Journal of Consumer Research, Inc.

Who is the Celebrity Endorser? Cultural Foundations of the Endorsement Process Author(s): Grant McCracken Reviewed work(s): Source: Journal of Consumer Research, Vol. 16, No. 3 (Dec., 1989), pp. 310-321 Published by: The University of Chicago Press Stable URL: http://www.jstor.org/stable/2489512 . Accessed: 12/02/2012 13:26
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

The University of Chicago Press and Journal of Consumer Research, Inc. are collaborating with JSTOR to digitize, preserve and extend access to Journal of Consumer Research.

http://www.jstor.org

Who

Is

the

Celebrity of

Endorser?

Cultural Process

Foundations

the

Endorsement

GRANT McCRACKEN*
This article offers a new approach to celebrity endorsement. Previous explanations, especially the source credibility and source attractiveness models are criticized, and an alternative meaning transfer model is proposed. According to this model, celebrities' effectiveness as endorsers stems from the cultural meanings with which they are endowed. The model shows how meanings pass from celebrity to product and from product to consumer. The implications of this model for our understanding of the consumer society are considered. Research avenues suggested by the model are also discussed.

he celebrity endorser is a ubiquitous feature of modern marketing. The actor Robert Young, the quarterbackJim McMahon, the dancer Mikhail Baryshnikov,the CEO Lee lacocco, the singer Whitney Houston, the test pilot Chuck Yeager, and the politician Tip O'Neill have all lent name and image to recent campaigns.Unfortunately, the popularityof this communications strategyhas not earned it extensive 'study.Nor has it inspired especially illuminating theoretical accounts. As a result, the received wisdom on celebrity endorsement is modest and imperfect, and existing models fail to capture several of the most interesting and central characteristicsof the endorsement process. This investigation of endorsement addresses these deficits from a cultural perspective. The argument is that the endorsement process depends upon the symbolic properties of the celebrity endorser. Using a "meaning transfer"perspective, these properties are shown to reside in the celebrity and to move from celebrityto consumer good and from good to consumer. This perspectiveis then used to addresscontroversial issues concerning the consumer society. Finally, research opportunities are reviewed. For present purposes, the celebrity endorser is defined as any individual who enjoys public recognition and who uses this recognition on behalf of a consumer good by appearing with it in an advertisement. We will refine this definition later, but for the moment it is deliberatelybroad to encompass not only the usual
*GrantMcCrackenis Associate Professor,Departmentof Consumer Studies, University of Guelph, Guelph, Ontario, Canada NIG 2W1. He wishesto thankPeterBennett,TaraCurtis,Bill Frisbee, Ron Goldman, Monty Sommers,Vic Roth, Donna Woolcott, and Rami Zwich as well as the anonymous JCR reviewersfor their very useful comments on earlierversionsof this article. 310

T

movie and television stars, but also individuals from the world of sport, politics, business, art, and the military. The term "celebrity"is also meant in this article to encompass a variety of endorsements, including those in the explicit mode ("I endorse this product"), the implicit mode ("I use this product"), the imperative mode ("You should use this product"), and the copresent mode (i.e., in which the celebrity merely appears with the product). Moreover, it includes a rangeof endorsement roles, such as cases in which the celebrity is also an expert (e.g., Bobby Unser recommending motor oil), is associated with the manufacturerin some long term capacity (e.g., Pat Summerall for TrueValue Hardware), or has no special knowledge of, or association with, the product in question (cf. Friedman, Termini, and Washington 1977). This definition is designed deliberately to exclude the "typical consumer" endorser (Friedman and Friedman 1979). The model presented in this article applies to all these variations but the last.

THE LITERATURE
Two models, the source credibility and the source attractivenessmodels, inform researchand reflection on the topic of celebrity endorsement. Both were devised originally for the study of communications and have been applied only latterly to the endorsement process. Both are designed to determine the conditions under which the message sender or source is persuasive.

