its truth. In this discussion
lies the following suppressed argument. Dead metaphors are in some cases
obviously true. (For example: the statement 'Mary flew into a rage'
would be true if Mary became very angry quite suddenly.) In some cases
of dead metaphor, however, the only relevant feature to change since its
inception is our becoming accustomed to it. These metaphors must, there-
fore, have been true while live. There are, therefore, true live meta-
phors. This argument is correct.

Having established that some metaphors are true, Goodman proceeds
much as did Kant with his synthetic-a-priori, to show how metaphors can

be true. In so doing, Goodman uses as a model the metaphor:

170

The picture is literally gray but only metaphorgcally sad.
But is it literally or metaphorically cold in color?

In attempting to say how it is possible that a painting could be sad, he
states:
Metaphorical possession is indeed not literal possession; but
possession is actual whether metaphorical or literal. The meta-
phorical and literal have to be distinguished within the actual.
This statement which is reminiscent of Meinong leads one to wonder
whether Goodman is introducing a new metaphysical theory. Although he
discussed possession on earlier pages, Goodman fails to distinguish be-
tween literal and metaphoric possession and the terms 'metaphroical ;
possession' and 'literal possession' enter the discussion withoUt intro-
duction or explanation. He provides no basis for determining whether a
property is possessed literally or metaphorically. We find that we are
left to attempt to make some clear sense of this notion of metaphorical
possession. We shall now make such an attempt.

To understand metaphorical possession and how it accounts for meta-
phoric truth, let us consider two statements. The first is rather rare.

(1) 'Metaphorically speaking, this painting is sad'.
This statement is rare because outside a philosophical discussion one

7 But even though this statement is rare, it

seldom labels his metaphors.
is easily understood. The phrase 'metaphorically speaking' is used to
alert the imperceptive hearer that what follows is not to be interpreted
literally. But now consider a statement employing Goodman's terminology
which purports to elucidate (l).

(2) 'This painting metaphorically possesses sadness'.

It is this second statement which we need to explain, since if we can

171

properly use the term 'metaphorical possession' we may be able to see how
this concept explains metaphorical truth. One could, of course, take the
position that (2) is synonymous with (1). But this position would result
in a circular argument since (2) is intended to explain (1), and (2) in
turn is explained as meaning the same as (l). Expanded, the circle would
be: 'the picture is sad, metaphorically speaking' is true if and only

if the picture metaphorically possesses sadness; and by 'metaphorical
possession of sadness' is meant 'is sad, metaphorically speaking'. Such
an explanation of metaphoric possession will not suffice.

An alternative to this explanation was suggested previously. State-
ment (2) is straightforward (that is, not in need of any unpacking) and
is based upon a new metaphysical theory. Such a theory would include the
notion of objects possessing qualities and that such possession would be
of two types, literal and metaphorical. But Goodman does not present
such a metaphysical theory and we see no ground upon which to base such
a theory.

It appears that to speak of metaphorical possession, without a com-
plete explanation of that notion, is to invite problems rather than solve
them. If metaphorical possession is to provide a basis for understanding
Inetaphor, some explanation of that notion is required; The Languages of
Art provides none.

The problems raised by the use of 'metaphorical possession' and
which are in need of explanations can be avoided by rephrasing Goodman's
discussion. Such a rephrasing would still make the same point.

The picture is gray (using 'gray' literally) and it is sad

(using 'sad' metaphorically). But is the word 'cold' correctly

applied to the picture when 'cold' is used metaphorically or
literally?

172

Avoiding the use of 'metaphorical possession' one could go on to say

'the picture has a property which is ascribed to it by the sentence

"this picture is sad" when that sentence is used metaphorically.‘ Further,
the property ascribed by the metaphoric use may differ from the property
which would be ascribed by the literal use of 'sad,' but nonetheless both
properties are real (that is, not fictitious). Goodman could then point
out that whether sad was being used literally or metaphorically would as
he suggests, depend "upon some such feature as its novelty." Inasmuch

as 'metaphorical possession' can be omitted from the discussion, it is
preferable to do so and avoid the accompanying problems.

As Goodman pointed out novelty may be our guide with regard to deter—
mining metaphoric use. But novelty“alone will not suffice. He correctly
explains that the respect in which a word is novel is critical.‘ Every
application of a predicate to a new event is in some sense novel, yet not
every application is a metaphor. The coining of a new word is not meta—

phor.

Metaphor, it seems, is a matter of teaching an old word new
tricks-—of applying an old label in‘a new way.:.. “Briefly, a meta-
phor is an affair between a predicate with a past and an object that
yields while protesting.9

In unpacking this metaphor, the following position can be constructed.
Three distinct claims appear to be made. (1) 'teaching an old word new
tricks" indicates that metaphor involves terms which have conventional
uses which are different from their use asmetaphor; (2) the"protesting
object' indicates that metaphors when interpreted conventionally turn out
to be false or at least an inappropriate use of language; (3) since the

'objects yield', one can with sufficient effort determine an interpretation

173

of the sentence which is fitting. Metaphor, Goodman says, involves an
a-priori denial of the metaphoric predicate to the object. "Application
of a term is metaphorical only if to some extent contra-indicated."10
Here Goodman is following Henle, Beardsley and Turbayne.

Goodman has claimed that there is a sense in which metaphors may be
true and once again, he sets about explaining 'truth' as applied to meta-
phor. His first proposal turned on a distinction between metaphorical
and literal possession of a property. The present explanation turns
upon a distinction between literal and metaphorical truth. Goodman uses
the 'sad picture' example to distinguish metaphor from mere falsehood
and to introduce the notion of metaphorical truth.

To say that our picture is yellow is not metaphorical but
merely false. To say that it is gay is false both literally and
metaphorically. But to say that it is sad is metaphorically true
even though literally falsez11

This distinction would make possible the denial of literal truth without
having to reject the statement. A speaker could maintain that we ought
to agree with his expression since it is metaphorically true.

Several problems are inherent in the proposed distinction between
literal and metaphorical truth and in the explanation of this distinction.
The first of these problems follows directly from Goodman's example.
Notice that to claim, as he does in his example, that 'the picture is
yellow' is not metaphorical but merely false is not to claim that some
specific use of the predicate 'is yellow' fails to apply to the picture.
It is rather to claim that there is no metaphoric use of the term yellow

which applies to the picture either truly or falsely and this may be a

hasty conclusion. If pictures can be described as sad they can also be

174

described as fearful. It may be that the painting is yellow after all,
metaphorically speaking. Goodman fails to realize that every predicate

‘2 It may be that one has

has metaphoric uses applied to any object.
failed to recognize many such uses and would deny that a predicate has

a metaphoric use in a given case. Such cases may require only insightful
interpretation. Goodman's example demonstrates that the distinction
between metaphoric truth and mere falsehood may not be as easy to make
as he indicates. A second problem with Goodman's notion of metaphoric
truth is that it apparently violates the law of the excluded middle.

The statement that the picture is yellow is, according to Goodman,
neither metaphorically true nor metaphorically false, "but merely false".
Some sentences it appears are neither metaphorically true or metaphor-
ically false.

A third problem with the notion of metaphorical truth is that the
law of non-contradiction apparently fails to apply to metaphorical truth.
One such violation of the law of non-contradiction is found in Goodman's
statement that the picture is gay. Goodman says this statement is meta-
phorically false. But suppose that picture was a sad scene of people
engaged in homosexual conduct. It would then be a gay picture and the
sentence therefore would be metaphorically true. The same sentence would
be both metaphorically true and metaphorically false.

The fourth problem with the proposal that we adopt a notion of meta-
phorical truth is that this notion is not sufficiently characterized.

We are not, for example, appraised of the truth functional relationships
between literal and metaphorical truth. (We do not know whether to count

(p > q) as true where p is literally true and q is metaphorically true but

175

literally false.) If we are to take this proposal seriously as an ex-
planation of truth, then that explanation must conform to the requirements
of standard logics or provide reasons for rejecting those requirements.

Each of the first three objections in itself provides a sufficient
ground for rejecting metaphorical truth as an explanation of truth in
metaphor. Even if these objections were rebutted, tremendous further
work would be required, as indicated in the fourth objection, to make
Goodman's proposal realistic and compatible without ordinary uses of
'true'.

Both Goodman's presentation of metaphoric truth and our replies rest

upon a fundamental error.13

That error lies in assuming that expressions
or uses of expressions have truth value which is independent of their
meaning. He says: "to say that our picture is yellow is not metaphori-
cal but merely false". In saying this he makes no reference to what one
might mean by such an utterance. Whether or not a painting were sad or
yellow would depend upon what one means by 'sad' and 'yellow'. It is
pointless to talk about the truth of a sentence until its meaning is
resolved. And if the sentence is ambiguous, then it is entirely possible
that differing truth values would be associated with the alternative
interpretations. As Goodman's discussion indicates, 'This painting is
sad' is ambiguous. We are not therefore surprised to find differing
truth values accompanying the different interpretations. But we find no
advantage in calling some of those values literal and others metaphoric.

At this point it is worth noting that the variety of metaphor upon

‘which Goodman bases his explanation is severely limited. In examining

176

the metaphors recognized by Goodman in this section, we find that they
all have a comnon feature. They are all dead. They have acquired by
repetition a second usage and are more properly interpreted as ambigui—
ties than metaphors. This is perhaps responsible for Goodman's willing«
ness to talk about a sentence being metaphorically false, or a picture
metaphorically possessing a property. Since the sense intended by the
dead metaphor is generally recognized, it is easy to overlook the fact
that when attempting to understand a live metaphor one lacks the guide-
lines provided by conventional use.

In Goodman's further explanation of metaphorical truth we find the
assumption that Goodman views metaphor as having only a single (or at
least a limited number) metaphorical use. He states:

Nothing can be both sad and not sad unless 'sad' has two dif-
ferent ranges of application. If the picture is (literally) not

sad and yet is (metaphorically) sad, 'sad' is used first as a

label for certain sentient things or events, and then for certain

insentient ones. To ascribe the predicate to something within

either range is to make a statement that is true either literally
or metaphorically. To ascribe the predicate to something in

neither range (I leave other ranges of metaphorical application

out of account for the moment)14 is to make a statement that is

false both literally and metaphorically.15
This explanation of metaphorical truth does appear reasonable. A state—
ment is literally true if and only if the object falls within the literal

‘6 A statement is metaphorically true if and

extension of the predicate.
only if the object falls within the metaphoric extension of the predi-
cate. But such an explanation assumes that there are a limited number of
Inetaphorical uses. The problem which arises with the realization that
there is no such limit is illustrated by considering Goodman's use of

'ascribe a predicate'.

177

Goodman is not clear about his use of 'ascribe the predicate to
something'. We do not know precisely what is to count as ascribing a
predicate. Specifically it is not clear whether ascription is of a
word token or of the extension of that token. This distinction may often
be overlooked in the discussion of the literal use of speech. In such
use the conventions of the language will determine the extension so that
in discussing the ascription of a predicate (word token) one may assume
the ascription of the word's extension as well. However, in the case of
metaphorical use there is not a single extension, but many. Therefore,
in ascribing a word token one fails to ascribe a single particular exten-
sion and it becomes important to specify whether one is ascribing a word
or ascribing its extension.

If the ascription of a predicate is to be construed as the ascrip-
tion of a word token, then the explanation of metaphorical truth must
be revised. Such a revision might be "A statement is metaphorically true
if and only if the object falls within one of the metaphoric extensions
of the predicate." But such a revision would be subject to two objec-
tions. (1) In attempting to determine the truth value of a metaphor one
will never have a guarantee that he has exhausted the possible extensions
(or ranges) of any given predicate which is being used metaphorically.
The determination that a particular metaphor is false would be, according
to this theory, impossible to make. It would be impossible since one
must show'that the object does not fall within any of the ranges and there
is INDIHEtHOd of determining when one examined every metaphoric range.

