How Does an Advertisement Mean -- Cue, Claim, Metaphor, Resonance? Discussant's Comments


Edward F. McQuarrie (1990) ,"How Does an Advertisement Mean -- Cue, Claim, Metaphor, Resonance? Discussant's Comments", in NA - Advances in Consumer Research Volume 17, eds. Marvin E. Goldberg, Gerald Gorn, and Richard W. Pollay, Provo, UT : Association for Consumer Research, Pages: 658-661.

Advances in Consumer Research Volume 17, 1990      Pages 658-661


Edward F. McQuarrie, Santa Clara University

Each of the papers presented today looks at a different aspect of how meaning can be conveyed through advertising. The Balasubramanian paper uses the concept of a cue and examines affective meaning in an experimental context. The Berger and Gilmore paper draws on the discipline of semantics to describe several types of claim that can be made in an ad. The Ward and Gaidis paper reviews a number of literatures in order to profile various theories about metaphor The concepts of cue, claim and metaphor encompass a great deal of contemporary discussion of meaning in advertising. My discussion begins with an examination of what these concepts do not cover; i.e., aspects of meaning in advertising which require other concepts for their explanation. I then offer some reflections about the kind of theory which consumer researchers need in order to take a comprehensive approach to meaning in ads. The discussion concludes with a review of the strengths and weaknesses of the individual papers.


"Cue" is an extremely flexible concept for describing aspects of a stimulus which may prove meaningful to a respondent. Any aspect of a stimulus which can be discriminated or recognized by a viewer can serve as a cue. This flexibility makes cue a useful notion in dealing with a variety of non-verbal aspects of advertising. Cue is also a useful notion in an experimental context: it suggests elemental or basic features of a stimulus which an investigator can add to or subtract from the stimulus in the course of devising manipulations. There are, however, two difficulties with the cue concept when the focus is on meaning. First, the very idea of a cue grows out of a tradition in psychology -- reinforcement and drive theory -which has often been hostile to the notion of mental process. While the use of cue in a conceptualization does not presuppose or inevitably entail a behaviorist position, cue brings with it the idea of an involuntary or mindless response. The very flexibility of the cue concept can also be a drawback. Since virtually any and every aspect of an ad could potentially serve as a cue, we learn very little about the internal structure of an ad or about how different cues may have different effects. In this sense, to describe an element of an ad as a "cue' provides very little information about what the role of that element may be in the creation of meaning.

A claim is an assertion about truth value. As Berger and Gilmore show, several different types of claims can be distinguished. The idea of a claim is clearly relevant in two contexts: 1) issues of truth in advertising; and 2) issues of the persuasiveness of particular messages. Thus, a study of claims may tell us what meaning is conveyed, and whether a meaning will succeed in being conveyed at all. "Claim," however, may be of more limited applicability than "cue" in the study of contemporary advertising. It seems to me that a study of claims limits one to a consideration of linguistic statements. The idea of claim seems to break down if we consider a television commercial: Where is the claim? Is it only present in words spoken or written on the screen? Or does the commercial as a whole constitute a kind of claim? The second problem with claim would appear to be its restriction to the explicit or the denotative. The idea of a tacit claim seems full of problems. If the claim has to be constructed by the recipient, is it fair to describe it as a claim made in the ad? If it is true that reports, inferences and judgments, as claims, are concepts that can be applied only with difficulty to non-verbal elements of ads, and to connotations of ads, then their utility may be quite limited in the assessment of meaning.

The idea of a metaphor is a rich one. It is clearly able to encompass both connotation and non-verbal elements in ads. Unlike cue, it also carries with it notions about internal structure. My problem with the discussion of metaphor in ads is simple: what about metonym? Metaphor is but one figure of speech; what about the others? I have to be suspicious of theories of metaphor that are not included within more general theories about figurative speech. In the absence of a more general framework, one fears that the specific phenomenon of metaphor will be so entangled with the general phenomenon -- figurative speech -- that one will be unable to tell whether the theory claims to explain the one, the other, or both.


At this point I would like to describe a device for conveying or constructing meaning in advertisements that does not seem to be either a cue, a claim, or a metaphor. Consider the following ads:

1. A man's biceps are shown flexed as he works at an exercise machine. The headline reads, "Arm yourself."

2. A shapely woman is shown from the rear as she works out on a stair machine such as those found in health clubs. The headline reads: "People are starting to stair."

3. An ad to Trident gum has the headline, "Fight cavities with a stick."

Magazine ads of this type are not uncommon (McQuarrie 1989, 1990). I have coined the term resonance to describe the aspect that these ads share. Resonance may be defined as an echoing or a doubleness within the meaning structure of an ad, such that one element takes on multiple meanings, or multiple elements are given a single meaning. A more common term, which describes some but not all instances of resonance, would be wordplay (Grinnell 1987). Two points can be made about the three ads described: 1) the picture and the headline work together to create meaning; 2) resonance requires both a particular structure in the stimulus, and constructive activity on the part of the reader, in order to occur. Thus, the Trident ad is not resonant until the reader supplies the phrase "a stick of gum.

