This article was downloaded by: [University of California Santa Cruz]

On: 10 November 2014, At: 08:22

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Metaphor and Symbolic Activity

Publication details, including instructions for authors and subscription
information:
http://www.tandfonline.com/loi/hmet19

Pictorial Metaphor in Advertisements

Charles Forceville
Published online: 17 Nov 2009.

To cite this article: Charles Forceville (1994) Pictorial Metaphor in Advertisements, Metaphor and
Symbolic Activity, 9:1, 1-29, DOI: 10.1207/s15327868ms0901_1

To link to this article: http://dx.doi.org/10.1207/s15327868ms0901_1

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis, our
agents, and our licensors make no representations or warranties whatsoever as to the
accuracy, completeness, or suitability for any purpose of the Content. Any opinions and
views expressed in this publication are the opinions and views of the authors, and are not
the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be
relied upon and should be independently verified with primary sources of information. Taylor
and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs,
expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or
indirectly in connection with, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial
or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply,
or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access
and use can be found at http://www.tandfonline.com/page/terms-and-conditions
 METAPHOR AND SYMBOLIC ACTIVITY Y ( I ) . 1-79
Copyright O 1994. Lawrence Erlbaum Associates. Inc.

Pictorial Metaphor In Advertisements

Charles Forceville
Vrije Universiteit Amsterdam
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

Metaphor has been first and foremost studied in its verbal variants. As Lakoff
and Johnson (1980) claimed, however, "metaphor is primarily a matter of
thought and only derivatively a matter of language" (p. 153). Presuming this
idea to be correct. this article makes an exploratory contribution to the study
of a different type of metaphor, namely, pictorial metaphor. With reference to
Black's (1962, 1979) interaction theory, several pictorial metaphors in adver-
tisements are considered with the following questions in mind: What are the
two terms of the metaphor and how d o we know? Which of the two terms is
the ("literal") A-term and which is the ("figurative") B-term? And what can be
said about the transfer of properties from B to A? Contextual factors of various
kinds help to answer these questions. A tentative subdivision is made into
metaphors with one pictorially present term and metaphors with two pictorially
present terms.

Because the study of metaphor is rooted in literary criticism and was subse-
quently taken up by philosophers and students of language, it is hardly
surprising that the overwhelming majority of articles and books on the
subject have concentrated on verbal metaphor. As early as 1936, however,
Richards (193611965) noted that "fundamentally [metaphor] is a borrowing
between and intercourse of thoughts, a transaction between contexts.
Thought is metaphoric, and proceeds by comparison, and the metaphors of
language derive therefrom" (p. 94). More recently, this insight was reiterated
by Lakoff and Johnson (1980), who stated that "metaphor is primarily a
matter of thought and only derivatively a matter of language" (p. 153). If
these authors are right, as I think they are, this implies that it is justified and
imperative also to research metaphor in its nonverbal and partly verbal
manifestations.
In this article, my concern is with pictorial metaphor. Interesting work in

Requests for reprints should be sent to Charles Forceville. Vrije Universiteit Amsterdam,
Faculty of Arts. P.O. Box 7161, 1007 MC Amsterdam, Netherlands.
 2 FORCEVILLE

this area has been done by the psychologist John Kennedy. Kennedy (1982)
argued in favor of adopting the concept of metaphor in pictures by present-
ing visual counterparts to a number of figures of speech known from rhetoric,
as presented under the heading "technical terms" in Fowler (1926, p. 597).
Sixteen devices, including allegory, anti-climax, catachresis, clicht, and meto-
nymy, are listed by Kennedy and interpreted in terms of Richards's well-
known tenor-vehicle distinction (Richards, 193611965, p. 96). This
distinction, however, is applied in a loose and unconvincing way. There is a
plausible reason for this loose application, because Kennedy's use of the term
metaphor is quite different from that of Richards. Whereas Kennedy takes
metaphor to be an all-inclusive term, somewhat synonymous withJigure of
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

speech or trope, Richards uses the term in the far more restricted sense in
which it is commonly used today. Thus, by metaphor, Kennedy either means
something different from Richards or he judges, mistakenly, that the tenor-
vehicle distinction, which Richards himself only applies to metapha in its
restricted sense, can without adaptation be transferred to other tropes. As a
result, Richards's vocabulary is too generally and, hence, imprecisely em-
ployed by Kennedy.
To avoid confusion, let me stress that I consider Kennedy's examples on
the whole to be convincing; it is only his terminology that I find unsatisfac-
tory. Therefore, although I concur with Kennedy (1982) that figures of
representation can "include both language and depiction" (p. 593), and
particularly sympathize with his emphasis on the importance of context in
pictorial metaphors, I think the study of this branch of metaphors would
benefit from a more systematic approach and a stricter application of termi-
nology than the one he provides.
Hence, my intention in this article is to make an exploratory contribu-
tion to a theory providing the mechanisms underlying pictorial metaphor
and to contribute to the development of a vocabulary enabling us to dis-
cuss this type of metaphor. Like Kennedy, I borrow my concepts and ter-
minology from studies in the domain in which metaphor has been most
extensively investigated: language. I do not, however, cast my net as wide
as he does and discuss metaphor in only its narrow sense of that pictorial
element in a picture that is represented in such a way that a viewer of the
picture is forced to understand or experience that element in terms of an-
other element (whether or not visibly present in the picture itself), without
there being a pre-existent or conventional connection between these two
elements. More specifically, I try to formulate answers to the following
questions: What are the two terms of the metaphor and how do we know?
Which of the two terms is the ("literal") A-term and which is the ("figura-
tive") B-term? And what can be said about the transfer of properties from
B to A?
 PICTORIAL METAPHOR IN ADVERTISEMENTS 3

ASSUMPTIONS

Discussing pictorial metaphors in terms of the vocabulary developed in the

area of verbal metaphors is fine, but, as theorizing about the latter is still in
full swing, it is impossible to apply a generally accepted, neatly organized set
of concepts to a corpus of pictorial metaphors and then record the results. In
the absence of an authoritative, all-inclusive theory, I briefly outline the
assumptions that underlie my analyses.

1. Throughout this article, the word metaphor must be taken in the spirit
of Lakoff and Johnson's (1980) description: "The essence of metaphor is
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

understanding and experiencing one kind of thing in terms of another" (p.