(2) Such a definition would be too restrictive. One may want to say of

a metaphor that it is false, even though the object may fall within one

178

of the ranges of the metaphoric predicate.

Goodman, for example, wants to maintain that it is false that the
painting is gay. And we would want to agree that the painting is not
gay, in one sense of 'gay' but it may be gay in another.

Both these objections indicate a problem with treating the ascrip-
tion of a predicate as a matter of ascribing a word token. Such an
ascription in the case of metaphor fails to ascribe a single or even
definite number of extensions or ranges. If truth is to be a matter of
an object's being within a range, then there must be some procedure for
specifying the intended range. These observations indicate that perhaps
one ought to view ascribing a predicate as assigning a particular range
or extension rather than a word token. It is clear that this is not
Goodman's position since the predicates which he ascribes may have more
than one range.17 Such a move even if permitted by Goodman, would not
be of appreciable assistance in explaining metaphor. Although an explan-
ation of truth would be provided, the theory will be uninteresting. If
one claims that to ascribe a predicate is to ascribe an extension or
range rather than a word, one avoids the problem of a great many exten-
sions and will insure a unique truth value. But this amounts to saying
that the provision of a specific meaning insures a unique truth value.
Such a statement nicely avoids the above mentioned problems; it also
avoids precisely that which needs to be explained about metaphor,
namely:

(a) How does one determine the appropriate extension of a metaphor?
(How does one choose the appropriate interpretation).

(b) Is there generally one interpretation that could be deemed the
appropriate interpretation?

179

Talk about ranges or extensions is useful only if there is some method
of determining the various ranges and which range is appropriate. We
can agree with Goodman when he says, "Whereas falsity depends upon mis-
assignment of a label, metaphorical truth depends upon reassignment."18
We can agree, that is, if Goodman will talk about truth rather than
metaphorical truth. But in metaphor the possibilities for reassignment
are limitless, both in variety and in number.

Goodman's talk about ranges does indicate a recognition that one
of the problems to be faced in a discussion of truth is metaphor is
ambiguity. But his discussion fails to account for the great variety of
ranges. When one's model of metaphor is a dead metaphor then the number
of ranges appears limited. It seems obvious in such a case that one has
a question of literal truth and another question of metaphoric truth.
But given the unlimited potential for interpretation there are an un-
limited number of truths to consider. 'The real problem of how to deter-
mine an appropriate interpretation remains. If one can settle upon an

interpretation, the question of truth in metaphor is the same as the

question of truth in the literal use.

Realm and Schema

While we find that Goodman's talk about ranges has failed to sub-
stantiate his position with regard to truth, we are still interested in
his discussion with regard to ranges. It is clear that any complete
discussion of metaphor must, in some way, account for the great variety

of uses of any particular expression when employed metaphorically.

180

Goodman's difficulty is after all not that he denied expressions had
different uses, but rather that he failed to take account of the great
variety of uses. We examine Goodman's discussion of the reassignment
of range with the expectation that it will account for the shift in
meaning which takes place in metaphoric use.

Goodman introduces the notion of "realms" as a means of explaining
the reassignment of metaphor. He indicates that even within the scope
of literal usage the sense of a term depends upon the general limits of
the current discourse. What will count as red, for example, will depend
upon whether one is distinguishing red from non-red over the entire
universe of colored objects or whether one is distinguishing red from
red-orange and red-violet. According to his discussion, the realm of a
label is a universe of objects among which that label is selective. In
the present example, the realm of 'red' would be the set of colored
objects. .Goodman, after providing this reasonably intuitive example,
defines 'realm' as follows:

The aggregate of the ranges of extension of the labels in a
schema may be called a realm. It consists of the objects sorted

by the schema -n- that iET—Ef the objects denoted by at least one

of the alternative labels.19
To understand this definition we find that we must determine the sense of
the term 'schema'. This term is not defined by Goodman, but his use of
it is sufficient to establish that (l) schemata sort objects, (2) labels
function within schemata, (3) a label may belong to any number of
schemata, and (4) schemata are constituted by labels. Goodman states:

The alternatives of a schema need not be mutually exclusive;

for instance, a set of color terms with some of their ranges over-
lapping and some included in others will serve.

181

It would seem then that a schema is a set of predicates, not a sentence
(or grammatical formula) and that a schema consists of the set of predi—
cates which sort the members of a given range. However fitting this
explanation may seem, we soon realize its circularity. The predicates
which compose the schema are determined by the realm and the realm is
constituted by the extension of the predicates which are members of the
schema. Further, it does not appear possible to explain without circu—
larity which predicates constitute any particular schema. One would
like to say it is composed of those predicates which are alternatives to
the label in question within the given realm. What the alternatives of
the label are will be determined by the context and custom. The circu-
larity of the explanation lies in the fact that 'realm' is defined as
the aggregate of the ranges of the extensions of the predicates in the
'schema' and the schema is the set of predicates whose extensions form
the realm. This circularity will not be fatal as long as further remarks
are sufficient for us to understand the position and as long as we base
no further conclusions upon the assumption that an explanation has been
provided.

The alternative labels which together with the label under discussion
are not generally specified, either in literal or metaphoric use. We
understand from context, for example, that 'red' is functioning as an
alternative to 'red-violet' or 'red-orange'. If we consider 'sad' we
find it used in contexts as an alternative to 'happy', 'angry', fjoyous',
etc. Generally, 'sad' with these alternatives would compose the 'schema',

the realm being people or other sentient beings.

182

Goodman makes use of our understanding of"realm' and “range' to
explain what transpires when metaphor is used. He states:
Metaphor typically involves a change not'merely of range but
also of realm. A label along with others constituting‘a“schema is
in effect detached from‘theghome‘realm“of that‘scgema and applied
for the sorting and organizing of an alien realm.
In the 'sad picture' example, the new realm is that of works of art.
Pictures are organized in terms of 'sad', 'happy'. etc. 'The range of the
term 'sad' now includes the members of the new realm, and it appears that
change in realm would indicate that the new realm was not an expansion of
the old realm to include pictures but that the metaphoric use of sad no
longer applied to sentient beings. This claim about his theory seems
justified by his willingness to separate metaphorical sadness from literal
sadness.

Consider this explanation of metaphor as it would be applied to the
following example:

The lion of England leaped upon the infidels.
We might say that 'lion' and its schema is detached from a home realm
where it serves to separate one beast from another; other members of the
schema might be 'cow', 'dog', 'dodo'. In its new realm, 'lion' sorts
people, perhaps kings, from commoners or the courageous from cowards.
But when we look for some justification of this analysis we can find none,
given Goodman's theory. Specifically there is no justification for the
choice of the home realm. In Goodman's theory, labels seldom operate in
a unique realm. ("... even a label with a unique range seldom operates

"’22

in a unique realm. We have no basis upon which to pick one realm

over another as the home realm. While it is true that the reference of

183

a metaphoric use of an expression is not that which we would expect from
the literal use, talk about home and foreign realms is helpful only if
we can establish what is to constitute the home'realm.1

Goodman's explanation of metaphor in terms of changing realms does
not square with what speakers of a language do when they utter or inter«
pret metaphor. There is no evidence that each time*one“speak5'metaphor«
ically that he has in mind some set of predicates (the schema) which he
intends to transfer to a new realm. 'Even if it could be established
that an author always had in mind a set of alternative predicates whose
referents establish a home realm, since authors nearly always fail to
announce those predicates, his audience would have no idea what they
were.

Explanations in terms of realms are subject to both theoretical
and practical criticism. Although it might seem rewarding to talk about
transferring a schema from its home realm, it is not clear that a predi-
cate has a home realm. It seems rather that predicates have many realms,
none of which is home. From the practical side we find that talk about
realms does not help in understanding what people do when they use or
understand metaphor. Goodman has provided no evidence that authors have
in mind sets of predicates when they use metaphor or that their audiences
recognizes those sets of labels. Our argument here is not that words
fail to have common or conventional uses or meanings. Nor is our argu—
ment that the metaphoric use of a term is dependent in part at least on
the way it is used literally. Our argument is that Goodman's talk about
schemata and realms is of no help in explaining the relationship between

literal and metaphoric uses.

184

Talk of shifting schemata between realms may have some merit with
regard to understanding particular metaphors. One can explain the sig»
nificance of a Particular metaphor, in some cases, by showing that the
word is used metaphorically to sort items in a way similar to that in

23 Consider Goodman's

which it is used to sort items when used literally.
discussion of the sad picture. It is the case that mentalistic terms
generally (perhaps even now by convention) are applied to works of art
in ways that suggest Goodman's analysis. That is, the opposites of
mentalistic terms when applied literally (say to humans) remain their
opposites when applied to art. For example, the opposite of a sad
picture would be a happy picture. A non-sad picture might be angry.
But this example is once again too narrow to form the basis of a theory
of metaphor. Consider another example to which the above analysis is
ludicrous. This example is the equally dead metaphor concerning King
Richard.

'Richard is a lion.‘
'Lion' is not used as the opposite of 'gorilla'. When Richard is called
a lion it is not to distinguish him from a gorilla, or some other beast,
but to indicate that he has some characteristics of the lion. If it
were said that Richard was not a lion, we would not set out to determine
whether he might not be a kangaroo in the way that we might attempt to
determine whether a non-sad picture was angry. It would not be silly to
attempt to determine whether a non-sad painting was happy. While 'sad'

continues to function among its usual alternatives (or schema) in the

metaphor examined, 'lion' does not.

185

Various metaphors may be explained in terms of realm and schema

24 But the adoption of

as is indicated by the 'sad picture' example.
such an explanation as a paradigm is unsuccessful and cannot be gen-
eralized to apply to all metaphors. It seems that this is the case
with Goodman's discussion of realms.

Overlooking the technical considerations and attempting to examine
the general position with regard to realms, we find Goodman's state-
ment so obvious as to be trivial. It is no surprise that metaphor is
a new use of a word, that it is applied to groups of objects to which
words of that type were not thought to apply, that some metaphorical
'statements are true and others false, and that there is a change of
extension when a word is used metaphorically. The important question

of how one is able to understand the new extension of terms used

metaphorically remains unanswered.

Understanding Metaphor;

We have, in this chapter, pointed out that the most significant
problem to be answered by a theory of metaphor is how one understands
metaphor. We claimed that a discussion of the truth of metaphor which
is independent of the discussion of its meaning leads to unreasonable
difficulties . But, in answer to questions regarding the understanding
of metaphor Goodman responds:

Presumably we are being asked, rather, for some general
aggoggt of how metaphorical use of a label reflects its literal

Goodman proposes two answers to this question. The first, which he

attributed to Cassier, is presented and dismissed. Metaphor, according

186

to that theory, is a matter of using a term, in an unconventional way,
but in a way that it was used at an earlier period in one's linguistic
development. Goodman provides the following example. An infant applies
the term 'mama' to everyone, later narrowing its application until it is
used to denote a single person. 'Mama' may later be used metaphorically
as applying to all females. Goodman's reply to this proposal seems
sufficient. 5

... however illuminating it may be, and however true for some
cases, it obviously does not explain the metaphorical application
of all or even most terms. Only rarely can the adult adventures
of a label be thus traced back to childhood deprivations.25
Goodman then considers the view that metaphor is essentially ellip-

tical simile. Although he does not say that this is his view, he does
not criticize it (intentionally). It seems, from the manner of presen-
tation, to be an explanation with which he agreed. In any event, we are
more interested in evaluating positions than attributing them. Goodman
states:

Metaphor has often been construed as elliptical simile, and
metaphorical truth as simply the literal truth of the expanded
statement.27

This is again essentially a statement of the comparison view and fails
to overcome the objection that one of the objects being compared is not
mentioned. This difficulty is particularly clear in Goodman's own
example.