Resonance could be described as an executional cue, but I do not see what is gained thereby. The stair ad contains any number of cues, and this categorization does not illuminate the action of resonance. There are probably claims made or implied by means of the resonant structure, but it is not clear that resonance itself is a type of claim. Similarly, resonance can be described as figurative speech, but it is not the same kind of figure as metaphor. However, this latter category takes us further than the first two (Polloy and Mainprize 1984; Stern 1988). Resonance definitely has a place within the matrix of figures of rhetoric developed-by Durand (1987) and discussed in Chebat (1989) and Dyer (1982); it is a false homology.

Of course, cue, claim, metaphor and resonance are not the only devices for generating meaning in advertisement. The purpose of introducing the notion of resonance was to demonstrate how diverse these devices are, and to show that they cannot be reduced to one another. Having given some idea of this diversity, the question now becomes: What kind of theory do we need if the goal is to encompass the whole range of devices whereby meaning can be generated in advertisement? At a minimum, the desired theory or body of theory must be able to: 1) handle non-verbal as well as verbal elements; 2) encompass all the various rhetorical figures, and not just metaphor; and 3) include affective as well as cognitive meanings. Anything less condemns us to a piecemeal approach to the study of meaning in advertisement.

Perhaps the best candidate to fill this role is semiotics, as introduced to the consumer literature by Mick (1986), elaborated in Umiker-Sebeok (1987), and further developed in the works of Barthes (1985), Eco (1976), and others. We could also benefit from more works that offer extended interpretations of specific real advertisements, as done most notably by Williamson (1978; see also Dyer 1982; Leymore 1975; Liess, Kline and Jhally 1986; Vestergaard and Schroder 1985). After reading a book like Williamson's it becomes difficult to believe that atomistic ideas such as cue or claim can ever capture the sheer complexity of the meanings that an advertisement can create. As consumer researchers become more comfortable with interpretive techniques (Holbrook 1987; Hudson and Ozanne 1988), the need for theories that can bring order to interpretive activity, whether these be derived from semiotics or from some other tradition, will grow. Since marketers are among the most prolific producers of texts that cry out for interpretation, perhaps we can even hope that some day consumer researchers will become the producers of, rather than just consumers for, theories of meaning in advertisement.


Experimental investigations of meaning in advertisements are not easy to pull off, especially when one departs from a strictly cognitive paradigm. The difficulty is compounded when real rather than constructed advertisements are used. What creates the difficulty, of course, is the many threats to internal validity that have to be countered. I think the most important of these threats in this study involve the sampling of stimuli -- an intrinsically difficult problem, as noted by Nunnally (1978). In this case we have a judgement sample of stimuli: thirteen ads selected by the author to meet several criteria. In the case of such a judgement sample, it is impossible to rule out the rival hypothesis that the effects observed for certain properties are really due to some peculiarities of the ads that include those properties, and not the properties themselves. A different judgement sample might yield entirely different results.

A more vexed problem is that a small number of ads chosen unsystematically from the universe of advertisements is unlikely to contain a great deal of variation on some or all of the cues of interest. A solution would be to have judges rate a large number of ads for the presence of, say, beautiful or ugly music, and then sample from the extremes to construct, in effect, treatment cells for each cue of interest. While this would inflate the observed effects of the cue or property, it maximizes the probability that an effect will be found if in fact such an effect exists. The alternative, which is to randomly sample a sufficiently large number of ads to insure that there is enough natural variation on all properties, might lead to an extremely unwieldy array of stimuli.

Still another problem with the stimuli used is the large number of properties (9) in relation to the number of stimuli 113, later reduced to 9). Although sampling at two second intervals does create a greater number of observations, the fact remains that many of the properties are probably only present in a small number of the ads studied -- perhaps only one ad in some cases. This again makes it impossible to rule out rival explanations. If only one or two ads contain a certain cue, then any effects observed may be a function of the ad rather than the property. Hence, the table of significant and null effects must be read as preliminary and suggestive only. It offers no conclusive evidence about the affective consequences of any of the cues investigated. No amount of sophisticated data analysis can overcome these sampling problems.

I have a couple of minor points to make about aspects of the experimental procedure. It does not appear that the order of presentation of ads was varied over subjects. It is easy to imagine that subjects experienced more positive affect and perhaps more involvement with the rating task early in the procedure. If so, and if certain cues were only present in ads shown at the beginning or the end of the sequence, then the effects uncovered in the analyses are biased to an unknown degree. On another point, inter-rater reliability for judges is just as important in the case of the continuous variables as the dichotomous variables. Test-retest reliability for a single judge is really no substitute. Again, this undermines our confidence as to whether an effect does or does not exist for any of the continuous variables studied.