5)-even though the use of the word thing is a bit unfortunate (see Assump-
tion 2).
2. As a general background, I use Black's modified creativity theory, as
presented in Ortony (1979, pp. 27-29). A paraphrase of the main point is: A
metaphorical statement has two distinct subjects, to be identified as the
"primary" subject and the "secondary" one. These subjects are to be re-
garded as systems rather than as individual things. The metaphorical utter-
ance works by "projecting on" the primary subject a set of "associated
implications," comprised in the implicative complex, that is predicable of the
secondary subject. The maker of a metaphorical statement selects, empha-
sizes, suppresses, and organizes features of the primary subject by applying
to it statements isomorphic with the members of the secondary subject's
implicative complex. In the context of a particular metaphorical statement,
the two subjects "interact" in the following ways: (a) The presence of the
primary subject incites the hearer to select some of the secondary subject's
properties; (b) invites him to construct a parallel implication complex that
can fit the primary subject; and (c) reciprocally induces parallel changes in the
secondary subject.
Moreover, Black's (1979) central claim that metaphors often do not reveal
a pre-existent similarity but rather create that similarity (p. 37) is understood
to be crucially correct. (See also Kennedy's, 1990, account of metaphor as
special-purpose classification.)
3. In metaphor, "similarity is a central concept" (Ortony, 1979, p. 189).
Miller stated: "Whether or not similarity is a defining property of metaphor,
no one could disagree with the claim that many metaphors are apperceived
in terms of resemblances" (Miller, 1979, p. 214). Also, elements such as
physical resemblance, spatial resemblance, resemblance pertaining to size,
and other properties that can be visualized, play a more important role in
pictorial metaphors than in verbal ones.
4. A notion of particular relevance for my present enterprise is the notion
 4 FORCEVILLE

of directionality. As Miller puts it: "As soon as we consider how authors use
analogies . . . we find that the context imposes a direction, that it is no longer
possible to rearrange freely the order of the terms" (Miller, 1979, p. 215). In
other words, the metaphorical transfer is primarily unidirectional.
5. Not the word but the statement is "the sole contextual milieu within
which the transposition of meaning takes place" (Ricoeur, 197511978, p.
65).

All these assumptions originate in studies on verbal metaphor as noted. To

make these assumptions work in pictures, as I propose to do, they have to be
applied with some flexibility and imagination. For example, the word state-
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

ment must be replaced by another term, but it is difficult to come up with a

satisfactory pictorial equivalent. Picture, for example, would not do. Just as
a text can be said to contain a verbal metaphor, so a picture can be said to
contain a pictorial metaphor; it is not, however, necessarily a metaphor itself.
All this may sound rather abstract, but I trust it will become clear once we
come to the actual pictorial metaphors.

MATERIALS

The pictorial material used for the analyses consists of a number of French,
Dutch, and British advertisements1(reproduced here in black and white). I
opted for advertisements because they usually have a clear-cut purpose, and
this has proved essential to a first exploration of the workings of pictorial
metaphor. My earlier research (Forceville, 1988) revealed that the analysis of
pictorial metaphors in a number of surrealist works of art encounters great
problems in the phase of their interpretation. Unequivocal verbalizations of
the metaphors contained in these artistic representations proved difficult to
realize. This is partly due to the subversive character of surrealist art and is
probably also a consequence of the more general fact that art has no unam-
biguously identifiable purpose. This matter of purpose poses no problem in
the study of advertisements. As Rossiter and Percy (1987, p. 3) stated, "Ad-
vertising and promotion are increasingly being viewed by marketing manag-
ers as highly interrelated yet distinctly specialized means of informing
customers about products and services and persuading them to buy" [italics
added]. This knowledge considerably facilitates the identification and inter-
pretation of signs (and, hence, of metaphors) in advertisements. I am encour-
aged by Barthes, who gives the same reason for focusing on advertisements
in developing his theories:

'See Appendix for the sources of the advertisements presented in this article.
 PICTORIAL METAPHOR IN ADVERTISEMENTS 5

We shall start by making matters considerably easier for ourselves: we shall

study only advertising images. Why? Because, in advertising, the image's sig-
nification is assuredly intentional: it is certain attributes of the product which
a priori form the signifieds of the advertising message, and these signifieds must
be transmitted as clearly as possible; if the image contains signs, we can be sure
that in advertising these signs are replete, formed with a view to the best
possible reading: the advertising image is frank, or at least emphatic. (Barthes,
196411986, p. 22)

DISCUSSION OF PICTORIAL METAPHORS

Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

In Figure 1, the foregrounded object is a shoe. We immediately see that there

is something odd about this shoe: It is located where we would ordinarily
have expected something else, namely, a tie. The viewer is invited to perceive
the phenomenon shoe not in its usual, literal sense but in terms of the very
different phenomenon tie. The metaphor can be verbalized as "SHOE IS
TIE. "

Three questions immediately arise. First, how do we know that the second
term of the metaphor is tie and not something else? After all, the tie is not
even visible. This becomes clear when we consider the next, reconstructed
picture (Figure 2). Even with the help of the text ("Regardez mes chaus-
sures!"-"Look at my shoes!"), we would not be able to identify the second
term of the metaphor. Indeed, in this picture we cannot even say that what
we have is a metaphor, because there is no (suggestion of a) second term.

FIGURE 1 "SHOE IS TIE" (Advertising agency: Publi-EstIGerstenhaber + Cie).

 6 FORCEVILLE
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

FIGURE 2 "SHOE IS TIE" (doctored version of Figure 1).

Thus, it is thanks to the pictorial context that we can establish which is the
(absent) secondary subject (SS) of the metaphor. Black's creativity theory is
entirely applicable here: The similarity between the two terms is not pre-
existent but-with the aid of this specific context-is created.
A second question is, How do we know the metaphor is "SHOE IS TIE"
rather than "TIE IS SHOE"? This question pertains to the order of the
metaphor's two terms: What is the metaphor's literal A-term, the primary
subject (PS), and what is its figurative B-term, the SS? At first sight, this
problem may seem characteristic of pictorial metaphors, because the ele-
ments in pictures are not linearly perceived. It is this linearity that is such a
great help in the distribution of PS and SS in the case of verbal metaphors.
In the type of verbal metaphors that theorists usually cite, the examples are
often one-line "A IS B" metaphors, in which the first term is invariably the
PS and the second the SS. This might lead us to think that all verbal meta-
phors feature this neat, linear order, with its simple distribution of PS and SS.
This, however, is not always the case. Brooke-Rose (1958, pp. 117- 118), after
remarking that "most equations with the copula put the proper term before
the metaphor" (i.e., metaphoric B-term), gave the following counterexample
from Spenser's Amoretti: "That ship, that tree, and that same beast am I"
(Brooke-Rose, 1958, p. 118). Of course the form of the verb to be employed
here is an additional clue-absent in pictorial metaphors-as to the order of
the terms.
Returning to our pictorial metaphor, we realize that, in the absence of the
information that usually suffices to assess the order of the terms in verbal
metaphor (linearity and grammatical clues), the key device in determining
this order in its pictorial counterpart has to be a different one, namely,
context. Here, context has to be taken in a wider sense than previously
 PICTORIAL METAPHOR IN ADVERTISEMENTS 7