But the simile cannot amount merely to saying that the picture
is like the person in some respect or other; anything is like any-
thing else to that extent. What the simile says in effect is that
person and picture are alike in being sad, the one literally, the
other metaphorically. Instead of metaphor reducing to simile,

simile reduces to metaphor; or rather the difference between
simile and metaphor is negligible. Whether the locution be

187

"is like" or "is", the figure likens picture to person by picking

out a certain common feature: ‘Th3t_the predicate "sad" applies

to both, albeit to the person initially and to the picture

derivatively.28
Goodman has here constructed for us the comparison which is to be the
explanation of his example: 'sad picture'. While considering the meta-
phor as a comparison may explain some meaning, Goodman fails to explain
how one knows that a person is that to which the picture is to be
compared. In the original metaphor, no persons were mentioned. There
is no evidence that the comparison need be between picture and person
at all. Many other objects are at times sad, dogs, trees, buildings,
chimpanzees, etc. The comparison theory of metaphor will explain our
use and understanding of metaphor only if it includes an explanation of
our knowledge of what is to be compared.

Goodman's own example gives rise to additional objections to the
comparison view since it appears to be a counterexample. When we say of
a painting that it is sad, we do not intend to ascribe to the painting
those characteristics which would lead us to say that a person was sad.
Goodman's explanation of the comparison turns upon a vacuous similarity.
Goodman says, we recall, "What the simile says in effect is that person
and picture are alike in being sad, the one literally and the other

29 We notice that what is alike is the fact that the

metaphorically."
same word is ascribed to both objects. But the fact that the same word
is ascribed to distinct objects fails to establish any further similarity.
Consider Goodman's line of argument applied to a different subject.

A turn to starboard is similar to a morally acceptable action in that the

word 'right' applied to both. We do not say that a starboard turn and a

188

good action are alike in being right, the one directionally, the other
morally. The fact that the same sign type (or replicas of a sign)
applied to two objects or events does not indicate that the objects
are similar (that they have anything in common other than the applica-

tion of a particular sign).30

To explain the comparison involved in
the comparison theory of metaphor one needs to do more than observe
that a word is applied to different objects since such an application
often.indicates no comparison at all. Even, as in the case above, the
fact that a word is correctly applied to two objects does not guarantee
a comparison or a similarity between the objects. If, as a matter of
fact, some metaphors are comparisons, more will be required to explain
metaphor than the observation that the same word is applied to distinct
objects.

Here, we need to carefully distinguish the explanation in Languages
gf_Art_from that offered by others who claim that similes are metaphors
or vice versa. Wheelwright, for example, is concerned to show that the
distinction between metaphor and simile is not a grammatical distinction.
As noted in the introduction, one often hears “a metaphor is a comparison
without the use of 'like' or 'as'." Such a statement would appear to
distinguish metaphor on the basis of grammar. But Wheelwright argues
that changing the grammatic structure of a metaphor so that it includes
'like' or 'as' will not change its meaning. The simile which results is
not a comparison, but rather a metaphor. Wheelwright does not claim that
there is no distinction between metaphor and simile, he claims rather

that aesthetically important similes are in fact metaphoric and that some

189

apparent metaphors are only similes. That is, he distinguishes meta-
phor from comparison.

In examining Goodman's position, we find that he is not following
Wheelwright in distinguishing metaphor from comparison, but is rather
collapsing them and, in effect, treating metaphor as elliptical simile.
Although the talk of literal and metaphorical application gives some
appearance of rejecting that claim, we again recall that he stated:

. the difference between simile and metaphor is negligible.
Whether a locution be 'is like' or 'is', the figure likens picture
to person by picking out a certain common feature: that—the
predicate 'sad' applies to both, albeit to the person initially
and to the picture derivatively.31

His claim that the difference between simile and metaphor is negligible

is essentially a matter of reducing both to comparison. This treatment

of metaphor as comparison provides the basis for Goodman's discussion of
the relationship between metaphor and literal speech, which we now con-

sider.

In explaining metaphor as elliptical simile, Goodman has made the
following statements:

(l) Metaphor is to be explained as comparison.

(2) That explanation is based upon the observation that the same
predicate is applied to two objects.

Holding this position one might quite naturally ask the question which
Goodman asks; namely:

"... What sort of similarity must obtain between what a predicate
applies to literally and what it applies to metaphorically."32

(We recognize that Goodman presents this question here only to be re-
jected.) Such a question would be asked since there seems to be more

than a fortuitous relationship between the metaphorical and literal

190

application of a term. But Goodman does not attempt to answer this
question. Instead he argues that there is no good answer to this ques-
tion. We cannot, for example, explain what objects must have in common
to have a single predicate apply literally. They must, of course, have
in common some property.
But what property? Obviously the property named by the predi-
cate in question; that is, the predicate must apply to all things
it must apply to. The question why predicates apply as they do
metaphorically is much the same as the question why they apply as
they do literally. And we have no good answer in either case,
perhaps that is because there is no real question.33
With this final argument we are to be satisfied as having an explanation
of a metaphor. That there is no good answer is perhaps a sufficient
basis for rejecting a question (assuming that the lack of an answer is
not a matter of ignorance). But we are not satisfied.

When a rejected question is the reasonable result of a previous
explanation, the rejection of the question is a strong argument for
rejecting the explanation. We recall that what was to be explained was

34 In

”... how metaphorical use of a label reflects its literal use".
following Goodman's line of reasoning, we find that metaphor is essen-
tially a comparison. (Goodman states: "the figure likgg§_picture to
person by picking out a certain common feature....")35 The common feature
he notes is the application of the same word to two objects (in his
example 'sad' is applied to both the person and the picture). But this
leaves unanswered the original question of how the metaphorical use
reflects literal use. Given his adoption of the comparison view, Goodman
interprets this question as being "what sort of similarity must obtain

between what a predicate applies to literally and what it applies to

191

metaphorically."36

And it is this question that Goodman rejects.
Notice, however, that this unanswerable question arises only from the
discussion of metaphor as comparison, comparison in terms of similarie
ties, and similarities in terms of properties. The fact that the at«
tempted explanation of how metaphorical use reflects literal use
results in unanswerable questions should constitute a reductiooad—
absurdum of that explanation. Rather than a complete explanation with
assurance that no further questions remain, Goodman's discussion serves
better as an argument against viewing metaphor as comparison.‘ When an
explanation leads to an unanswerable question,it does no good t0‘claim
that 'there is no real question,‘ particularly if that question is the
one which initiated the'attemptedexplanation.37 Whatis required is
that one attempt some other explanation.

Goodman's argument, by way‘of'error, doesindicate an important
aspect of metaphor. Goodman claims "The question why predicates apply
as they do metaphorically is much the same as the question why they

38"It'is, of course, the case that the

apply as they do literally".
questions are much the same (they differ only in the occurrence of one
word). But from this, Goodman infers that the answers to those ques«

39 It may be the case, as he suggests, that

tions will be much the same.
no general statement will explain what similarities things must have to
be denoted by a single predicate: But there is, on a different level,
an answer to the question of how in literal use, predicateS‘apply to
objects. That answer is that‘the literal use of predicates iS'a matter

of convention. If we can speak a language, if we are familiar with its

192

conventions we can apply predicates (in literal use). Should we, however,
encounter a word for which we did not know the conventions we would be
unable to apply it. We could only guess at its meaning. This explana«
tion is of course trivial. But this trivial explanation does not apply
to metaphor. The reason a predicate in metaphorical use applies as it
does is not explained entirely as a matter of convention. One of meta-
phor's key features is transcendence of convention in some respect.
Nonetheless, one must be familiar with conventions to understand the new
metaphoric use. But we do not need to learn a new set of conventions to
understand metaphor. We are able to determine the new reference, that is,
the reference of the term used metaphorically, based upon our knowledge
of the reference of the literal use of that term. It is obvious that
the application of a predicate in metaphoric use must be explained on
some basis other than, or in addition to, familiarity with conventions.
The fact that a trivial answer will satisfy one question (even if only at
an elementary level) but will not satisfy another is an indication of
a significant difference between these questions. Given that the same
explanation will not suffice for metaphoric use, we are still left with
the question of the relationship between the metaphorical and literal
uses of language.

Goodman concludes his discussion of this issue on the following note
of finality:

At any rate, the general explanation why things have the prOperties

liifi'filealf‘.m?§'§“‘é§lfia}’.ih‘éfinififii ‘33 lilie'ta iii 22133033630
But this comment surprises us since it seems that the question has been

reversed. Our question began as one concerning the relationship of

193

literal and metaphoric use (and perhaps how the literal and metaphoric
use of predicates makes reference to the properties which things have).
We find that it has now been reversed. It is Goodman's analysis of

this issue which leads from consideration of the use of language to

the consideration of cosmology. We agree that such questions ought to
be left to a cosmologist. But since Goodman's analysis has led to a
dead end (one might have expected this since he examined only dead meta-
phors) we recommend that some other analysis be attempted which avoids

cosmological questions.

Conclusions with Regard to Goodman's Theory

The conclusions resulting from Goodman's theory are essentially
advice to avoid certain problems. First, we must take care in using the
adjective 'metaphoric'. It is all too easy to avoid the serious ques«
tions and provide what appears to be an explanation by modifying the
terminology employed in the philosophy of language with that adjective.
One might, for example, be tempted, as was Goodman, to say that a meta—
phor is true if and only if the subject falls within the metaphoric
extension of the predicate. But such a move only pushes the question
back to the question, how does one know which of all the'possible‘exteno
sions of the predicate is indicated by a particular metaphoric use?
Questions of metaphorical truth fare a similar result. What needs to be
explained is how metaphors can be true where truth is construed to be
one of two mutually exclusive and jointly exhaustive truth values;‘ Given
this requirement, the introduction of 'metaphorical truth' will be

either a circumlocution which could be eliminated, or it is the proposal

194

of another truth value. Neither of these will enlighten us with regard

to the original question. In our discussion, we shall be careful to use
'metaphorical' to modify 'use' and attempt to explain how the metaphor-

ical use of a sentence can be true.

Second, Goodman's work should also serve as a further caution
against basing our efforts upon a model which does not fairly represent
metaphor in general. We notice in particular that Goodman's example of
metaphor lacks those qualities which the poet finds most important,
particularly the limitlessness of possible interpretation. It is surpris-
ing that Goodman overlooks this since it is this characteristic which
often makes metaphor the nemesis of Philosophy. Even if our objections
to the theory in Lauguages of Art fail to hold, that work would fail to
account for an important (perhaps the most important) aspect of metaphor.
If examples are to be used as models, they must be chosen so as to
reflect the great variety in metaphor. One must then adopt either a
theory to explain them all or several theories to explain several dis-

tinct types of metaphor.

Elliott and Mew
While Elliott considers metaphor, he does not purport to offer a
complete explanation of metaphor. His article is written as a response

4] But in that response

to Arnold Isenberg's, "The Problem of Belief".
he considers the relationship between the interpretation of a metaphor
and the truth of the metaphor. Of particular interest to us is his
objection to Isenberg's contention that understanding is independent

of truth. Elliott says of Isenberg:

195

He maintains that if we wish to affirm or deny something which is

said to us, we must understand what is said before we can look

for truth or falsity. This shows that it is possible to under—

stand a poem without being concerned with its truth or falsity. 2
In reply to this argument, Elliott argues that in many cases of metaphor,
one cannot understand the expression without determining its truth.

