Rhetorical Strategies in Advertising

John Deighton, Dartmouth College
ABSTRACT - While some advertising acts on pre-experience mental states like awareness or intention to purchase, other advertising influences what the consumer takes from the consumption experience itself (so-called transformational advertising). This paper applies rhetorical theory to identify attributes of the advertisement which must be present for transformation/embellishment of the consumption experience to occur. It proposes a typology of advertising arguments built upon characteristics of argument form and structure.
[ to cite ]:
John Deighton (1985) ,"Rhetorical Strategies in Advertising", in NA - Advances in Consumer Research Volume 12, eds. Elizabeth C. Hirschman and Moris B. Holbrook, Provo, UT : Association for Consumer Research, Pages: 432-436.

Advances in Consumer Research Volume 12, 1985      Pages 432-436


John Deighton, Dartmouth College


While some advertising acts on pre-experience mental states like awareness or intention to purchase, other advertising influences what the consumer takes from the consumption experience itself (so-called transformational advertising). This paper applies rhetorical theory to identify attributes of the advertisement which must be present for transformation/embellishment of the consumption experience to occur. It proposes a typology of advertising arguments built upon characteristics of argument form and structure.


Some experiences resist verbal description. They may generate what Bem (1972) called "weak, ambiguous, or uninterpretable" cues. Or they may offer strongly felt cues which fit no ready categories in language because the cues are new, unfamiliar or unaccepted (Schachter 1964, Valins 1966). This paper is concerned with the role of advertising in making sense of consumption experiences of these kinds. The argument is that there is a class of advertising, that following Wells ( 1983) can be called T transformational', whose function is to prefigure ambiguous consumption experience and shape the meaning consumers give to it. This paper aims to identify the rhetorical characteristics of communications that affect such sense-making.

Advertising and Sense-Making

Webster (1971) proposed that the function of advertising was "to convey meanings". Two recent papers have argued that some advertising may go further and construct meanings, by influencing the conscious apprehension of experience. Deighton (1984) refers to 'directed inference'. Puto and Wells ( 1983) use the concept of 'transformational advertising' to characterize communications whose effect is to embellish the consumption experience, distinguishing this kind of communication from 'informational' advertising.

These papers claim that, while some advertising merely generates awareness of choice options or supplies reasons to try one over another, some advertising can affect what is experienced when the product is consumed. Puto and Wells ( 1983) ascribe this effect to vicarious classical conditioning, by which the portrayal of a social model's responses to the product induces the same responses in the observer, and to Clynes' ( 1980) model of generalized emotion, by which the advertisement evokes an emotional state and attributes it to the advertised product. Deighton ( 1984) attributes it to heuristic-driven, confirmatory processing of experience cues driven by advertising-induced expectations. In this view the advertisement affects what is attended to, and what is recalled from, the experience of using the product.

This paper, however, is not primarily concerned with the psychology of the phenomenon. If the distinction between informational and transformational communications exists, then it ought to be possible to recognizeS in the advertisements themselves, the rhetorical components that account for the distinction. This paper therefore develops a coding scheme for a content analysis of advertising as part of a general investigation of the issue. It identifies attributes of argument structure and form which, from research in linguistics and logic, appear to be necessary conditions for the transformation effect to occur.

Argument structure

Advertising is argumentation, in the sense that it intends to influence an audience with substantiated assertions. An analysis by Toulmin ( 1958) suggests a dimension of argument structure that may account for some advertising's ability to construct meanings, rather than simply present information. The reader is referred to Toulmin's 1958 thesis and to Toulmin, Rieke, and Janik (1979) and Bettinghaus (1966), for fuller treatments of this topic.

Toulmin's concern is how should an argument be laid out if we want to locate the sources of its validity? Three elements of his structure are a claim, grounds and a warrant or license to infer from the grounds to the claim. The claim is a conclusion whose merits must be established. The grounds are the specific data, not in dispute, to which the arguer appeals. The warrant is a general principle, rule or law by which the grounds support the claim. If the warrant is disputable, then the argument will contain backing. The function of this element of the structure is to supply grounds for the authority of the warrant. If the claim is qualified, the argument may contain a modality, to specify the claim as necessary or probable or presumable, and the warrant may contain a rebuttal, to record the conditions which might defeat its general authority. Figure 1 illustrates this structure.



The structure of the advertisement for Tylenol in Appendix A is easily captured in this model, as Figure 2 illustrates. The ad in Appendix B presents more of a challenge to the model. This ad relies less on verbal argument. However as Figure 3 indicates it can be accommodated. The elements of the American Tourister ad that make it difficult to model as reasoned discourse are (1) that a large part of the grounds is presented not in words but in a metaphorical visual image (the gorilla abusing the case), (2) that the warrant is unstated and must be inferred if the claim is to hold, and (3) that it is the warrant that seems to be at issue, and not the claim





Toulmin (1958, p. 120) anticipates this third difficulty. He distinguishes between warrant-using and warrant-establishing arguments. The Tylenol ad represents the former: the warrant is a matter of common currency. It is the experience of Ann Johnson as grounds for the claim that make the claim compelling. The American Tourister ad is an example of the latter. Toulmin describes a warrant-establishing argument as one which makes clear the acceptability of a novel warrant by applying it to cases in which both the grounds and the conclusion have been independently verified.