mentioned, when we invoked the pictorial context to determine the second,

pictorially absent, term. Invoking the pictorial context helps little to deter-
mine the order of the terms. It is the awareness that the advertisement is one
for shoes rather than for ties that guides the viewer's distribution of PS and
SS as "SHOE IS TIE" instead of "TIE IS SHOE. " Hence, it is the combina-
tion of the viewers' classification of the picture as an advertisement and their
understanding of the verbal context that is responsible for the distribution of
PS and SS. Later, I discuss the role of context in advertisements.
A third point in reference to this metaphor is the matter of property
transfer. What property or properties are, in Black's terminology, projected
from the implicative complex of the SS (tie) upon the PS (shoe)? Of the three
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

issues raised, this one is potentially the most controversial. Our example,
however, seems rather clear-cut. Understanding the advertisement to hint
that Clerget shoes are so beautiful that they should be worn as an ornamental
rather than as a functional piece of clothing, I propose the property trans-
ferred can be formulated as "nonfunctional beauty."
Similar reasoning can be applied to Figure 3. As in Figure 1, there is
something odd in this advertisement. Normally, one would expect corn to be
found in a can or a jar, whereas the usual (literal) container for wine is a
wineglass. The sweet corn in the wineglass, then, is a deviation from what is
expected (i.e., fromconvention, from literal use). In the resultingmetaphor, the
second term of the metaphor is wine, which we know because of the pictorial
context (the glass itself and the typical position of the hand holding the

FIGURE 3 "SWEET CORN IS WINE" (Advertising agency: Beverly-Markagri, Groupe

"Publicis").
 8 FORCEVILLE

wineglass), although wine is also verbally reinforced ("L'Appellation trks

contr6li.e"). The order of the terms can be inferred from the fact that the
advertisementis for sweet corn rather than for wine; that is, the viewer is invited
to perceive sweet corn in terms of wine rather than wine in terms of sweet corn,
yielding the metaphor "SWEET CORN IS WINE. "The properties projected
from wine upon sweet corn can be phrased as "refinement," "having a delicate
taste," "preciousness," and "deserving respectful treatment."
In both examples discussed, the PS of the metaphor coincided with the
product advertised. This is hardly surprising, because one would expect
advertisers to make some sort of claim about their product, and the PS is the
term of the metaphor about which something is predicated. These do not
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

appear to be isolated examples. Figures 4, 5, 6, 7, 8, and 9 would have to be

verbalized as "MACHINES ARE CHESS PIECES"; "TUBE IS CUSH-
ION''; "GAS NOZZLE IS GUN''; "CAR IS WITCH"; "TICKET IS DECK
CHAIR"; and "TICKET IS MONOSKI, respectively. In these cases, it is
"

the product-or something that refers metonymically to the product-that

is the pictorially present PS of the metaphor. The viewer is invited to see this
product in terms of something else, the SS, and this SS is pictorially absent.

ROLAND BARTHES'S THREE KINDS OF MESSAGES

At this stage, it seems useful to introduce some labels that bear on the verbal
component in the advertisements under consideration, and that will facilitate
our discussions of them. Barthes provided some very important and relevant

FIGURE 4 "MACHINES ARE CHESS PIECES" (Advertising agency: ACTE I).

 PICTORIAL METAPHOR IN ADVERTISEMENTS 9
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

FIGURE 5 "TUBE IS CUSH-

ION" (Advertising agency:
BCRC).

concepts. In his analyses of advertisements, he distinguished between three

messages: the linguistic message, the denoted message, and the connoted
message. The linguistic message comprises everything expressed in language,
whether it appears as text, caption, or headline in the image or outside the
image. Barthes subdivided two functions that the linguistic message may
have in relation to the pictorial message: relaying and anchoring. In its
relaying function, mainly manifested in cartoons and comic strips, "language
. . . and image are in a complementary relation" (Barthes, 196411986, p. 30).
The anchoring function of language is, according to Barthes, more com-
mon. In this function, the linguistic message guides the identification and the
interpretation of the pictorial components of the image.

The text directs the reader among various signifieds of the image, causes him
to avoid some and to accept others; through an often subtle dispatching, it
teleguides him toward a meaning selected in advance. In all these cases of
anchoring, language obviously has a function of elucidation, but such elucida-
tion is selective; it is a matter of a metalanguage applied not to the whole of the
iconic message but only to certain of its signs. (Barthes, 196411986, p. 29)

As far as the pictorial side of the advertisement is concerned, Barthes distin-

guishes between two messages. First, there is the sum of often discontinuous
signs that inhere in the image as a whole and in its parts, that is, all the
associations evoked by it. Barthes called this the "connoted" or "symbolic"
image. Underlying this image is the second, "denoted" or "literal" image; it
 10 FORCEVILLE
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

FIGURE 6 "GAS NOZZLE IS

GUN" (Advertising agency: DDB,
New York).

is "what remains in the image when we (mentally) erase the signs of connota-
tion" (p. 3 1). This is the image in its innocent state, hypothetically seen before
it is charged with the meanings of connotations. Barthes realized that the
connoted and denoted aspects of the image cannot be separated: "we never-
at least never in advertising-encounter a literal image in the pure state" (p.
31). The distinction between the literal message and the symbolic message is
merely operational: It corresponds to the twofold division into signifier and
signified in the worldlverbal sign.
We return now to Figures 1 and 3 and consider how Barthes's concepts
shed more light on the matter of context. We saw earlier how the pictorial
context unambiguously suggested the absent term but does not give decisive
clues as to the order of the terms. At first, we might think it is merely because
they are the visible terms that we understand the sweet corn as wine and the
shoe as tie, instead of the other way around. However, this criterion of plus
or minus visibility is not sufficient for the distribution of PS and SS, as is
convincingly shown by Figure 10. In this advertisement, we see the unex-
pected life buoys instead of the expected tires. We know tires is the other term
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

FIGURE 7 "CAR IS WITCH"

(Advertising agency: Publicis).

FIGURE 8 "TZCiYET IS DECK CHAIR" (Advertising agency: H C M ) .

 12 FORCEVILLE
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

FIGURE 9 "TICKET IS MONOSKI" (Advertising agency: HCM)

FIGURE 10 "TIRES ARE LIFE BUOYS" (Advertising agency: Abbott VickerslSMS

Ltd.).

of the metaphor because of the relevant pictorial context of the car. The order
of the terms is inferred from the text, "Dunlop tires" (the advertisement is for
tires, not for life buoys), so that the verbalization of the metaphor is "TIRES
ARE LIFEBUOYS. "The property projected from SS upon PS is "providing
safety." Unlike the previous examples, however, it is here the B-term of the
metaphor instead of the A-term is pictorially rendered. This case proves that
visibility of a term does not automatically lead to its being accorded PS
status.
 PICTORIAL METAPHOR IN ADVERTISEMENTS 13