Of the various interpretations of a metaphor, some will result in
truth, others falsehood or obscurity. According to Elliott, when we
attempt to determine the meaning of a metaphor, we attempt to interpret
it so that it is true. Elliott says:

We begin from a preliminary or initial grasping of meaning and

proceed to establish the 'full' or exact meaning which is the_

meaning of the poetic sentence. It is as having that meaning

that the poetic sentence is verified, but the process of verifi-

cation is identical with the process by which the meaning is

established.43
In considering a particular metaphor we arrive at an interpretation be-
cause that interpretation results in a true statement. On Elliott's
view, one assumes the truth of the metaphor and seeks an interpretation
to support that assumption. Elliott says:

. verification of the poetic sentences is accomplished in the
process of discovering their meaning. In all such cases, the veri-
fication movement is a necessary one.... In many cases the only
alternatives are truth and obscurity. Where alternative interpre-
tations are available, we normally prefer that which not only gives
a poem meaning but enables it to mean truly.44

The import of Elliott's position is that, if he is correct, asking for
the truth value of an interpreted metaphor is unnecessary because in
determining its meaning, we presupposed its truth. Elliott's remarks
were all to be applied to poetry. But it seems reasonable to consider
them as recommendations for understanding metaphor in general. We will

do so with the recognition that some of the criticisms of his theory

196

would not be justified if we limited his theories application to poetry
as he did.

There are several confusions in Elliott's work. The first is con-
fusion of truth with belief. He fails to clearly distinguish truth
from belief and speaks as though our opinion as to what is true were
always correct. At best, one will be able to interpret the metaphor
so that it appears to be true. A change in his explanation from state-
ments about truth to statements about belief would save, for a time, his
theory. His statement could then be revised from:

. we normally prefer that interpretation which not only gives
a poem meaning but enables it to mean truly,45

to

we normally prefer that interpretation which not only gives a
poem meaning but enables it to mean something which we believe
to be true.

On this revision, the interpretation will be based upon belief rather
than truth.

46 criticizes yet

Peter Mew, in his article "Metaphor and Truth",
another aspect of Elliott's theory. According to Elliott, we often
determine the sense of metaphor by recognizing its truth. If, however,
no true sense were perceived, we should be unable to determine its full
meaning. Mew says:

One obvious and crucial defect in Elliott's view is that it can

provide no explanation of how it is that certain false open meta-

phorical statements are not generally regarded as vague (rough,
imprecise). Of the world, Hamlet says:

... tis an unweeded garden
That grows to seed; (Hamlet, Act I Scene 2)

And of life, Macbeth says:

197

... it is a tale
Told by an idiot, (Macbeth, Act 5, Scene 5)

Perhaps these statements are true, but I know many people who
believe them false, yet who have never complained of their
vagueness, imprecision or obscurity.47
Elliott has failed to account for the fact, according to Mew, that some
people understand the meaning of metaphors, yet nonetheless consider
them to be false.
Mew, however, reminds us of a distinction that ultimately provides
the basis for our objection to his argument. That distinction is between
open metaphor and the closed metaphor. He explains this distinction,

making use of a couplet by Pope:

So well-bred spaniels civilly delight
In mumbling of the game they dare not bite.

The-Pope metaphor is an example of an open metaphorical statement
in that nothing is done explicitly, or implicitly through the

use of context, to specify the characteristics of civil behavior
which are thought to satisfy the same description of the spaniel's
behavior. An example of a closed metaphor would be:

(c) The heavyweight champion of the world is an elephant,
he's massive and trundles along clumsily wherever he goes.

In this second case, the specification of elephant characteristics
is made.48

The context of the closed metaphor determines which sense is intended.
The truth value of such metaphors under various interpretations will have
no bearing upon which interpretation is correct. Thus Elliott's position
is refuted.

Elliott has available a reply to this criticism. He could apply
Mew's distinction between the open and closed metaphor to support his

own position. This could be done by broadening somewhat the notion of

198

context to include matters other than immediate literary context.
Those examples from Shakespeare which were offered by New as counter-
examples are in fact closed metaphor. They are closed by prior inter-
pretation. The understanding of those who find them false is a bor-
rowed understanding. The sense of these metaphors was discovered by
persons who found them to be true. Had the people, mentioned by Mew,
come upon these metaphors in isolation without the assistance of some-
one who perceived their truth, they would not have discovered the mean-
ing. Having shown the counter-examples to be closed metaphor, Elliott
then needs only to weaken his claim so that it applies solely to open
metaphors.

While his explanation appears not to apply to the closed metaphor,
Mew's discussiun does not show that it fails to hold for open metaphors.
That is, it may still hold for metaphors which are still in need of
interpretation. But there is yet another objection to Elliott's theory.
It appears to lead to paradox. Consider the following pair of open
metaphors.

(22) 'Green ideas sleep furiously'

(23) 'It is not the case that green ideas sleep furiously'

(23) is the direct denial of (22). If the notions of true, false, and
negation are to maintain their usual significance then (22) and (23)

must have opposite values. But on Elliott's analysis, they are both to
have true interpretations. Faced with the choice of abandoning Elliott's
analysis or abandoning the convention use of 'true', 'false' or 'denial',

we choose to abandon Elliott's analysis.

199

It seems that Elliott felt that our choices were between truth and
obscurity asopposed to truth and falsity. It seems clear from the
last example that our choice is between a meaningful interpretation
(without regard for truth value) and obscurity. In short, we should
choose an interpretation which makes sense. It seems that Elliott
failed to realize that an interpretation which makes sense could be
false.

The appeal of Elliott's suggestion that we select a true interpre-
tation for the metaphor lies in our confidence in the poet. "When a
poet makes a statement which is obviously false we normally assume that
he did so intentionally and look for the point of the falsehood."49
Our assumption is that the poet's statements must be true in some sense.
We believe he has an understanding of some phenomenon and is attempting
to communicate that understanding to us. But in spite of our confidence,
the possibility that we or the poet are in error remains. It is possible
that the author is misinformed about facts or has made some other mis-
take. It is clear that an author could be incorrect. A correct theory
must therefore leave open the possibility of the falsehood of metaphor.

Even though Elliott presents an interesting argument, little is
established concerning the truth value of metaphors. His discussion
does lead us to realize that either from confidence in, or charity
towards an author we attempt to interpret metaphor so that we believe it
to be true. Failing at that, we attempt to interpret the metaphor so
that it makes sense. Our argument on the previous page establishes that
one has the choice between recognizing that metaphor may be either true

or false, or giving up some basic considerations of logic.

200

ers
Concern with truth often leads one to consider the logical relation-
ships holding between metaphors and other statements. Earlier in this

chapter, we criticized Goodman for failing to take such relationships

50

into account. An article by C. Mason Myers discusses the possible

ascription of truth value to metaphor by investigating such logical rela-
tionships. The result of his consideration is the claim:

In philosophical discussions it will generally be a mistake to
regard a statement that is only mediately informative about x

(a metaphor) as true or false of x in the ordinary sense of the
words. Rather we should speak of the usefulness or lack of use-
fulness of such a statement for the understanding of x.51

This claim advises against our earlier position that metaphors could be

either true or false. Myers' claim purports to have a logical basis

and invites consideration of the logical implications among statements

and beliefs. Myers relies heavily on the work of Henle, and Myers'

vocabulary of technical terms closely parallels that of Henle.
If S is (an immediately informative statement), it will imply a
determinate set of statements about some object of thought x, the
falsity of any one of which implies the falsity of S, but if S is
(a metaphor) then a given interpreter I] can be guided by S to
accept a certain set C] of statements about x and a second inter-
preter 12 can be guided by S to accept a certain set of CC of state-
ments about x such that C] can contain statements not in 2 and ,
vice versa, and where it is possible a member of C1 is incompatible
with a member of C2. If P is a statement in C1 (or C2), then the
falsity of P does not necessarily imply that S is a false sggte-
ment about x but only that I] (or 12) has been misled by S.

One notices that Myers, in characterizing the relationships involved in

literal speech, discusses implication; in discussion of metaphoric use

he does not mention that which is implied but rather that which people

will be guided to accept. What is not discussed is that which one will

201

be guided to accept by a literal statement. Consider the relationships
in such a case. They will be essentially the same as in the case with
metaphor. A literal statement would lead interpreter I] to accept a

set of sentences C1 and interpreter I to accept a set C2. And the

2
statements in C1 need not be consistent with statements in C2. In fact,
it may be that sets C1 and C2 are not themselves consistent (that is, a
statement may lead one to believe (unwittingly) contradictory things).
It would be the case that the falsity of a member of C1 or C2 would not
imply the falsity of S. But this provides no basis for withholding the
assignment of truth value to S. In short, that which people are led to
believe has nothing to do with logical relationships among statements
and can not form a basis for withholding the ascriptions of truth values
from metaphor.

If we draw the parallel in the other direction, namely, consider
those sentences implied by the metaphor, we still find no basis for the
intended distinction. The problem with determining which sentences
are implied by a metaphor is the result of the ambiguity (openness to
alternative interpretation) of the metaphor. Prior to engaging in a
discussion of the logical relationships of any ambiguous expression one
attempts first to resolve the ambiguity, that is, settle upon a single
interpretation. In Elementary Logic,53 Quine suggests resolving ambig-
uity as follows:

[some words] have the effect, in these examples, of allowing the

truth value of a sentence to vary with the speaker or scene or

context. Words which have this effect must be supplanted by un—
ambiguous words or phrases before we'can accept a declarative

sentence as a statement. “It is only under such revision that a
sentence may as a single sentence in its own right, be said to have

202

a truth value.... The methods of technical analysis will be

fashioned in conformity with the understanding that'a*statement

is a sentence which is uniformly true or uniformly false inde-

pendent of context, speaker, time and place of utterance.54
On reading Quine, we find the motivation for denying that sentences
which are ambiguous have truth values; ‘The"methods of technical analy«
sis are appropriate only to sentences which have "in their own right" a
single truth value. In line with our terminology we would say that
such a sentence would have the same truth value in any of its uses or
contexts. Such a requirement would, however, deny truth values to a
very large number of literal English sentences, which because of context
were not considered ambiguous. Such a move in discussing English has
the disadvantage of failing to correctly represent the facts of language
and its use. This move we take Myers (but not Quine) to be making.
Myers may be attempting to follow this move of formal logic, but has
failed to notice that the logician is primarily interested in formal
languages and techniques for their application; Myers is purporting to
deal with natural languages. Quine's move is not to declare that am-
biguous sentences of natural language have no truth value and are to be
typed useful or not useful; he rather directs us to resolve the ambiguity
prior to the attempt to apply the technical analysis. In this respect,
metaphor differs little from simple ambiguity or amphibole. To return
to the considerations of what sentences would be implied by metaphor,
we note that implication is a technical term of logic appropriate to the
methods mentioned by Quine. However such methods presuppose a resolu-

tion of the ambiguity of metaphor. That is, they presuppose the assign-

ment of a single sense to the sentence and further a sense which will

203

determine a unique truth value. We find then that the discussion of
implication, far from determining whether metaphors have truth value,
presupposes that a unique truth value has been determined. The methods
of formal analysis are applicable only after the question of unique
truth value has been settled.

We find then that Myers argument against the ascription of truth
value to metaphor fails. It fails because the formal methods which he
attempts are applicable only on the assumption that the conclusion he
reaches is false. We will argue with Quine that the uses or utterances
of statements are the bearers of truth values. He states:

Strictly speaking, what admit of truth and falsity are not state-

ments as repeatable patterns of utterance, but of individual

events of statement utterance. For, utterances that sound alike

can vary in meaning with the occasion of the utterance. This is

due not only to careless ambiguities, but to systematic ambiguities

which are essential to the nature of language. 5
and

Strictly speaking, as urged earlier, what admit to meaning and

of truth and falsity are not the statements but the individual

events of their utterances.56
In terms of logical relationships, then, metaphor is on an equal footing
with literal speech. The methods of classical formal logic are amenable

only to those utterances of statements whose meaning guarantees a

unique truth value.