In conclusion, the author is to be applauded for undertaking to measure temporal variation of cues within an advertisement. The use of the joystick rating method certainly has promise, and none of the difficulties mentioned above is insuperable. However, although continuous rating such as those provided by a joystick are rare in academic consumer research, they have been utilized for many years by several commercial providers of copy testing services, such as ASI. It would be interesting to know what sort of data archives exist that might reward study with respect to temporal variation. If a large archive exists, then this might provide a solution to many of the sampling problems discussed earlier. A model for such work would be the study of executional cues by Stewart and Furze (1986).


This paper attempts to integrate literature in the areas of linguistics and semantics with the more familiar (to consumer researchers) literatures on cognitive and social psychology, and also to validate some of the semantic distinctions identified. It is nice to see this effort to go outside of the customary and familiar psychological literature. But the absence of semiotic theory is keenly felt, and I would urge the authors to expand their horizon beyond general semantics. There is an enormous European literature on some of these issues (Eco 1976), and its omission is bothersome.

The fundamental difficulty I have with this work concerns the usefulness of the semantic trichotomy. Are these really three equally fundamental and tightly linked concepts that have to be analyzed as a unit? The connection between a report and an inference seems much more integral than that between a judgement and either one. I doubt that the three types of claim lie on a single continuum, nor do they seem to be related in any geometric or combinatory way, such as primary colors in a color wheel. I would ask the authors: What do we gain by treating all three together? Reports, inferences, and judgments are certainly of fundamental interest to advertisers, just as the authors argue; but to speak of a trichotomy is to suggest that these represent an exhaustive and mutually exclusive list of all the fundamental types of claims that can be made in an advertisement. I am not convinced that this is so.

The linguistic approach that underlies the semantic trichotomy also seems unduly limiting. Too much of the meaning generated in advertisements is both non-verbal and tacit. The problem is particularly acute in the case of television advertisements, where the question becomes what is claimed, and how, and by whom? Should we focus on the sponsor, the presenter, or the text (broadly construed) of the ad itself? The inference category is particularly problematic. Whose inferences should we be studying: those offered about the product within the ad, or those made by the respondent in response to the ad? I would have appreciated the inclusion of real advertisements analyzed according to the author's scheme. The lack of concrete examples drawn from specific advertisements made the exposition difficult to follow at a number of points.

I would also point out what strikes me as an important flaw in the second experiment described. The whole point of manufacturing stimulus materials in advertising research is to exert the tightest possible control over what is being presented to subjects so that causal inferences are facilitated. The examples in Table 3 do not satisfy this standard; they seem to me fatally prolix. Too much is going on that is neither report,.inference nor judgement, or that represents multiple occurrences of one or another of these. Hence, rival explanations for why subjects were able to successfully sort the statements can be offered. I suspect that the sort was done in terms of the extremity or the degree of exaggeration in the claims, and not because the three types of statements reflect some emic distinction made by respondents among types of advertising claims. I would also note in passing my skepticism that anyone in the study really believed that a research physician would use language such as that used in the Judgement text with respect to an over the counter drug.

In conclusion, I think the components of the semantic trichotomy are well worth investigation on their own, particularly in the context of research on misleading or deceptive advertising, or on the persuasive impact of each of these types of claims. In my opinion there would be more value in such pursuits than in attempting to establish the semantic trichotomy as a fundamental organizing principle for claims made about products.


I have less to say about this paper because personally I found it the most satisfying of the three. Again the authors have gone out and tapped into literatures not commonly discussed or cited at ACR. The value added is enhanced because they then go further and integrate some of this work into three distinct models of how metaphors work. The discussion of the models is balanced, illuminating, and rich in implications for future research. I congratulate the authors on a fine review paper.

I would like to second their comment about the role of the marketing function in producing a substantial percentage of the metaphors to which people are exposed. That says to me that consumer researchers potentially can make a significant contribution to the study of metaphors; it is home territory for us. Perhaps this is an area where the positivistic tradition, with its understanding of experimental design, and the interpretivist tradition, with its firm grasp of the role of the reader in generating meaning, can usefully come together. Both will be necessary to understand what metaphor does or can do in advertisements.

I want to repeat my comment that investigations that do not clearly distinguish metaphor from other types of figurative speech are fraught with danger. We need both: general theories about how figurative speech works, and specific theories about how individual tropes or figures such as metaphor work. But if we do not keep the two levels of explanation separate we will have problems. Hence, a useful test to apply to each theoretical model of metaphor might be: Can this be made to work as an explanation for other rhetorical figures as well? If the answer is yes, then we are probably looking at an explanation of figurative speech rather than an explanation of metaphor per se. For such a general theory, the question would become: How many types of figures does this model illuminate? The problem here is that authors outside the literary tradition tend to be ignorant of the great variety of figures and tropes examined over the years. This is a natural but potentially dangerous blind spot.