As the analysis of the metaphor in Figure 10 suggests, there is a more

decisive element that guides the viewer in the distribution of PS and SS in the
advertisement metaphors under discussion, and that is the linguistic message
that anchors the picture and thus directs the viewer's interpretation. In all
three advertisements (Figures 1, 3, and lo), we find words in the anchoring
text that refer to the PS of the metaphor used in the advertisement. Therefore,
if a consideration of the purely pictorial context should allow doubt as to the
PSISS distribution of a metaphor, the wider context of the anchoring text
dissolves this doubt. Notice that it is not merely the text that helps us out: The
realization that the picture-cum-text is something called "an advertisement"
plays an important role, too. This realization is not merely a result of com-
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

bining image and text; the place of this image-cum-text in a magazine or

newspaper as well as its layout certainly contribute to this. Thus, it is neces-
sary to distinguish another, still higher level of context: Beyond the pictorial
context and the anchoring textual context, there is the context of what
Barthes calls "world knowledge." This world knowledge pertains to an enor-
mous number of facts, beliefs, conventions, and experiences. One relevant
aspect of world knowledge is the awareness that the picture one is facing is
an advertisement. The contextual levels of text and world knowledge, then,
guide the viewer into attributing PS position to the term-if there is one-in
the pictorial metaphor that is or represents the product.
In returning to the metaphors previously discussed, we see that they fit this
hypothesis well. We have not only a clear-cut PSISS distribution but also
another important element from Black's (1979) theory of metaphor in the
form of a projection of properties from the implicative complex of the SS
upon the PS. See Scheme to follow.

SCHEME
Figure PS SS
- Projected Property(ies)
1 Shoe Tie Designed for aesthetic, not
functional, purposes
3 Sweet corn Wine Refinement," "preciousness,"
"having a delicate taste,"
"deserving respectful treat-
ment
4 Machines Chess pieces Internal cooperation ensuring
best results
5 Tube Cushion Providing rest (for the feet)
6 Gas nozzle Gun Being potentially lethal
7 Car Witch Possessing magic powerslpro-
viding rapid transport
8 Ticket Deck chair Holiday 1exoticness
 14 FORCEVILLE

9 Ticket Monoski Holiday /exoticness

10 Tire Life buoy Providing safety

Some general remarks must be made. First, although the description of the
properties may sound strained, this unnaturalness merely results from having
to "translate" something into words that emerges from looking at the pic-
tures. Second, it is clear that the description of the property or properties at
stake often is not the only possible one. Other formulations are conceivable,
and it is possible that there are more properties than mentioned here that can
be sensibly projected from B to A. In Barthes's terminology, I have been
looking for the connoted message(s) of the SS that are applicable to the PS.
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

As Barthes admitted, "the number of readings of the same lexia (of the same
image) varies according to individuals" (p. 35). What matters is that "varia-
tion in readings is not anarchic, it depends on the different kinds of knowl-
edge invested in the image (practical, national, cultural, aesthetic
knowledge), and these kinds of knowledge can be classified, can join a typol-
ogy" (Barthes, 196411986, p. 35). Although Barthes is talking about the
image as a whole, the same holds true for the one element of the picture I am
talking about-the SS. Because only one element is concerned, the potential
readings are probably even more restricted than in the case of an entire
image.
The metaphors previously considered give rise to fairly satisfactory read-
ings from a theoretical point of view. I have established that, in all but one,
the pictorially present term of the metaphor is the PS; that the PS is, or refers
to, the product advertised; that the absent term is unambiguously suggested
by the pictorial context; that if the pictorial context should leave us in doubt
as to which term is the PS and which is the SS, we need only one element from
the next layer of context (the anchoring text) to resolve this doubt, namely,
a realization of what the product of the ad is; and finally that there is a
projection of nameable properties from the implicative complex of the SS
upon the PS.

METAPHORS WITH TWO PICTORIALLY PRESENT TERMS

Hitherto I have examined advertisements containing metaphors in which

only one term of the metaphor was pictorially present. There are also adver-
tisements, however, that contain metaphors that feature both terms pictori-
ally. Figures 11 and 12 show fairly straightforward examples. In Figure 11,
we easily recognize the two terms of the metaphor as earth and candle, both
of which have been partially rendered. There is no pictorial context that may
help us to establish the order of the terms, so we have to consult the anchor-
ing verbal context to decide which of the terms is PS and which is SS. The
 PICTORIAL METAPHOR IN ADVERTISEMENTS 15
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

FIGURE 1 1 "EARTH IS CAN-

DLE" (Advertising agency:
McCann-Erickson).

heading ("We extract energy from the earth as if it were inexhaustible") and
the awareness that the authority which, in 1976, commissioned this warning
advertisement was the Dutch Ministry of Economic Affairs, suggest not only
that earth is the PS of the metaphor, but also supplies the feature transferred
from the SS candle to the PS earth: "providing an exhaustible amount of
energy."
Figure 12 provides another clear example. Both terms are depicted (the
bricks in their entirety, and the earphones only partially) in the electrical cord
and in the part of the earphones that goes over the man's head. The ear-
phones represent the advertised product and thus occupy the PS position,
leaving the SS position for the bricks. The metaphor, then, is "EARPHONES
ARE BRICKS. The headline, which reads, "Most earphones have one big
"

disadvantage," reinforces the feature transferred from SS to PS-heaviness.

Notice that the PS's referent is obviously earphones that are not promoted
by the manufacturer. This example proves that a PS can also refer antonymi-
cally to the advertised product.
The metaphor in Figure 13 belongs to the same series of advertisements
 16 FORCEVILLE
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

FIGURE 112 "EAR1 ARE

BRICKS " (Adveri ,ency:
KVH/G(3K).

as those in Figures 8 and 9. However, because part of the Indian headdress

is pictorially present, this metaphor, unlike the other ones, is a metaphor with
two pictorially present terms (MP2; a metaphor with only one pictorially
present term is labeled MP1). Note that the Indian headdress does not
connote holiday; it is the verbal context that activates, or rather, creates this
connotation. Furthermore, the SS in Figure 13 connotes holiday for a partic-
ular audience only (e.g., a European, adult audience living in the latter half
of the 20th century). That is, pictorial metaphors, like their verbal counter-
parts, are culturally determined.
To interpret the metaphor, we have to establish that ticket and deck chair,
monoski, and Indian headdress refer to the airline company issuing the ticket
and to holiday, respectively. This is exactly what happens in (part of) the
anchoring text: "Air France: Vacances." Figure 14 shows how the concept
Indian headdress can serve very different purposes. "Yellow Pages" are pre-
sented here in terms of an Indian headdress. As in the preceding advertise-
ment (Figure 13), the Indian headdress metonymically refers to Indian, but
in Figure 14, it is not the Indian's connotation of exoticness or far-away-
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

FIGURE 13 "TICKET IS IN-

DIAN HEADDRESS" (Advertis-
ing agency: HCM).