Conclusions
In this chapter we have been primarily concerned with the question
of truth. In considering Goodman's view, we found an argument showing

that at least some metaphors are true. In summary, that argument is

204 ,

(1) some dead metaphors are obviously true, (2) the only relevant
facts to change since the metaphor was fresh is our becoming accustomed
to it, (3) therefore, those metaphors were true when alive. Establish-
ing that some metaphors are true is a first and necessary step in es-
tablishing a place in philosophy for the metaphor. But a second and
equally important consideration is the establishment of criteria by
which one can determine the truth of metaphor. We rejected Goodman's
attempts in this regard. We recall that he distinguishes metaphoric
truth from literal truth. We supposed that 'literal truth' was used
synonymously with our ordinary use of 'truth'. Goodman's introduction
of metaphoric truth provides no basis for determining whether metaphor
is true. Besides, metaphoric truth seems to run afoul of various rules
of logic. We should prefer any other explanation of the truth of meta-
phor over one which is incompatible with the assumptions of two valued
logic.

We found that talk of metaphoric possession of properties is of
no help explaining metaphor. While it is true that some metaphors assert
that certain objects have certain properties, we here use 'have' and
'possess' in the same sense as we would in describing literal speech.
We do realize that under a metaphoric interpretation, the objects or
qualities mentioned may differ from those mentioned, given a literal
interpretation. Nonetheless, it is not the case that a different sort
of possession is involved.

It becomes clear in considering Goodman's view that he failed to
take into account, the wide variety of interpretation to which any given

metaphor is open. While he does mention some difference in the ranges

’

205

of a predicate, his discusSion treats the expression aS‘though“it had
only two, a literal and a metaphoriC'range; Such a view would indeed
suggest that the truth of metaphor could be explained’by“saying that

a statement would be true if its subject were within one of the ranges
of its predicate. But this move fails tO‘explain“how’one*is to
determine the relevant range. That‘is,‘it”failS"to provide advice on
how one knows what the metaphor says; 'And”given“variou3‘interpretav
tions the metaphor will turn out alternatively true‘and”false. We are
increasingly aware as we study Goodman's view that the difficulty in
determining the truth value of a metaphor is based upon a difficulty in
determining its meaning. Elliott, however, gives us reason to recon-
sider a claim that the determinations of meaning must precede the
determination of truth.

Elliott's article emphasized the interrelationship between inter-
pretation and truth. While Elliott was not able to establish his point
that metaphors ought generally be given interpretations under which they
will be true, this advice may be sound in guiding our interpretation.
In considering his arguments, we found that in the interpretation of
metaphor one ought to seek an interpretation which has a truth value
(assuming metaphor purports to be a declarative sentence).

Myers' arguments against ascribing truth value to metaphor failed.
They failed because the methods he used presupposed that sentences have
a unique truth value.

Our overall conclusion is that at least some metaphors are true and

that they are true in the same sense of 'true' as are some literal

206

expressions. Similarly, some metaphors are false and false in the same
sense of 'false' as are some literal expressions. But determining the
truth or falsity of a metaphor depends in part on determining its mean—
ing. We suspect that once one understands a metaphor, that is, once
one has settled on one interpretation, that determining the truth value
of a metaphor will be the same as determining the truth value of a

literal expression.

10.
11.
12.
13.

14.

15.
16.

207

NOTES TO CHAPTER VI

. Nelson Goodman, Languages of Art (Indianapolis: The Bobbs-Merrill

Company, InET), 1968.

. R. K. Elliott, "Poetry and Truth", Analyses, January l967, pp.

75-85.

. Peter Mew, "Metaphorical Truth".

. C. Mason Myers, "Metaphor and Mediately Informative Expressions",

Southern Journal of Philosophy, Fall, l968.

. Goodman, 68.
. Ibid.

. Here it is worth remembering that we want to understand the use of

metaphor by people in general and are not especially interested
in those problems to which only a philosopher's distinction can
give rise.

, 69.

H

U" 0'
d do
O. O.

H
U"
..a

D.

Ibi

O.

This point was demonstrated by Beardsley in the previous chapter.

We presented those replies to defeat the position on its home ground
as well as to emphasize the point to be made presently.

This parenthetical remark indicates some awareness of multiple meta-
phoric uses. But the fact that no further comment is made in
this regard indicates that Goodman considers this to be of little
significance. But since there is an unlimited number of ranges,
one could never (with any degree of confidence) say as He does
that a statement is metaphorically false.

Ibid.

Such a theory presumes that all names have been eliminated from the
language and one is only concerned with predicates, variables
and logical words.

17.
18.
19.
20.
21.
22.
23.

24.
25.
26.
27.
28.
29.
30.

31.
32.
33.
34.
35.
36.
37.

38.

208

ijg,

191g,

ijQ,, 72.

1919,, 73.

121g,, 72.

191g,

Notice that such an explanation would be an elucidation of the

sense of that particular metaphor rather than metaphor in
general.

Assuming that one is clear about the use of 'realm' and 'schema'.

Ibid., 76.

H
U"
A

D.

o, 770

H
U
..a

D.

F!
U'
_u
D.

.9 77'78.

H

bi

O.

, 77.

We recognize that at least some metaphors are designed to reflect
similarities, not create them.

ijb,, 78.

Ibid,

ijg,

1219,, 76.

191g,, 78.

191g,

We recall that it was a need to discuss the relationship between

literal and metaphoric use which prompted the discussion in
terms of comparison.

Ibid.

39.

40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.

56.

209

Given this line of reasoning, one would expect“the"following‘ques«
tions to have much the same answer:“

'Why“did‘Germany lose

World War I'? and 'Why did Germany lose World War 11'?

Ibid.

Arnold Isenberg, "The Problem of Belief."

Elliott, 77.
ijg,, 81.
ijg,, 83-84.
191g,

3

8W.

0—:

bid., 193.

bid., 191-192.

Elliott, 79.

Myers.
Ibid., l66.

H

Ibid.
W. V. Quine, Elementary Logig (New York:
Ibid., 6.

W. V. Quine, Methods of Logic (New York:
1959), p. xi.

Ibid., xvi.

Harper & Row, l965).

Holt, Rinehart, Inc.,

CHAPTER VII

CONCLUSION

A definition of metaphOr is not forthcoming. A great variety of
figures of speech are referred to by people when they use 'metaphor'.
A definition which would fit one metaphor likely will not fit another.
C. Mason Myers provides an interesting comment on this point which
could stand as a summary of our investigation of other views. Observing
the great variety of metaphors, he goes on to say: '
... Thus it is not surprising that advocates of a number of dif-
ferent theories of the nature of metaphor have no trouble finding
examples which seem to fit only their particular theory. They
are right about the special class of metaphors to which their
examples belong, but wrong if they assume they have extracted an
essence belonging to metaphors in general.1
While we do not claim to have extracted the essence of metaphor (we are
not certain what that would be), we can discuss several characteristics
which all metaphors share. Metaphors are uses of expressions. by
'uses of expressions' we mean the occurrence of grammatically correct
strings of symbols of language within a particular context. The words
occurring in those expressions have meanings supported by convention
and/or custom. Nonetheless, metaphor is a use which goes beyond those
conventions. The meaning or sense which is associated with a particular
metaphoric use is in part, but only in part, dependent on convention.

It is dependent as well on a wide variety of relationships between ob-

jects, uses of language, feelings, and opinions. It is not dependent

210

211

on the recognition of any new set of conventions. An expression which
is used metaphorically appears absurd, contradictory or inappropriate
when interpreted literally. On the other hand, it is a use of words
which is generally understood within a more or less limited linguistic
community. Metaphors are therefore meaningful. Any use of words which
fails to meet this last requirement will be nonsense and not a metaphor.
While the foregoing is no definition, it is a correct and concise
statement of our position. 'Metaphor' is, as we suggested in the
initial chapter, an imprecise term and we must still guard against a
definition which is at odds with ordinary use. Even though we provide
no definition, conclusions from our investigation of other theories
provide analysis of the key issues involved in metaphor. We consider

those analyses in the remaining paragraphs.

Restriction of 'Metaphorical' to 'Use'

Since our analysis of other views is so negative in general, it is
fitting that we begin negatively. It is tempting to use 'metaphoric'
and 'metaphorical' quite widely. It is particularly tempting to use
these words as modifiers of 'truth', 'referent', 'meaning', and other
terms generally associated with language. In view of comments made in
discussing Goodman's work we refrain from such use. Generally,
'metaphoric' and 'metaphorical' modify 'use' or 'interpretation'.
'Metaphoric expression' is best regarded as a device for avoiding the
circumlocution 'an instance of an expression used metaphorically' or
perhaps 'an expression whose uses are generally metaphoric'. We shall,

on occasion, speak of metaphors when it is apparent that we are

212

referring to expressions rather than their uses. On such occasions,

we shall be referring to expressions whose use would generally be
recognized as metaphoric. We do not desire to depart from the ordinary
use of language. Since 'metaphor' is often eonstrued as referring to
expressions, rather than their uses, we show how, in the light of our
position, such a construal makes sense. We do not recognize two 1
radically different kindS‘of“sense"sometimes‘spoken“of"by“others as
'literal sense' and 'metaphoricrsense'; 'The”expression“imetaphoric
sense' is best interpreted as ‘the‘sense’of“an"expression“which is

used or interpreted metaphorically'; it should'notw e interpreted to

a separate kind of sense. We recohnize that metaphor doe3"not‘constie
tute a separate language'resting onsa unique type“of“meaning“associated
only with metaphors. It generally occurs within the context of literal
speech and its meanings are interrelated with those of literal use.
Some metaphors are meaningful and true; and when we so state, we in-
tend the terms 'meaningful' and 'true' to have the same significance
that they do for the literal use of language. A similar motivation is
behind our decision with regard to meaning and reference. Although the
interpretation of metaphor requires us to consider matters other than
conventional meaning, we maintain that the metaphoric use of language
is meaningful and that some metaphoric uses refer. We insist that
metaphors can be truthful and sensible in the same sense as literal
statements. Not only does this position help explain metaphor as a part
of single language, it avoids the problems generated by a two truth or

two meaning theory.2

213

No Examples as Paradigms

A second negative point suggested by our"variou3‘analyse5‘13'that
we avoid basing a theory of metaphor onra‘single“examplex"An“example
will be even less likely'tO'characterize“metaphor than will a‘definie
tion. As with examples in other areas, examples of metaphor will sug-
gest simultaneously explanations which are too broad and too narrow.

An explanation which provides a basis for understanding one metaphor
may be totally inappropriate for another. We, therefore, take no
examples of metaphor as paradigm.

Since the literature concerning metaphor is replete with hasty
generalization we are careful not to generalize rashly from examples.
Those points we do draw from examples will be points claimed true for
only some metaphors. Arguments by example will generally be used as
counter-examples to universal claims.

We notice that those theories taken primarily from examples tended
to treat such metaphors as the only type of metaphor and attribute their
characteristics to all metaphors. We are convinced that there are as
many types of metaphor as of literal uses. Many of the views correctly
explained some aspect or some type of metaphor. The comparison view,
for example, is correct for some metaphors.

The claim that all metaphor is comparison may derive from the
explanation of our understanding of comparison metaphors (that is, meta-
phors which state comparisons). Our explanation of the understanding
of metaphor will not depend on comparison; however, comparison will

still be available as an explanation of some metaphors.