One other point is relevant to consumer research on metaphor, and that is the changed meaning of "aptness" when we move from literature and everyday speech to paid messages devised by a corporation to further its goals of profitability, etc. In the traditions summarized by the authors, "aptness" represents an aesthetic judgement. In consumer research aptness will have a second meaning: effectiveness in advancing the sponsor's corporate goals. This raises several questions for research. Are aesthetically apt metaphors also more effective from a managerial standpoint? Does a metaphor have to be apt in an aesthetic sense to effectively serve the manager's pecuniary goals? We can imagine metaphors which are bad from any poetic standpoint but which still do the job asked of them by the person paying for the ad. These are questions about metaphor that may be unique to the consumer context. If we don't pursue them, no one else will.


Barthes, Roland (1985), "The Rhetoric of the Image," in The Responsibility of Forms, New York: Hill and Wang, 21-40.

Durand, Jacques (1987), "Rhetorical Figures in the Advertising Image," in Marketing and Semiotics: New Directions in the Study of Signs for Sale, ed. Jean Umiker-Sebeok, New York: Mouton de Gruyter, 295-318.

Dyer, Gillian (1982), Advertising as Communication, New York: Methuen.

Eco, Umberto (1976), A Theory of Semiotics, Bloomington, IN: Indiana University Press.

Grinnell, J.D. (1987), National Register of Advertising Headlines & Slogans, Westbury, NY: Asher-Gallant.

Holbrook Morris (1987), 'The Study of Signs in Consumer Esthetics: An Egocentric Review" in Marketing and Semiotics: New Directions in the Study of Signs for Sale, ed Jean Umiker-Sebeok, New York: Mouton de Gruyter, 73-122.

Leymore, Varda L. (1975), Hidden Myth: Structure and Symbolism in Advertising, New York: Basic Books.

Liess, William, Stephen Kline and Sut Jhally (1986) Social Communication in Advertising: Persons, Products and Images of Well-Being, Toronto: Methuen.

McQuarrie, Edward F. (1989), "Advertising Resonance: A Semiological Perspective," in Interpretive Consumer Research, ed. Elizabeth C. Hirschman, Provo, UT: Association for Consumer Research, 97-114.

McQuarrie, Edward F. (1990), "Advertising Resonance: Criteria for Effective Use," unpublished manuscript, Santa Clara University, Santa Clara, CA.

Mick, David Glen (1986), "Consumer Research and Semiotics: Exploring the Morphology of Signs, Symbols and Significance," Journal of Consumer Research, 13 (September), 196-214.

Nunnally, Jum C. (1978), Psychometric Theory, 2nd ed., New York: McGraw-Hill.

Pollay, Richard W. and Steve Mainprize (1984), "Headlining of Visuals in Print Advertising: A Typology of Tactical Techniques," paper presented at the American Academy of Advertising, Denver.

Stern, Barbara B. (1988), "Figurative Language in Services Advertising: The Nature and Uses of Imagery," in Advances in Consumer Research, Vol. 15, ed. Michael J. Houston, Provo, UT. Association for Consumer Research 185-190.

Stewart, David W. and David H. Furze (1986), Effective Television Advertising: A Study of 1000 Television Commercials, Lexington, MA: Lexington Books.

Umiker-Sebeok, Jean (1987), Marketing and Semiotics: New Directions in the Study of Signs for Sale, New York: Mouton de Gruyter.

Vestergaard, Turben and Kim Schroder (1985), The Language of Advertising, New York: Basil Blackwell.

Williamson, Judith (1978), Decoding Advertisements: Ideology and Meaning in Advertisements, Boston: Marion Boyars.



Edward F. McQuarrie, Santa Clara University


NA - Advances in Consumer Research Volume 17 | 1990

Share Proceeding

Featured papers

See More


Thou Shalt Not Look! When Processing the Odds Visually Biases Gambling Behavior

Rod Duclos, Western University, Canada
Mansur Khamitov, Nanyang Technological University, Singapore

Read More


Thank You = Trust Me: When Gratitude Expressions Help Promote New Products

Maria Ng, University of Houston, USA
Dejun Tony Kong, University of Houston, USA
Vanessa Patrick, University of Houston, USA

Read More


To Touch or Not to Touch?: How Touch Influences Decision Confidence

Sang Kyu Park, University of Florida, USA
Yang Yang, University of Florida, USA

Read More

Engage with Us

Becoming an Association for Consumer Research member is simple. Membership in ACR is relatively inexpensive, but brings significant benefits to its members.