JRE 14 "YELLOW PAGES ARE INDIAN HEADDRESS" (Advertising agency:

nd Public).
 18 FORCEVILLE

holiday-destinations that matters, but rather the Indian being associated,

through strip cartoons and films in the Western genre, with the particular
pose of looking into the distance adopted by the man wearing the Indian
headdress. It is this cliche that is played on by the anchoring text ("With the
Yellow Pages I am on the right track"); without that, the metaphor would
not be clear. It is through a chain of metonyrns and cliches, then, that the
pictorial metaphor makes sense. The anchoring text and "world knowledge"
are indispensable for its interpretation.

MPls AND MP2s

Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

Previously, I made a distinction between MPls and MP2s, but to what extent
is this a clear-cut division? Let us look once more at Figures 1,3, and 10 and
test whether these MPls could be reverbalized as MP2s, that is, as metaphors
featuring (parts of) both terms. I argue that for "SHOE IS TIE" we have to
take recourse to a strained rendering like "MAN'S TORSO WITH SHOE IS
MAN'S TORSO WITH TIE"; that "SWEET CORN IS WINE" could be
rephrased as "JAR OF SWEET CORN IS GLASS OF WINE"; and that
"TIRES ARE LIFE BUOYS" could be transformed into "WHEELS ARE
LIFE BUOYS. " Because this would mean that now the "man's torso with
shoe," the "jar of sweet corn," and the "wheel" are (partially) present, we
would have to say that the metaphors belong in the MP2 category.
I previously argued that the verbalizations employed in this article are no
more than approximations to render into language what is presented pictori-
ally and that other formulations are therefore conceivable. Here, however,
we are talking about something else than minor variations in the wording
chosen to render the pictorial metaphors in language. The verbalizations
juxtaposed reflect different ways of experiencing the metaphors under consid-
eration. The question is, What belongs to the metaphor, and what belongs to
its context? The question whether one of the pairs previously quoted is a
more adequate verbalization than the other is therefore by no means a trivial
one. If the answer is negative, the entire distinction between MPls and MP2s
is irrelevant, but if we believe that one verbalization is a more appropriate
characterization of what we "see" than another, the distinction holds, and
this tells us something about the ways we channel our perceptions into
categories and concepts.
I cannot adduce decisive evidence, but I argue that in all three cases the
alternatives suggested seem to miss the point of the metaphor. What matters
in the advertisements is the shoe rather than something like "man's torso
with shoe" (it is also significant that there is no simple word or expression to
describe this phenomenon; apparently it does not exist in our culture as a
concept that requires a separate word); the (quality of the) sweet corn rather
 PICTORIAL METAPHOR IN ADVERTISEMENTS 19

than the way it is packaged; the tire rather than the wheel (the advertiser
advertises tires, not wheels). Therefore, it appears to make sense to stick to
the original verbalizations.
When re-examining the metaphors that were classified as MP2s, we con-
front the fact that these, too, are technically open to conversion, that is,
into MPls. Therefore, the MP2 "EARTH IS CANDLE" could also be
rendered as "UPPER PART OF EARTH IS UPPER PART OF CAN-
DLE," and "EARPHONES ARE BRICKS" as "EAR-COVERING
PARTS OF EARPHONES ARE BRICKS." In the case of the two Indian
headdress advertisements, we might opt for the rephrasings "TICKET IS
UPPER PART OF INDIAN HEADDRESS" and "YELLOW PAGES
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

ARE UPPER PART OF INDIAN HEADDRESS," respectively. Here

again, I suspect that the original verbalizations render the metaphors more
adequately, because the alternative verbalizations draw attention to an ir-
relevant level of the metaphor. I tentatively conclude that, although differ-
ent verbalizations are often possible to render the pictorial metaphors,
there is one privileged plane on which the two elements/concepts most
meaningfully interact, that is, correspond best to what the viewer experi-
ences as the concepts significantly conflated.
To further support the claim that the subdivision into MPls and MP2s is
valid, it should be added that the distinction seems to have a parallel in verbal
metaphors, namely, in the opposition between metaphor in praesentia and
metaphor in absentia. Ricoeur (19751 1978, p. 186) gave the examples "Jim
is an ass" and "What an ass!" In the first sentence, both PS and SS are given;
in the second sentence, the PS is not given and must be recovered from the
context. There is one crucial difference between the verbal and the pictorial
metaphors, however: Whereas in the verbal specimens it is the PS that must
be recovered from the context, in pictorial ones, this is almost always the SS.
(See Brooke-Rose, 1958), for an elaborate discussion of these types of verbal
metaphor.)
But does a comparison between Figures 8, 9, and 13 not make the entire
subdivision seem trivial? According to my classification, Figure 8 and Figure
9 contain MPls, whereas Figure 13 contains an MP2 just because a tiny part
of the SS Indian headdress happens to be pictorially present, but we intui-
tively feel that the three advertisements (and the metaphors they contain) are
based on the same idea. Granting all this, we are left with one important
technical difference between the MPls and the MP2s. Whereas in MP2s, no
contextual features (pictorial or verbal) are necessary to establish the two
terms of the metaphor, in MPls contextual features are absolutely indispens-
able to establish one of the terms, namely, the pictorially absent term of the
metaphor. Although it cannot be more than a hypothesis that will have to be
tested experimentally, I propose that, in their most plausible verbalizations,
some metaphors are MPls and some are MP2s; therefore, neither of the
 20 FORCEVILLE

categories can be incorporated into the other. The validity of this hypothesis
is taken for granted in the remainder of the argument.

TWO DIFFICULT CASES

After discussing a number of relatively straightforward examples of pictorial

metaphor, we now turn to two advertisements that at first sight seem to
contain metaphors but turn out to be problematic. First let us try to process
them as metaphors. In Figure 15, we see a lady's shoe with a wineglass where
we would have expected the shoe's heel. That the pictorially absent term of
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

the metaphor must be high heel is forcefully suggested by the only contextual
pictorial element in the advertisement: the rest of the shoe. Although the key
word Bordeaux in the anchoring text is sufficient to establish that wine is the
product advertised, and hence to assess what is the PS of the metaphor, the
resulting "GLASS OF WINE IS HIGH HEEL" does not make sense until we
have consulted the anchoring text in its entirety: "Marchez aux Bordeaux"
and "La couleur du bon gofit." The second slogan straightforwardly trans-
lates as "the color of good taste," and the first slogan is an exhortation that
could be rendered as "run on Bordeaux wines!", a jocular parallel to, say, a
car running on gas. The text corroborates that the wine refers to the subject
of the advertisement (Bordeaux) and explains the presence of the shoe (mar-
chez). Are there any features projected from SS upon PS here? We could say
that it is the connotation "eleganceu-suggested by the verbal "bon gofit"-
of the "high heel" that is to be projected upon the glass of wine, but if we can
talk about feature transfer in this case, this transfer is not of the straightfor-
ward kind we saw in most of the earlier examples. What seems to be occur-
ring here is the establishment of a rather loose associative chain in which
verbal and pictorial elements are cleverly combined. In any case, the anchor-
ing text is absolutely indispensable to make sense of the pictorial metaphor.
--- - - - - --