214

Recognition of Metaphor

Metaphors do not generally have labels. This rather obvious fact
is overlooked by several writers who proceed to explain how one handles
metaphor without explaining how it is that one recognizes a metaphor.
If, as in the model presently to be discussed, interpretation is con-
strued as a relatively mechanical procedure, then an explanation of our
ability to recognize metaphor is of great importance. One would not
apply the procedure unless he thought that he was dealing with a meta-
phor. An explanation of our ability to understand metaphor must include
an explanation of our ability to recognize metaphor. An explanation
which misses this point is considered by Black in his discussion of the
'comparison view.‘ He provides directions to apply a formula to meta-
phors to determine their meaning. Such an application however presup-
poses that one is aware that the use of language under consideration is
metaphoric. Any such explanation fails because recognition of metaphor
often does not precede an understanding of it. That is, one often does
not recognize that the use of an expression is metaphoric until he under—
stands that use. And one may understand a metaphor without recognizing
that it is one. (This would be most likely in the case of one who en-
joyed literature but had escaped formal training in interpretation.)
Two considerations show that one often recognizes cases of metaphor only

3 First, consider Beardsley's point

after he understands the metaphors.
that the use of an expression is to be considered a metaphor only if it
is not nonsense. The only assurance that an expression is not nonsense
is that someone understands it. 'An intelligent individual is not likely

to be assured that the use of an expression is not nonsense unless he

215

understands it, whether or not it mustfbe explained to‘himz‘”Therefore,
in such cases, one must understand“the use of an”expression“before he
will recognize it as metaphor.

Second, it is only after understanding'a“figure of speechrthat we

are able to distinguish‘metaphor‘from‘other‘figuressucbfasmyperbole4

or irony.5

We are able to distinguish these figureS'from“metaphpr due
to certain facts concerning their meanings.

Two essential factor5°are“t0“be noticed; The first i3“tbat the
recognition of metaphor is not‘a‘matter'of‘syntax: There“is not for
example, a metaphor-mark“as*there'is‘a‘question‘mark; “Whiiermemaphorical
and literal uses must be grammatically correct, no syntactical devices
are generally employed to distinguish them from each other. A second
factor is that our recognition of a use of language as metaphoric is in
part a matter of our recognition of its sense. Any explanation of meta-

phor which attempts to deal with either recognition or interpretation

independently will fail.

Recognition of Figurative Use
While there is no general test for metaphor prior to its under-
standing, there is a cue by which we are alerted to possible figurative
use. That cue is described by Henle:
One ordinarily takes words in their literal sense and this is
impossible in metaphor. This impossibility is in fact what drives
one to seek a figurative sense.
While 'impossible' is too strong a word, it is the inappropriateness of

the literal interpretation which causes an abandonment of that interpre-

tation and initiates a search for a figurative interpretation.

216

The inappropriateness of the literal interpretation may be due to a
contradiction in that interpretation or a recognition that such an
interpretation would result in absurdity in the given context. In
either case the ineptness of a literal interpretation indicates that
(in the absence of an apt literal interpretation) only a figurative
interpretation will be suitable.

In considering the inappropriateness of the use of an expression
we are reminded of Henle's point. He noticed that a wider context than
the sentence itself must be taken into account in interpreting some
metaphors. Since many such linguistic uses are recognized as metaphors
in this wider context, we shall consider such context a part of the

metaphor.

Context

'Context' has occurred in other views.previously treated. Richards,
for example, argues against Eastman's theory concerning impractical
identifications and bases his argument on the observation that certain
contexts make identifications less impractical than they might other-
wise appear. Black's discussion of frame and focus is equally a discus-
sion of context. Our own consideration of context will follow Black's
lead in considering the mataphor to be composed of both focus and frame.
We find, however, that Black construed the notions of focus and frame
too narrowly. Because of this narrow construal, Black was unable to
explain some of his own examples of metaphor.

Black considers a single word as the focus of the metaphor, the

remainder of the sentence the frame. Black noted that "... the presence

217

of one frame can result in metaphorical use of the complementary word,
while the presence of a different frame for the same word fails to

result in metaphor."7

With this we agree. The status of an utterance
depends on more than a single word within that utterance. Black's
discussion of frame was, however, limited to the sentence in which the
focus occurred and consideration of focus was limited to a single word.
It seems that both these notions can be expanded and we expand them in
the following way. Focus includes the expression which is used meta-'
phorically while frame includes the context which makes that use meta-
phoric. We agree that in his example:

The chairman ploughed through the discussion.
the focus is 'ploughed' and the frame is the remainder of the sentence.
But in the case of some metaphors the focus is not limited to a single
word nor is the frame a single sentence. Consider the sentence:

His heart is not in the right place.
If this sentence were to occur in a novel about a cardiac patient, and
further, occurred during the recounting of-a heart transplant, then,
while we may be concerned for the patient, we would find nothing meta-
phoric. If the novel concerned a wicked man who beat his children, we
would recognize the metaphor here. The literary context in these hypo-
thetical cases determines that one use of the sentence is metaphoric
while another is not. Often the relevant context is not literary. It
is rather the actual circumstances under which an expression is uttered.
Considering the same example; if the phrase were actually uttered in the
consideration of an E.C.G. report it would be interpreted differently

than at a discussion concerning the United Fund and only one of these

218

would be considered metaphoric. Not only physical surroundings, but
the "system of associated common places" of Black,8 or Henle's "minor
facets of the general culture, such as what characteristics are upper-
most in people's mind"9 must be considered as a part of the frame or
context.

If we consider these factors as part of the frame of some metaphors
we can discuss an example of which Black said "I shall have to neglect

"10 Given a wider notion of frame and

such complexities in this paper.
focus, the complexities of his example can be easily dealt with.
Black's example is:

Light is but the shadow of God (Sir Thomas Browne)

Light is often associated with hope, joy, understanding and happiness.
Since light is God's shadow the possession of these qualities by God
must be incomparably great. Someone of course who understands light
and relationship between light and shadow differently may interpret
the line quite differently. Nonetheless, it is the frame or context,
including general attitudes and beliefs which account for our under-
standing of this and other metaphors. All of these must be considered
a part of the frame.

We invite the reader to notice that the inclusion of both the focus
and frame as parts of metaphor is consistent with our explanation of
metaphor as use. The use of an expression is in part determined by the
context in which that sentence is uttered. So, even given our wider

notion of frame a difference in frame will result in a difference in use

and therefore a difference in what metaphor is used.

219

Confidence in the Speaker

While our examples indicate that the consideration of context will
be a requirement for understanding metaphor, context alone explains
neither our desire nor our ability to understand metaphor. Recall
Black's question regarding one's motivation for providing his readers
with a puzzle to solve. We now turn this question around to inquire
about the reader's motivation for solving the puzzle. We noted that
puzzle solving itself may be a rewarding experience. But metaphor often
occurs when one is not at all interested in playing such games. Some
other reason for our attempt to understand metaphor must be found.

Our willingness to attempt the interpretation of metaphor is based
in part on our confidence that the speaker has something worthwhile to
communicate. It will not matter whether that value is cognitive or
emotive. We agree with Richards' argument against Eastman that we would
soon grow tired of bafflement. Eastman perhaps believes that we are
mountain climbers who accept the challenge because it's there. But just
as the existence of mountains is not sufficient motivation for most of
us to climb, the mere existence of figurative language is not sufficient
to motivate interpretation. That motivation has as its base Henle's

principle that people generally try to talk sense.

Relevance of Conventional Use

Desire to understand another may explain our motivation to cope with
metaphor. It does not, however, explain our ability to understand meta-
phor. The major appeal of the relational theory as stated by Henle was

an explanation of that ability. According to his view, one gains

220

understanding of a metaphor by the following procedure. He notices the
customary referent for one of the terms of the expression and substitutes
for it an object which is similar. Considering the new object results in
a meaningful expression. Such an explanation, as we discussed earlier

is at best suitable for a limited type of metaphor.

Some modification of the relational view will increase its range of
applicability. Our major criticism of this view is that it depends upon
relationships between objects, and the understanding of some metaphors
does not depend upon such relationships. The obvious suggestion would
be to drop the discussion of resembling objects. In so doing, we could
say that when one encounters a metaphor, he picks a meaning which makes
sense in the given context. But such an explanation fails. One could
satisfy its requirements by deciding in advance to interpret every meta-
phor as synonymous with:

Excuse me, can you direct me to the restroom?

While such a remark would not make sense in every'context, it would be a
safe bet in many. This objection illuminates the fact that in our pro-
posed explanation the meanings of the terms in their literal use is not
taken into account and, as we argue in our discussion of Eastman, those
meanings are relevant. We must at least modify our explanation by saying
that one picks as a sense, something that makes sense, taking into
account the conventional meanings as well as the context. This rather
obvious remark would find opposition from Foss. We recall that he holds
that in metaphor symbols lose their familiar meaning; their meanings are
destroyed. But, as our argument against Foss established, if the indi-

vidual terms of the expression had lost their conventional sense,_

221

metaphors would not be meaningful. Our understanding of a metaphor is,
in part, due to our understanding of the conventional uses of its terms.
What we should like to say is just how conventional meaning is taken

into account. In our criticism of the comparison'views, we suggested
several ways of taking conventional meanings into account‘other than
comparison or similarity of their referents.” Such other bases are our
feelings towards the things mentioned, our beliefs about those things,
and similarities between the meanings of the various terms; ‘Yet this is
not a complete list of possibilities.' It seems that Henle, while less
precise, may be more correct in explaining our basis of interpretation

as 'characteristics uppermost in peoples' minds.‘ The factor most
responsible for our interpretation of metaphor is what comes to mind when
the conventional meaning'of the terms in the metaphor'are‘considered.
Context is here relevant in providing a mental set which may direct our
thoughts as well as providing a test of sensibleness of the interpreta-
tion. Why a linguistic community would be reminded of, or experience the
feelings, objects or events they do could only be explained on“a case to
case basis. Nonetheless, the ability to communicate via'metaphor is
based upon such similarities among the audience and the‘speaker.' Communi—
cation through metaphor will be successful only when the sorts of objects,
events, feelings, etc., suggested to the author are similar to those
suggested to the audience. Only in these cases will the metaphor be
understood in the same way by author and audience. Notice that we have
not spoken of determining the real or true meaning of a metaphor, nor

has there been talk of correct interpretations.‘ We find that while

these notions may make some sense“in the discussion of‘literal usage,

222

that sense is inappropriate in the consideration of metaphor. When one
speaks of a true meaning or correct interpretation he is making refer-
ence to conventional usage. An interpretation is considered correct
insofar as it is in accordance with such conventions. Metaphor is
however, in part, a departure from convention. Therefore given the
usual understanding of true meaning or correct interpretation, metaphor
will have neither of these. We therefore make no attempt to explain
them. The remaining question in this regard, "how is communication
possible?" was answered above. 'When the set of associated commonplaces
in the minds 0f the audience and speaker are sufficiently similar, they
will interpret a metaphor similarly., .

We recognize, too, that communication is not the only goal of meta-
phor. Often the value of the metaphor lies in the suggestion of a
relationship which the audience must follow out and fill in. Its value
in literature in this respect is that the same line in the same work
need not have the same significance at each reading and to each audi-
ence. This is in part what accounts for the notion that a good metaphor
is lively. In reading literature which contains such metaphors one is
able to see different things with each reading.