This distinction parallels the distinction between deductive and inductive reasoning. The Tylenol ad argues deductively, from a fact about hospitals (they prefer Tylenol) and a concrete instantiation of that fact, to the conclusion that Tylenol can be trusted. The American Tourister ad invites an inductive inference, from the vivid instantiation of a suitcase's rough treatment (the grounds) and the demonstrated survival of the case (the claim), to the warrant that the case is durable.

Notice in the case of an argument which persuades inductively that the ad does not interpret experience, it effectively supplies the experience. If the ad persuades, it is as if the event has actually occurred. Subsequent experience is assimilated incrementally: if one sees a newlooking American Tourister suitcase at an airport, it is not unreasonable to decide that the case may actually be quite old, and therefore a confirming instance of the advertised proposition.

Whether an advertisement establishes a warrant or uses a warrant, thus, indicates whether the argument appeals to evidence or constitutes the evidence. It is the latter case which can be described as transformational: the advertisement either adds to or interprets past experience, or supplies a datum to which future experience will be assimilated.

Rhetorical Form

The previous section has proposed that argument structure determines whether a communication influences the interpretation of experience. This section examines how argument form can do so. Among the many aspects of form which are rhetorical in their purpose is the use of tropes. Tropes, or figures of speech, are necessary when the domain of discourse has no well-established language conventions. When experience has been schematized, as for example in a mature science like physics, discourse has well-defined terms to employ. Discussion is perhaps colorless, but it is free of surprises. When language is set the task of conveying less well-disciplined experiences, or when the aim of discourse is to change the schema, then the objects of communication must be evoked creatively, often by reference to similar or contiguous objects. These unexpected substitutions can be classified in terms of the four 'master tropes' (Burke 1962): metaphor, metonomy, synechdoche and irony.

Metaphor is the tactic of denoting a concept by a word or phrase with which it shares some points of similarity. A metaphor succeeds as a way of telling when the respondent recognizes the intended analogy when a train of inference is initiated by the rule that if the things agree in some respects, they will agree in others. Metaphor abounds in advertising. American Tourister asserts that busboys are gorillas when they handle luggage. IBM denotes the naive personal computer user by the Chaplin tramp. Merrill Lynch invoked the thundering herd metaphor, found that some of the inferences it induced were counter productive, and modified it to 'a breed apart'.

It is not only in media advertising that a marketer is concerned to direct the construction of meaning. The adoption by consumers of new products can be smoother if metaphor is used to build continuity between past experience and the new phenomenon and if product design is careful to have analogical inferences confirmed in use. Designers of computer systems have been quite explicit about the need to control inference by metaphor. Office software has employed the pointer, the menu, the desk and more recently the kitchen as metaphors for the control interface. In a particularly insightful paper, Clanton (1983) proposes of software, "The device should document itself. It should create the metaphor that directs its use, then give feedback that allows the user to learn more." The tactic of presenting the product to the consumer metaphorically, as something it is not, is clearly not deceptive - it is a powerful way to induce comprehension and speed adoption.

Synecdoche employs an attribute or part of the referent to stand for the referent. The effect is to make salient some aspect of the referent at the expense of others, and particularly to direct the thinker to comprehend a mere aspect as a representative aspect. The consequences for inference are suggested by Kahneman and Tversky's (1972) discussion of the representativeness heuristic.

While synecdoche focuses comprehension by reduction, metonomy does it by integration. It denotes the referent not by one of its parts, but by some whole to which it is physically or existentially linked. It makes an abstract referent more vivid or concrete, and gives it attributes that belong to the other. Thus when 'the White House' is used to denote the executive branch of government, inferences are made possible that depend on properties of a building (solidity, prominence, endurance) that the abstract referent may not share.

Irony, in which 'the author. signals in advance a real or feigned disbelief in the truth of his own statements' (White 1973), turns figurative language against itself to communicate the unsoundness of the whole image-building process. Ironic language invites the audience to see through the figuring tactic to some more sophisticated and often cynical view of reality. Irony is rare in advertising, because it is potentially destructive of the whole shared enterprise. Ironic advertising, such as for example the 'lemon' campaign for Volkswagen in the 1970's, depends for its success on the excesses of other advertisers. Furthermore it is only rarely a good idea to foster a cynical interpretation of consumption experiences.

The general claim in this section of the paper is that tropes prefigure the apprehension of reality and so shape the sense that is made of it. A study by Glick and Holyoak (1983) is evidence for that claim. The counterpart of experience in this study was a -problem to be solved by the subjects, and the sense that was made of the experience was revealed by whether or not subJects solved the problem. The study investigated whether exposure to an analogous solved problem (corresponding to a solution metaphor) would affect the rate of problem solving. The finding was that the rate of problem solving was higher only when at least two different analogs were presented. The authors attributed this result to a process by which exposure to two analogs drew subjects' attention to the critical similarities. These recognized similarities became the elements of a new schema for the comprehension of subsequent experience which now included a solution element.