1 MAFCHEZ
AUX BORDEAUX.

FIGURE 15 "GLASS OF WI:VE

IS HIGH HEEL ." (Advert151ng
- LACOULEUR DU BON GOUS:
-- agency. Ted Bdtcs).
 PICTORIAL METAPHOR IN ADVERTISEMENTS 21

Consider Figure 16. A tennis player who is recognizably Bjorn Borg holds
an object in his hand that is simultaneously a tennis racket and a fishing net.
Both terms are identified not only by the formal similarities the hybrid object
shares with both tennis racket and fishing net-handle, ending in a more or
less round frame, provided with a mazed structure-but also by the presence
of contextual features belonging to the realms of tennis (Borg, tennis ball,
lines of court) and fishing (fish), respectively. As none of the terms is, or
(directlyrefers to, the product advertised ("Ahoy Hall"), the rule that product
equals PS does not apply. The pictorial context establishes the dominance of
the realm of tennis so that we would be inclined to opt for "TENNIS
RA CKET IS FISHING NET. "
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

But let us now take into account the next layer of context, the anchoring
text. The top caption can be translated as "That's only possible in Ahoy,"
Ahoy being a huge hall in Rotterdam, The Netherlands, where all kinds of
big events-exhibitions, concerts, shows, and sporting matches-take place.
'The translation of the bottom text is, "Ahoy . . . that's top-class sport, that's
fishing, that's all in one." The terms of the pictorial metaphor-if that is
what it is-do recur in the words top-class sport andjishing. It is interesting
that our assessment of the pictorial dominance of the tennis context, and

FIGURE 16 "TENNIS RACKET

AS' FZSlW N G NET" (Advertising
agency: Baas en van Haastrecht).
 22 FORCEVILLE

hence the labeling of tennis racket as the PS of the metaphor, is corroborated

by the text, as "top-class sport" is mentioned before "fishing."
But as in the preceding advertisement (Figure 15), we run into difficulties
when assessing what features are projected from fishing net upon tennis
racket. The best we could come up with is "capable of catching fish." This,
again, is highly unsatisfactory. This is not to say that the advertisement is
necessarily unsatisfactory. Let us interpret the advertisement to suggest that
Ahoy hall is so well equipped that it can host events as different as tennis
tournaments and, say, fishing exhibitions, with perhaps the added message
that these can take place simultaneously. The surprise caused by the confla-
tion of racket and fishing net is then quite functional. After all, "That's only
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

possible in Ahoy."
In neither of the two previously discussed cases can we-as in the other
metaphors-formulate a clear-cut property that is transferred from B to A.
In Black's scheme, the projection of elements from the implicative complex
of B upon A is an essential aspect of verbal metaphor. The same holds true
for pictorial metaphor, indeed it is a defining characteristic of metaphor in
general. The conclusion would then have to be that Figures 15 and 16 should
not be labeled metaphors. These advertisements therefore constitute a warn-
ing that a presentation of one object in terms of another should not automati-
cally be processed as a metaphor. Perhaps the description "(verbo-) visual
pun" is more appropriate. A clear-cut division, however, is still difficult to
make. More research will have to be done to distinguish pictorial metaphors
from (verbo-) visual puns and other pictorial tropes.

DISCUSSION

What other observations can be made when we survey this small corpus of
pictorial advertisement metaphors? First, a degree of physical similarity be-
tween PS and SS is often involved. It must be realized, however, that even this
similarity can only to a limited extent be said to be "preexistent." In several
instances, the PSs have been bent, folded, cut, or modified by other tech-
niques in order to create (physical) similarity with the SS. But physical
resemblance is by no means a necessary condition for pictorial metaphors.
For example, there seems to be very little intrinsic physical similarity between
machines and chess pieces (Figure 4) or between a car and a witch (Figure 7).
Therefore, when intrinsic physical similarity is largely or entirely absent,
identification of the terms, of their order, and of the projected feature(s)
depends completely on contextual features.
Second, the role the pictorial metaphors play in the advertisements in
which they appear varies widely. In some cases there is a fairly far-fetched
relation between the metaphor and the message conveyed by the advertise-
 PICTORIAL METAPHOR IN ADVERTISEMENTS 23

ment as a whole. An example is provided by Figure 6, in which the metaphor

"GAS NOZZLE IS GUN" needs the complement of the verbal message ("Or
else buy a Volkswagen!") before the advertisement as a whole can be under-
stood. By contrast, in Figures 1 and 3, the pictorial metaphor captures the
essence of the advertisement message as a whole ("Clerget shoes are so
beautiful you ought to wear them where you ordinarily wear your tie";
"Asgrow sweet corn is as delicate, well sampled, and precious as good wine").
Although it seems impossible to quantify the importance of the anchoring
role of the text in each case, it seems clear that these advertisements differ
considerably in the extent to which the text is necessary to clarify the purpose
of the pictorial metaphor in each advertisement. Those advertisements in
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

which the anchoring function of the text is not very strong, that is, in which
the textual message reinforces rather than directs or curbs the iconic message,
the pictorial metaphor comes close to being able to stand on its own. Al-
though this would have to be corroborated by more extensive research in-
volving experiments with unprejudiced viewers, I suspect that cases such as
Figures 1 and 3 constitute advertisements that could be understood even
without the anchoring text. Imagine Figures 1 and 3 without any text and
thus without the name of the product. There would appear to be two poten-
tial clues left to establish the order of the terms. In the first place, as we saw,
in many MPls the present term is, or refers to, the product advertised and
hence constitutes the PS. However, as clearly shown in Figure 10, where it is
the absent term that refers to the product advertised-and hence is the
PS-this rule does not work in all cases. But we have a second potential clue.
In most cases, only one of the two possible variants "A IS B" and "B IS A"
makes any sense. In Figures 1 and 3, reversing the terms to "WINE IS
SWEET CORN" and "TIE IS SHOE" would in the present (advertisement)
context make it difficult to find properties that could be transferred from SS
to PS.
Let it be emphasized that the ability to sort out the order of the terms in
Figures 1 and 3 presupposes the realization that the metaphors under consid-
exation are part of an advertisement. After all, a particular pictorial meta-
phor may, when displayed in a pop-art collage rather than in an
aldvertisement, have to be processed with reversed terms. Imagine that we
encounter our metaphor "SHOE IS TIE" as an artistic collage in a series in
which the tie is substituted by a different "alien" object in each collage (e.g.,
a1 broomstick, a candle, a tube). The metaphors would in that case probably
be processed as "TIE IS BROOMSTICK," "TIE IS CANDLE," and "TIE
IS TUBE, " that is, with tie in the PS position.
After surveying the whole corpus of the pictorial metaphors investigated,
we can state that it is immediately (i.e., before consulting the text) clear to the
viewer what the two terms of the metaphor are. In fact, it is only because the
viewer recognizes the terms (whether simultaneously present or not) that they
 24 FORCEVILLE

are aware of an anomaly. In this respect, there is no difference from verbal

metaphors. Mutatis mutandis, the following observation by Novitz (1985, p.
101) is true for pictorial metaphors as well:

It seems plain enough that we can only recognize that an utterance is a meta-
phor if we know that it should not be taken literally; and this, of course,
requires familiarity with the literal meanings of at least some of the words and
phrases deployed in the utterance.