It is in great part this feature of metaphor which, while creating
great literature, causes metaphor to be abhorred by the philosopher.
The philosopher, in his search for*truth, demands a language of precise
meanings and meanings which do not fluctuate with context.‘ Further, he
demands a use of language which is amenable to critical analysis and
refutation. Metaphor will allow the slippery opponent to say 'that's

not quite what I meant.‘ While we make room for this sort of liveliness

223

in our discussion, we are still most concerned with metaphor's useful-
ness in philosophy. We turn therefore to a more central issue, that of

the truth of metaphor.

Truth of Metaphorical Statements

Explaining our ability to recognize metaphor as being, in part,
a matter of recognizing absurdity leads us to question the possibility
of ascribing truth values to metaphor. It is clear that that which is
absurd is surely not true. However, man is able, by the use of meta-
phorical expression, to communicate useful information. But, if meta-
phors are false, it is paradoxical that they could be so communicative.
Consider the following example:

1] It is

(1) "His grandfather is a child when it comes to money."
not possible for anyone's grandfather to be a child; sentence (1) is
false. But we know from this sentence that the grandfather does not
keep good accounts of his money and that he does not spend it wisely.

That some information is provided indicates that there is some sense in
which (1) could be true. Our attempt to interpret metaphor is directed
towards providing an interpretation under which it will be true. But it
must also be possible for (l) to be false in the same sense. Consider
the following example:

(2) Martha Mitchell has the mouth of Martha Rae.

This is a false statement, if literally interpreted. However, (2) does
indicate something considered by some to be true, that Martha Mitchell's
l2

manner and topic of speech has caused embarrassment to those around her.

On the other hand

224

(3) 'Senator McGovern has the mouth of Martha Rae.‘
is false, both conventionally interpreted and metaphorically interpreted
(at least in the sense in which (2) is true). Given that we make this
distinction, a theoretical basis for it may be established.

We have been concerned with the distinction between literal and
metaphorical language, and those problems unique to metaphor. The ques-
tion of truth when applied to literal speech is problematic. There is,
at least, no universally accepted theory of truth. Rather than con-
sidering theories of literal truth, or the general problem of truth, we
will assume that some satisfactory explanation is available. Further,
we shall assume that such an explanation would closely relate truth
and meaning in such a way that given the state of the universe, a state-
ment with a particular meaning would have a particular truth value.
Given such an assumption, it seems that the special problem of truth
of which a theory of metaphor must take account is the problem of deter-
mining the meaning of a particular utterance when such an utterance
constitutes a metaphorical use. If we can show how to determine the
meaning of a metaphor, we will assume that a similar relationship holds

between meaning and truth, as holds in the case of literal speech.

Literal Equivalents and Truth

The question of literal equivalents is tied directly to the question
of truth, on assumption that the problem of truth has been solved for
literal statements. If every metaphor has a literal equivalent, even if
cognitively and not emotively, then the answer to the question of truth

becomes easy: the truth value of the metaphor is identical to the truth

225

value of its equivalent. However, in our discussion of Black we argued
that no serious theorist held that every metaphor has a literal equiva-
lent. Our intention here is not to support such a view. We hold
neither the view that every metaphor has a literal equivalent nor the
view that no metaphor has a literal equivalent. But our position in
this matter is not the result of a lack of sufficient consideration.

We hold neither of these views because we deny them both.

Both Wheelwright and Black set forth views that metaphor is indis-
pensible without loss of expressive power since metaphors (in Black's
case interaction metaphors) have no literal equivalent. But neither
Wheelwright nor Black establishes his claim. Each, in fact, offers lit-
eral equivalents for various examples of metaphor.

We shall consider one further attempt to establish that metaphors
have no literal equivalent. Abraham Kaplan, in "Referential Meaning in

"13

the Arts, attempts to refute the view that metaphors have literal

equivalents. He presents that view as follows:

.. it [a view that all metaphors have literal equivalents] treats
metaphor as having a type of cognitive meaning differing from non-
metaphoric reference only in complexity. Metaphor is compound
reference, referring to its tenor by way of its vehicle. The sense

‘of the metaphor could be given directly, and it is this sense
which is important. But the literal statement would not have the
psychological effect of the metaphoric one. Poetry can be, and in
critical practice should be paraphrased; the paraphrase will lack
only the ornamental and emotional appeal of the original.]

In objecting to this view Kaplan offers essentially the same objection as
does Black. Both, in recognizing the ambiguity of metaphor, claim that
a paraphrase will fail to say precisely those things said by the metaphor.

Kaplan states:

226

Of course to describe Richard of England as a "lion" is to say

that he is brave and kingly. But it is to say a great deal else

besides. And when enough paraphrases have been added to exhaust

the content of the metaphor, all of them taken together now are

likely to say too much--each phrase has its own senses not involved

in the metaphor, in addition to those appropriate to it.15
But Kaplan, like Black, overlooks the possibility of various disjunc-
tions of phrases combined with indications of how seriously each of them
is to be taken. We still have no conclusive argument that some such
device will not work. Further, we notice that Kaplan's conclusion is
itself weak. He says, "we are now likely to say too much." This does
not rule out the possibility that with a sufficient amount of study and
attention an equivalent literal statement could be devised.

In considering Beardsley's work, we examined several arguments with
regard to paraphrase, where paraphrase was interpreted as producing a
literal equivalent.

Arguments on both sides were inconclusive. No sound argument had
as its conclusion 'no metaphor can be paraphrased' nor was it established
that every metaphor can be paraphrased. A similar question could be
raised with regard to the literal use of language: namely can every
literal expression be paraphrased? This question too, appears to be un-
resolved. Here we are tempted to ask: 'By whom and under what circum-
stances?’ We, on occasion, find ourselves called upon to explain some-
thing which we have said by providing a paraphrase. And we find that we
cannot construct a paraphrase of our expression even though we may offer
some explanation of our meaning. In being unable to paraphrase we may
find some difficulty in getting another to understand our original state-

ment. We do not however, assume that one's inability to paraphrase is

an indication that he is confused, does not know what he is talking about

227

or that his words are without sense or truth value. But this is often
thought to be the case when considering metaphor. Stewart expresses
clearly this position which may be at the heart of the philosopher's
concern with paraphrase.

... given that the new meaning of the word cannot be stated or

explagnfigaggngxgggliEgedeinsznggher expression it is likely that
If, when dealing with literal usage, an explanation is provided which
allows us to make sense of an utterance, even though no paraphrase is
provided, we assume that it was not nonsense. Similarly, the absence of
a literal expression with the same meaning as a metaphor is not a suf«
ficient ground for declaring it to be meaningless. Paraphrase in lan-
guage is based on an accidental, not necessary feature of language.

That feature is the redundancy of many of its expressions. That a
particular expression has an equivalent is a fortuitious matter and has
no bearing on whether the expression is meaningful.

In light of these facts, one wonders about the concern over para-
phrase of the metaphoric use of expressions. In claiming that a rela—
tion exists between literal equivalents and truth we are doing no more
than claiming that two synonymous expressions have the same truth value.
Even were it the case that no literal expression is equivalent to any
metaphor it would not be established that such metaphors had no truth
value. Many literal uses of language have no literal equivalents yet
their truth value is clear. The appeal to literal equivalents to solve
the question of truth in metaphor will fail since it cannot be establish-

ed that every metaphor has a literal equivalent.

228

Paraphrase and Understanding

We indicated, in a criticism of Beardsley, that our interest in
paraphrase of metaphor could well be that we do not fully (or perhaps
even partially) understand some metaphors. We look to paraphrase for
an explanation of a metaphor's meaning. We are also interested in
literal paraphrase as a check on whether another is uttering nonsense.
(In philosophy we often claim literal utterances to be nonsense.)

A key difference exists between attempting literal paraphrase of
a literal use and a similar paraphrase of a metaphoric use. That dif-
ference is that in considering conventional use, our misunderstanding is
likely to be due to our lack of familiarity with various conventions.
In such cases the O.E.D. may be of great assistance. Perhaps a good
thesaurus may be of help. These tools often provide information about
conventions and synonymous expressions which are the basis of the con-
struction of a meaningful paraphrase. Of course if one did not under-
stand the original expression, one will not know if this paraphrase is
precisely synonymous. One is satisfied with a paraphrase that makes
sense and would have been appropriate in the context. In addition, the
reference to standard lexicographic sources provides an argument for
the correctness of the interpretation. Metaphor is different. We
assume that one's difficulty in explaining a metaphor is not with the
understanding of the conventions of language and:therefore the O.E.D. and
thesaurus will be of no help. While we recognize that the sense of a
metaphoric use rests in part on convention, convention will not determine
what the metaphor says. Therefore the tools based on these conventions

will not generate paraphrase. Of course, the availability of the O.E.D.

229

does not guarantee our success in paraphrase of a literal expression nor
does its disutility guarantee failure in the case of metaphoric expres-
sion. While the O.E.D. may suggest possible interpretations of literal
expressions, it is the fittingness of one of them that governs our final
choice. In the case of metaphor as in the case of literal use the
fittingness of the paraphrase is our only guarantee of correct para-
phrase.

A significant difference between literal and metaphoric use is meta-
phor's openness to alternative interpretation. This feature is not of
course unique to the metaphoric use of language. But it is this feature
which is one source of the life of the poetic metaphor and one of the
reasons for its use. Metaphor is 'live' because it remains open to a
great many interpretations. It remains interesting because one is able
to find meaning in poetic metaphor which was overlooked earlier. This
is the basis of the argument of Kaplan and perhaps the source of
Beardsley's comments concerning that portion of the metaphor which is
understood. Here the argument against paraphrase is an argument that no
paraphrase will be open to precisely the same interpretations as will
the original. And it may well be that we will have no guarantee that a
paraphrase had included all the meanings which will ever be discovered
in the original. Poetic metaphor is designed for this purpose. When a
metaphor dies it becomes conventional. But in considering poetry the
term 'dead' applies in quite the opposite way that 'live' was said to
apply. The repeated use of a metaphor tends to center upon one of its

possible senses, others cease to be considered and since the metaphor

230

becomes in this way a conventional word alternative interpretation is
not invited. Consider:

Richard is a lion
We would not think that this could mean that Richard is cowardly, lazy
(in letting his wife do the hunting and provide for the family) or bad
smelling. But all these are qualities of lions which are easily
attributed to humans. But 'lion' is a conventional term indicating
that an individual is a very brave or strong person who may possess other
kingly qualities. By custom this term no longer suggests the other
possible interpretations, except perhaps as a joke.

Metaphor not occurring in poetry shares this openness to alterna-
tive interpretation. Rather than an advantage as in poetry, openness
results in difficulties in philosophy and other studies where truth is
of the essence. But in these cases we expect that the speaker has
something specific to say. One of our interests in attempting to para-
phrase is to find out which of several things a speaker said. If this
question were asked of the poet he would probably answer, if at all,
that he said all of them and more. But notice here that it will matter
little whether our paraphrase is considered a literal or metaphorical
use of language. Such paraphrase is of value, in the absence of the
author in keeping straight the various possible interpretations for

discussion with others and for decisions on our part.

Metaphor vs. Ambiguity

 

Talk about openness to alternative interpretation suggests that,

in some respects, metaphor may parallel literal ambiguity. It will be

231

worthwhile to briefly consider ambiguity to determine how it differs
from metaphor, since such a consideration may indicate a method of
determining meaning in the case of metaphor. Ambiguity can be summar-
ized as follows: an ambiguous word has more than one distinct use and
it may not be evident which use is intended; an ambiguous word has
more than one extension and it may not be evident which extension is
intended. Ambiguity is governed by convention, and its various uses
are generally few and well isolated. Since the uses are few and con-
ventional, it is not generally difficult to determine, by the use of
context, the intended meaning. Metaphor, on the other hand, goes beyond
convention, and the possible senses of metaphor are not restricted.
And in the case of poetic metaphor, even context may not determine a
single correct interpretation. In the case of poetry this is an ad-
vantage. But outside of poetry we often desire a single sense to con-
sider. In such a case, the departure from convention and the accompany-
ing departure from the limitations of those conventions makes the task
of selecting a single interpretation considerably more difficult. To
exhibit this difficulty, we consider two examples; the first is an
example of ambiguity, the second, an example of metaphor.