Subjects who saw the two analogs were able to grasp the metaphor, and perceived the target problem in a different and more productive way.

The Glick and Holyoak study is a rare example of experimental analysis of the effect of communication on sense-making. It suggests a paradigm for systematic investigation of this important issue in advertising.


This paper has made two distinctions among persuasive communications, a distinction of structure between warrant-establishing and -using arguments, and a distinction of form between figurative and literal language usage. Some evidence and argument have been advanced to link both of these attributes to the transformation effect, suggesting that necessary (but of course not sufficient) conditions for a transformation effect are the presence of warrant-establishing structure and figurative form.

Figure 4 sets out a taxonomy of arguments that results from this set of distinctions.



In this table, a report is an argument which proceeds inductively (its grounds and its findings are used to support the warrant or hypothesis which is in dispute) and has at its disposal a well-defined set of terms to describe experience. Perhaps the purest example of this mode of argument is the scientific paper recording the test of an hypothesis. If the same set of terms were applied to argue deductively (to use or draw implications from, rather than to support, the warrant) the result would be a syllogism. If the structure lacked a well-schematized language but retained its deductive structure, the argument could be characterized as persuasion by allegory or fable: it would argue for a claim by deduction, using symbolic terms.

In the final cell (transformation/myth), the object of the communication is not to dispute a claim. It deals with the value or warranty to be given to an event, when the event is recounted figuratively. In argumentation by myth, the story transforms the event's meaning. Eclipses and thunder are common examples of events whose experiencing has been transformed in many cultures by mythic argument.

Each of these modes of argumentation is encountered in advertising. The simple announcement ad, common in retailing and in the marketing of prosaic industrial products, is usually syllogistic (though often with the minor premise implied). Comparative advertising for appliances and industrial equipment usually adopts the report mode to establish the warrant of product superiority. These two modes of advertising have the advantage that communicator and receiver share a common mapping of experience into language. This kind of advertising is seldom accused of being manipulative (though it may be accused of being wrong) because the argument is explicit. On the other hand, mythic/ transformational or allegorical advertising often arouses public concern, especially when it is applied to 'important' issues like politics or expensive consumer product choices. Note, however, that the choice of the figurative form is inevitable when there are no literal terms in which to articulate reality. Similarly, warrant-inducing advertising is not of itself manipulative: some events simply are open to more than one interpretation.

The hypothesis derived in this paper is that advertising capable of transforming the consumption experience will have recognizably different characteristics from advertising capable only of influencing beliefs, intentions, expectations, or other pre-experiential mental states, and that these characteristics include form (the use of figurative discourse) and structure (warrant-establishing argument) A test of this hypothesis will make progress toward a typology of advertising tasks and illuminate the question of the various ways in which advertising works.






Bettinghaus, Erwin P. (1966), "Structure and Argument," Perspectives on Argumentation edited by Gerald R. Miller and Thomas R. Nilsen. Chicago: Scott, Foresman and Company.

Clanton, Chuck (1983), "The future of metaphor in man-computer systems," Byte pp. 263-279.

Clynes, Manfred (1980), "The communication of emotion: Theory of sentics" in R. Plutchnik and H Kellerman (Editors), Emotion: Theory, Research, and Experience. New York: Academic Press.

Deighton, John (1984), "The interaction of advertising and evidence," Amos Tuck School of Business Administration Working Paper Series.

Glick, Mary L. and Keith J. Holyoak (1983), "Schema induction and analogical transfer," Cognitive Psychology, Vol. 15, pp. 1-38.

Kahneman, David and Amos Tversky ( 1972), "SubJective probability: A judgment of representativeness," Cognitive Psychology, Vol. 3, pp. 430-454.

Puto, Christopher and William D. Wells (1983), "Informational and transformational advertising: The differential effects of time," Advances in Consumer Research. Association for Consumer Research

Schachter, Stanley (1964), "The interaction of cognitive and physiological components of emotional state," in L. Berkowitz (Editor), Advances in Experimental Social Psychology Volume 1. pp. 49-80, New York: Academic Press

Toulmin, Stephen E. (1958), The Uses of Argument. London: Cambridge University Press.

Toulmin, Stephen E., Richard Rieke and Allan Janik (1979), An Introduction to Reasoning. New York: Macmillan Publishing Co., Inc.

Valins, Stuart (1966), "Cognitive effect of false heart-rate feedback," Journal of Personality and Social Psychology, V. 4, pp. 400-408.

Webster, Frederick E. (1971), Marketing Communications. New York: Ronald Press,

Wells, William D. (1983), "How advertising works." Mimeographed. Needham, Harper and Steers Advertising. Inc. Chicago.

White, Hayden (1973), Metahistory: The Historical Imagination in Nineteenth Century History. Baltimore: The John Hopkins University Press.