Novitz (1985, p. 101) proceeds: "Equally obvious is the fact that we cannot
understand or be appropriately affected by a metaphor unless we are ac-
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

quainted with the literal meanings of the terms within it." For our purposes,
we would have to reformulate this last phrase as " . . . unless we are ac-
quainted with the way in which the terms are conventionally used." It must
be emphasized once more that, although pictures are certainly more "interna-
tional" than words, understanding the terms of a pictorial metaphor requires
knowledge of the cultural background in which it is embedded. In looking at
Figure 13, young children or people from a radically different culture might
be at a loss because they would be unfamiliar with the concept Indian or else
associate entirely inappropriate (i.e., inappropriate for understanding the
metaphor) connotations with it.
Finally, I briefly discuss two matters that, although important, have re-
ceived insufficient attention. I have talked about the language component in
the advertisements, primarily in its function of "anchoring" the image and
the pictorial metaphor contained in it. On close inspection, however, it
becomes clear that in a number of advertisementsthere is some sort of formal
resemblance between the pictorial metaphor contained in it and some part of
the anchoring text, usually the heading. We can witness this feature for
example in the Air France advertisements (Figures 8, 9, and 13). The recur-
ring slogan in these advertisements is "Air France Vacances, " with an obvi-
ous rhyming pattern. In my analyses, I have argued that in each of the three
advertisements it was the connotation "exoticness", 'choliday'7 from the
SS-whether deck chair, ski, or Indian headdress-that was projected upon
the PS ticket. I also claimed that the ticket metonymically referred to the
airline company issuing them: Air France. We can see now that the meta-
phorical relationship between ticket and its three SSs, in which the ticket is
physically represented in terms of something else, has a verbal parallel in the
slogan's phrasing. That is, the pictorial resemblance between ticket and deck
chair, ski and Indian headdress, respectively, is echoed in the verbal resem-
blance of "Air France Vacances." This phenomenon occurs often but unfor-
tunately not in our examples. One more example must suffice. In Figure 17,
which is an advertisement of a foundation promoting dental care, the word
zoetigheid ("sweetness") is placed above rottigheid ("rottenness," but also
 PICTORIAL METAPHOR IN ADVERTISEMENTS 25
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

FIGURE 17"TEETH ARE LZQ-

UORICE" (Advertising agency:
ACT).

"misery"), paralleling the liquorice above the rotten teeth. This observation
should serve as a reminder that the overall effect of an advertisement as a
whole can depend on considerably more than just the pictorial metaphor it
contains. On the basis of other advertisements (which are not discussed in
this article), the hypothesis might be hazarded that the more an advertise-
ment shows some sort of verbal parallelism to the pictorial metaphor it
contains, the weaker this pictorial metaphor tends to be. The idea is that the
makers of the advertisement are constrained by the formal requirements they
have to meet on the verbal level.
A second matter requiring thought is the distinction between pictorial
metaphor (the MP1 variant) and pictorial symbolism. Note that, here again,
the concept of context seems to be a key notion for distinguishing between
the two. It appears that a symbolic PS is far less context dependent than a
metaphorical PS when it comes to assessing the identity of the absent term.
 26 FORCEVILLE

As we saw, elimination of pictorial context in MPls (see Figure 2) leads to

nonmetaphor, in which there simply is no (hint of a) second term. On the
other hand, the symbol seems to have a far more independent existence: A
red rose, within certain cultural boundaries, always potentially suggests love.
Therefore, the MPI type of metaphor we examined depends on the context
to supply the second term, and this second term could never have been
identified without it, whereas a symbol evokes its second term with far less
or even no context, and this second term is a specific, fixed one. This idea,
however, requires more extensive research.
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

TENTATIVE CONCLUSIONS AND FURTHER RESEARCH

Metaphors not only occur in language but also in pictures. Advertising

provides a fruitful area for research pertaining to pictorial metaphor because,
unlike artistic representations, advertisements reveal more or less clear inten-
tions. A number of criteria delimiting the concept of metaphor can be derived
from the literature on verbal metaphor (notably Black, 1962, 1979) and
subsequently, with some flexibility, applied to pictorial metaphors.
The first questions to be answered are the following: What are the two
terms of the pictorial metaphor, and how are they to be verbalized? What is
the PS, and what is the SS in each metaphor, and how do we know? And what
properties, if any, are projected from the implicative complex of the SS upon
the PS? Because pictorial metaphors, unlike many verbal metaphors, have
nothing corresponding to either the latter's linearity or their (form of the)
verb to be, studying the context proves to be of crucial importance for
answering these questions. At least three context levels must be distinguished:
pictorial context; linguistic context ("anchoring text," as Barthes calls it);
and world knowledge, cultural knowledge, or encyclopedic knowledge. An
important aspect of the latter is the awareness of the kind of picture involved
(here, it is advertisements). Although a viewer would seem to make sense of
a pictorial metaphor by bringing to bear all levels more or less simultaneously
upon it, it appears probable that a hierarchical order in the contextual levels
can be distinguished. Roughly speaking, lower levels are modified by higher
levels.
Although the distinction is not unproblematic, it makes sense to subdivide
pictorial metaphors into MPls and MP2s. Theoretically, it turns out to be
possible to convert (i.e., reverbalize) MPls into MP2s and vice versa, but
these conversions seem to be counterintuitive. The two categories may corre-
spond to the verbal categories of metaphor in absentia and metaphor in
praesentia, respectively. The twofold division does not seem to entail differ-
ences for the manner of processing the metaphors. The difference pertains
primarily to a technical matter: In the MPls, we have no metaphor when the
 PICTORIAL METAPHOR IN ADVERTISEMENTS 27

context is excluded, as it is the context that supplies the second term, whereas
MP2s in this respect need no context, because the pictorial metaphor yields
both terms. The division into MPls and MP2s may well reflect our ways of
conceptualizing and "cutting up" the visual world around us in meaningful
units-which, in turn, are reflected in language-and in this sense the divi-
sion into MPls and MP2s may have significant implications.
Experiments with unprejudiced viewers are recommended to test the hy-
potheses in this article. What is particularly interesting and eminently testable
is the relative importance of each of the contextual levels for a viewer asked
to identify and distribute the terms and to assess the properties transferred
from SS to PS. To investigate this, the pictorial metaphors must be presented
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