(4) 'Sam jones will run.‘
We can reduce our number of possible interpretations to two. The state-
ment either says:

(5) 'Sam Jones will be a candidate' or

(6) 'Sam Jones will, on foot, 90 faster than a brisk walk'
The context, whether one is talking about how fast Sam will move, will

decide between (5) and (6). But if we consider

232

(7) 'Richard is a lion'
as it might have been as a live metaphor, there were an unlimited number
of interpretations.

(8) Richard is a king.

(9) Richard is brave.

(l0) Richard is yellow.

(ll) Richard needs a haircut.

(l2) Richard devours his neighbors.

(13) Richard has offensive body odor.

(l4) Richard requires his wife to support the family.

(15) Richard is fierce.

(16) Richard roars.

(17) Richard goes about naked.
Context will be of assistance in eliminating some of these interpreta-
tions. Some interpretations may be inconsistent with others and context
may isolate some interpretations as incorrect. But as in (8) and (9)
above there may well be no basis to select one, the other or a conjunc-
tion of the two. This suggests another difference between ambiguity
and metaphor. The selection in the case of ambiguity involves choosing
only one of the meanings. In (4), for example, Sam Jones is running a
lOO-yard dash or he is running for office but he is not doing both. To
intentionally-confuse the meanings (that is, to see both meanings as
being correct) is a joke, a pun. In the case of metaphor, many of the
possible interpretations can be viewed as correct at the same time with

no joke intended. Indeed, we recall that Henle says, 'Thus the aptness

233

of metaphor depends on the capability of elaborating it--of extending
the parallel structure.‘17
Choosing an appropriate interpretation of metaphor bears some
similarity to resolving literal ambiguity. In both, one considers
the context of a statement and selects a meaning which makes sense.
Metaphor may not so easily be resolved as ambiguity, since often a
wider variety and greater number of interpretations make sense.
However, the system of associated commonplaces is often effective in
reducing the number of reasonable options. Nonetheless, the final

determining factor in selecting an interpretation will be the sensible-

ness of the contemplated interpretation.

Final Note on Truth

To return to the question of truth, we find that although the
specification of a literal equivalent would solve the problem, there
may be no cases in which metaphors have such equivalents. This does not
bear upon the wisdom of ascribing truth values to metaphor. We notice
that with literal speech, understanding of the use of an expression is
sufficient (together with the knowledge of several facts) to determine
its truth value. There is no dependence upon literal equivalents. One
does not attempt to paraphrase a sentence to determine whether it is
true; the existence of another synonymous series of signs of the lan-
guage is fortuitous and irrelevant. With literal speech we use para-
phrase to get someone to understand a sentence or to agree with our

interpretation and therefore, in view of a common understanding of the

234

relevant facts, share our ascription of truth value. In either of these
situations, a literal equivalent is not required as a paraphrase. In
this regard, metaphor is no different from literal usage. Once one
understands the meaning or meanings of a metaphor he will be able to
assign truth values appropriately (if he also is aware of certain facts).
He needs make no recourse to literal equivalents. Such recourse, as is
the case with literal speech may be helpful in inducing another to share
that understanding or in an attempt to determine which of several things
may have been intended. But equivalents are not required to guarantee
that one's expressions are meaningful nor are they required for an

expression to be truthful.

Metaphor in Philosophy

 

Our most important question is whether metaphor may be used in
philosophy. Turbayne has emphasized the sorts of errors into which
philosophers have fallen when employing metaphor. But we recognize that
these errors stem from a misinterpretation of those metaphors. In par-
ticular, Turbayne points out the following sort of error. A philo-
sophical position may include metaphor (perhaps because no literal
expression seemed to convey the intended theory). In evaluating the
position one errs by attempting to interpret the metaphor as though it
were a literal use. Such an error would result in misunderstanding the
position. Ones arguments would not be, of course, in opposition to the
view stated but probably against a straw man. This error would, as we

suggested earlier, amount to a fallacy of equivocation. But this sort

235

of difficulty is at most a warning and is not a prohibition. The fallacy
of equivocation is nothing new.

We sum up our position on this matter with the following analogy:
Making a philosophical point with metaphor is comparable to driving
a finishing nail with a large pipe wrench. It is possible to drive
nails in this manner, but many will bend. If one had a hammer, he would
be advised to use it. Similarly, we have argued that metaphors are
meaningful and may have genuine truth value. They may therefore be
effectively used in communication, including philosophic communication.
But we have also likened the metaphor to literal ambiguity. While one
may make a point, there will be question as to which point he is making.
Parties may easily assent to the same metaphor while disagreeing over
the matter at hand. We advise then, that one is wise to confine his
effort to literal speech when his point can be made with literal speech.
But since we reject the position that all metaphors have literal equiva—
lents, we must admit that there may be occasions where only the metaphor
will express the point. In such cases the metaphor is entirely appro-

priate.

A

12.

13.

14.

15.
16.
17.
18.
19.

oxooowoscn

236

NOTES TO CHAPTER VII

. Myers.

. These problems were discussed in our evaluation of Goodman's position

in the previous chapter.

. One may determine that a line must be metaphor due to the type of

material he is reading or perhaps because some authority de-
clared it to be metaphor.

. Example of Hyperbole: 'Waves mountain high broke over the reef.‘

Example of Irony: 'This is nice weather' uttered on a stormy day.
Henle, 182.
Black, 276.

. Ibid., 287.

Henle, 186.

. Black, 275.

. Owen Thomas, Metaphor and Related Subjects (New York: Random House,

1969), p. 22.

We are not claiming that this is true of Martha Rae, but she, too,
is noted for having a big mouth.

Abraham Kaplan, "Referential Meaning in the Arts," Journal of
Aesthetics and Art Criticism, XII (l954), pp. 457-74.

Ibid., 469, Kaplan uses the terms 'vehicle' and 'tenor' in much the

same manner as does Richards.
Ibig.
Stewart, “Metaphor and Paraphrase", p. 113.
Henle, 180.
Wheelwright, l09.
Ibig,

SELECTED BIBLIOGRAPHY

Books

Beardsley, Monroe C. Aesthetics. New York: Harcourt, Brace, and World
Inc., 1958.

 

Eastman, Max. The Literarnyind. New York: Charles Scribner's Sons,
1931.

Empson, William. Seven Types of Ambiguity. Meridan, 1955.

. The Structure of Complex Words. Ann Arbor: The University
of Michigan Press, 1967.

Foss, Martin. Symbol and Metaphor in Human Experience. Lincoln:
The University of Nebraska Press, 1949.

Goodman, Nelson. Languages of Art. London: Hutchinson University
Library, 1949.

. The Structure of Appearance. 2nd ed. Indianapolis:
BoBbs-Merrill Company, Inc., 1966.

Henle, Paul. Lagguage, Thought and Culture. Ann Arbor: The University
of Michigan Press, 1958.

Ogden, C. K. et al. The Foundations of Aesthetics. London: George
Allen and Unwin Limited, 1922.

 

Pierce, Charles Sanders. Collected Papers. Edited by Charles Hartshorne
and Paul Weiss. Cambridge:

Quine, Willard Van Orman. Elementary Logic. New York: Harper and Row
Publishers, Inc., 1965.

. From a Logical Point of View. New York: Harper and Row
Publishers, Inc., 1963.

Methods of Logig, .New York: Holt, Rinehart, Inc., 1959.

Richards, I. A. How To Read A Page. New York: W. W. Norton and Company,
Inc., 1942.

. The Philosophy of Rhetoric. New York: Oxford University
Press, 1965.

237

238

Richards, 1. A. Principles of Literary Criticism. New York: Inter-
national Library of Psychology, 1925.

Practical Criticism: A Study of Literary_Judgment. New York:
Harcourt, Brace and World Company, 1929.

Ryle, Gilbert. The Concept of Mind. London: Hutchinson University
Library, 1949.

Santayanna, George. The Sense of Beauty. New York: Danes Publications
Inc., 1955.

 

Shibles, Warren. Essays on Metaphor. Whitewater: The Language Press,
1972.

 

. Metaphor: An Annotated Bibliography and History. Whitewater:
The Language Press, 1971.

Thomas, Owen. Metaphor and Related Subjects. New York: Random House,
1969.

Turbayne, Colin Murray. The Myth of Metaphor. Columbia: University of
South Carolina Press, 1971.

Urban, Wilbur Marshall. Language and Reality. London: George Allen
and Unwin Limited, 1939.

 

Wheelwright, Phillip. The Burning_Fountain. Bloomington: Indiana
University Press, 1968.

 

Metaphor and Reality. Bloomington: Indiana University Press,

 

 

1969.

Articles

AIdr1Ch. Virgil. "Form in the Visual Arts," British Journal of Aesthet-
iEé; II (1961), 215-226.

Beardsley, Monroe C. "The Metaphorical Twist," Philosophy and Phenomen-
ological Research, XXII (1962), 293-307.

 

Black, Max. "Metaphor," Proceedings of the Aristotelian Society, LV
(1954-1955), 273-294. Reprinted in Models and Metaphor. Ithaca,
New York: Cornell University Press, 1962.

. "Some Questions About Emotive Meaning," The Philosophical
Review, LVII (1948), 111-126.

239

Blocker, Gene. "The Meaning of A Poem," British Journal of Aesthetics,
X (1969), 337-343.

 

Elliott, R. K. “Poetry and Truth," Analysis, XXXV (1967), 77-85.

Isenberg, Arnold. "The Problem of Belief," The Journal of Aesthetics
and Art Criticism, XIII (1955), 395-407.

 

. "On Defining Metaphor," The Journal of Philosophy, XXI
(1963), 609-622.

Kaplan, Abraham. "Referential Meaning in the Arts," Journal of Aesthet-
ics and Art Criticism, XII (1954), 457-474.

 

Khatchadourian, Haig. "Metaphor," The British Journal of Aesthetics,
VIII (1968), 227-243.

 

. "Symbols and Metaphors," Southern Journal of Philosophy,
VIII (1968), 181-190.

 

Loewenberg, Ina. "Truth and Consequences of Metaphors," Philosophy and
Rhetoric, VI (1973), 30-46.

Mészaros, Istvan. "Metaphor and Simile," Proceedings of the Aristotelian
Society, LXVII (1967), 127-144.

Mew, Peter. "Metaphor and Truth," British Journal of Aesthetics, XI
(1971), 189-195.

Miller,(Frederick Max. "Metaphor," Lectures on the Science of Language,
I 1873 , 3.

Myers, C. Mason. "Metaphors and Mediately Informative Expressions,"
Southern Journal of Philosophy, VIII (1968), 169-176.

Olscamp, Paul J. "How Some Metaphors May be True or False," Journal
of Aesthetics.and Art Criticism, XXIX (1970).

Quine, Willard Van Orman. “On What There Is," Review of Metaphysics,
XIV (1948), 21-38.

Reiser, Max. “Analysis of Poetic Simile," Journal of Philosophy, XXXVII
(1940). 209-217.

Steward, Donald. "Metaphor and Paraphrase," Philosophy and Rhetoric,
IV (1971), 111-123.

Wheelwright, Phillip. "Symbol, Metaphor, and Myth," The Sewanee Review,
LVIII (1950), 678-698.

"A Symposium on Emotive Meaning," The Philosophical Review, LVII (1948),
111-157.