to recipients in different forms; in each case, one or more of the contextual

levels are to be left out or substituted. In Figure 1, for example, one could
reproduce the picture in such a way that only a shoe is visible, leaving the text
untouched (see the reconstructed Figure 2), and tell the viewers that the
picture is an advertisement (i.e., supply a fact on the context level of world
k~nowledge).In this case, it is to be expected that viewers will not be able to
identify the second term (tie), but a reconstruction containing the whole
picture minus the anchoring text will still very probably elicit the metaphor's
SS and, possibly, the order of the terms. In the interpretation of this pictorial
metaphor, then, pictorial context is more important than anchoring (con)text
for the identification and ordering of the terms, and possibly even for the
determination of what properties are transferred. One could vary the pres-
encelabsence of the different context levels and find out which contextual
information is necessary or sufficient to identify and order the terms, as well
as to determine the property transfer.
Because many advertisements contain pictorial hybrids of some kind, it is
necessary to develop criteria for distinguishing pictorial metaphors from
nonmetaphors (be they instances of other pictorial tropes or not). Postulating
tlhe transfer of features as a defining characteristic of pictorial metaphor
should help distinguish metaphor from nonmetaphor.
Like verbal metaphors (cf. Lakoff & Johnson, 1980, chap. 5), pictorial
advertisement metaphors are culturally determined. The SS of a given picto-
rial advertisement metaphor must evoke specific connotations and not others
in order to be correctly understood by the viewers it wishes to persuade. Age
and cultural background (cf. the discussion of "TICKET IS INDIAN
HEADDRESS" in the section "Metaphors With Two Pictorially Present
Terms") are factors that are important to ensure this. In principle, it is
thinkable that a particular metaphor acquires a different meaning when
apperceived again 10 years later, even for the same recipient, simply because
the connotations of an SS have changed.
Finally, although these advertisements were primarily chosen because
their inherent intentionality makes them a suitable corpus for researching the
 28 FORCEVILLE

principles of pictorial metaphor, it is not merely the theory of metaphor that

may benefit from this kind of investigation. Like all metaphors, advertise-
ment metaphors present one phenomenon in terms of another phenomenon
and highlight some aspects of the PS (often the product) while hiding others.
An investigation of what PSs are used in connection with what SSs, and
subsequently which properties are projected from SS upon PS and which
properties are not, will no doubt shed more light on the persuasion tech-
niques employed in advertising.

ACKNOWLEDGMENTS
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

This article was written as part of Research Project No. 301-180-039, fi-
nanced by the Netherlands Organization for Scientific Research (NWO).
I thank Elrud Ibsch (Vrije Universiteit Amsterdam), Lachlan Mackenzie
(Vrije Universiteit Amsterdam), Peter de Voogd (Rijksuniversiteit Utrecht),
John Kennedy (University of Toronto), and an anonymous reviewer of
Metaphor and Symbolic Activity for their valuable comments on an earlier
version of this article. Furthermore, I am indebted to Marieke Nieuwland
(Nederlandse Vereniging van Erkende Reclame Adviesbureaus, Amsterdam)
and Marie-Lou Florisson (Art Directors Club Nederland, Amsterdam) for
helping me procure the illustrations used in this article.
This manuscript was accepted November 1990.

REFERENCES

Barthes, R. (1986). The rhetoric of the image. In R. Howard (Trans.), The responsibility of forms
(pp. 21-40). Oxford, England: Blackwell. (Original work published 1964)
Black, M. (1962). Models and metaphors. Ithaca, NY: Cornell University Press.
Black, M. (1979). More about metaphor. In A. Ortony (Ed.), Metaphor and thought (pp. 19-43).
Cambridge, England: Cambridge University Press.
Brooke-Rose, C. (1958). A grammar of metaphor. London: Secker & Warburg.
Forceville, C. (1988). The case for pictorial metaphor: Reni: Magritte and other surrealists. In
A. Erjavec (Ed.), Vestnik (Vol. 9, No. 1, pp. 150-160). Ljubljana, Yugoslavia: InStitut za
MarksistiEne Studije.
Fowler, H. W. (1926). A dictionary of modern English usage. Oxford, England: Oxford Univer-
sity Press.
Kennedy, J. M. (1982). Metaphor in pictures. Perception, 11, 589-605.
Kennedy, J. M. (1990). Metaphor-Its intellectual basis. Metaphor and Symbolic Activity, 5,
115-123.
Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chicago: University of Chicago Press.
Miller, G. A. (1979). Images and models, similes and metaphors. In A. Ortony (Ed.), Metaphor
and thought (pp. 202-250). Cambridge, England: Cambridge University Press.
Novitz, D. (1985). Metaphor, Derrida, and Davidson. The Journal of Aesthetics and Art Criti-
cism, 45(2), 101-1 14.
 PICTORIAL METAPHOR IN ADVERTISEMENTS 29

Ortony, A. (1979). The role of similarity in similes and metaphors. In A. Ortony (Ed.), Metaphor
and thought (pp. 186-201). Cambridge, England: Cambridge University Press.
Richards, I. A. (1965). Thephilosophy ofrhetoric. New York: Oxford University Press. (Original
work published 1936)
Ricoeur, P. (1978). The rule of metaphor (R. Czerny, Trans.). London: Routledge & Kegan Paul.
(Original work published 1975)
Rossiter, J. R., & Percy, L. (1987). Advertising and promotion management. New York:
McGraw-Hill.

APPENDIX
Downloaded by [University of California Santa Cruz] at 08:22 10 November 2014

Sources of Advertisement Illustrations

Reclame jaarboek 1976. Amsterdam: Art Directors Club Nederland
(Figure 11 [Advertising agency: McCann-Erickson]).
Reclame jaarboek 1980. Amsterdam: Art Directors Club Nederland
(Figure 12 [Advertising agency: KVHIGGK]).
10 ans de publicit6 (Stratbgies No. 294, October 30, 1981). Paris:
Strattgies (Figure 6 [Advertising agency: DDB New York] and Figure
7 [Advertising agency: Publicis]).
Vademecum voor adverteerders 1981. Amsterdam: VEA (Figure 17
[Advertising agency: ACT]).
Vademecum voor adverteerders 1983. Amsterdam: VEA (Figure 16
[Advertising agency: Baas en van Haastrecht]).
Dossier agences 1985. Paris: Stratkgies (Figures 1 and 2 [Advertising
agency: Publi-EstIGerstenhaber + Cie], Figure 3 [Advertising
agency: Beverly-Markagri (Groupe "Publicis")], Figure 4 [Advertis-
ing agency: ACTE I], and Figure 5 [Advertising agency: BCRC]).
Dossier campagnes 1985. Paris: Strattgies (Figure 15 [Advertising
agency: Ted Bates]).
Dossier campagnes 1986. Paris: Publications Professionelles Fran-
~aises(Figures 8,9, and 13 [Advertising agency: HCM] and Figure 14
[Advertising agency: Grand Public]).
British design and art direction. London: Polygon Editions SARL
(Figure 10 [Advertising agency: Abbott VickersISMS Ltd.]).