The source credibility model rests on research in social psychology (Hovland and Weiss 1951-1952; Hovland, Janis, and Kelley 1953). The Hovland version of the model contends that a message depends for its effectiveness on the "expertness" and "trustworthiness" of the source (Hovland et al. 1953, p. 20; cf. Dholakia and Sternthal 1977; Sternthal,Dholakia,
? JOURNAL OF CONSUMER RESEARCHo Vol. 16 * December 1989

CELEBRITY ENDORSER: CULTURAL FOUNDATIONS

311

and Leavitt 1978). Expertness is defined as the perceived ability of the source to make valid assertions. Trustworthiness is defined as the perceived willingness of the source to make valid assertions. The Hovland model holds that sources exhibiting expertness and trustworthiness are credible and, to this extent, persuasive. triggerthe past associations with the sponsor and stimulate the rememberingof past messages. He would integratenew messageswith the old so as to build a unifying, coherent, sustained, and consistent image of the brand.

The source attractivenessmodel also rests on social psychological research. The McGuire (1985) model contends that a message depends for its effectiveness chiefly on the "familiarity," "likability," and/or "similarity" of the source (McGuire 1985, p. 264; cf. Baker and Churchill 1977; Debevec and Kernan 1984; Friedman, Santeramo, and Traina 1978; Joseph 1982; Kahle and Homer 1985). Familiarityis defined as knowledge of the source through exposure, likability as affection for the source as a result of the source's physical appearanceand behavior, and similarity as a supposed resemblance between the source and receiver of the message. The McGuire model holds that sources who are known to, liked by, and/ or similar to the consumer are attractive and, to this extent, persuasive. The source models (as we shall call the source credibility and source attractivenessmodels together)have been confirmed by research.The Hovland model has been validated by several parties (Atkin and Block 1983; Kamen, Azhari, and Kragh 1975; Klebba and Unger 1983; cf. Finn 1980, p. 779). The McGuire model also demonstrated its value (Friedman and Friedman 1979), and it appearssafe to say that celebrities owe some of their effectivenessas marketingdevices to their credibility and attractiveness. The source models are, to this extent, a necessary part of our understandingof the endorsement process. But, they do not captureeverythingat issue in the endorsement process. Indeed, there is reason to think these models cannot explain endorsement's most fundamental features. The evidence for this skepticism is everywhere.The researchitself is littered with puzzles and peculiaritiesthe source models cannot explain. For instance, the research by the Friedmans produced results that are not consistent with the source models. They found that some product categories were incompatible with celebrity categories (e.g., that MaryTyler Moore served as a poor celebrity endorser for vacuum cleaners). But the source models make no such provision. For the models' purposes, as long as the credibility and attractivenessconditions are satisfied, any celebrity should serve as a persuasivesource for any advertisingmessage. According to the model, the persuasiveness of the celebrity has everything to do with the celebrity and nothing to do with the product. Kamen, Azhari, and Kragh (1975, p. 18) suggest that the spokesperson acts as a kind of "core around which the substantive messages are positioned." In this capacity, the spokespersonhelps

This position implies that the celebrity serves the endorsement processby taking on meanings that then carry from ad to ad, and that the celebrity is capable somehow of serving as a site in which meanings cohere. Plainly, neither possibility is consistent with either source model. After all, these models make assertions only about the credibility and attractiveness of the message sender and none about the endorser's role as a message medium or the continuity of the message from ad to ad. In the language of Kuhn (1962), the paradigm is beginning to accumulate anomalies. Scholars have been compelled either to abandon or transformthe source models. But if the internal evidence for skepticism is strong, the external grounds are even stronger.The scholarly and professional literature is littered with data that cannot be explained by the source models. There are mysteries everywhere. Bill Cosby failed as an endorser for E.F. Hutton despite his evident success for Kodak and Coca Cola. John Houseman failed as an endorser for McDonald's despite his effective work for Smith Barney (Marshall 1987). George C. Scott proved, mysteriously, to be the wrong choice for Renault, as did Ringo Starr for Sun Country Classic wine coolers (Motavalli 1988). The source models, as the present guardian of current endorsement practice, did not forewarnadvertisingpractitionersof the inappropriatenessof these celebrity choices. Nor can they, as the received academic wisdom on the endorsement process, help us understand what went wrong. The source models have not servedas a practical or theoretical guide to celebrity endorsement. Consider, for instance, the example of John Wayne as a celebrity endorser for the pain reliever Datril. "Waynehad nothing to do with the product, and sales of the analgesic languished.. . . (It was a) classic .

mismatch between star and product" (Kaikati 1987, p. 6). This is offered as a kind of explanation of what went wrong in the Datril case. But what does it mean to say that the celebrity "had nothing to do" with the product? What "mismatch" between celebrity and product is being assertedhere?The source models do not tell us. They cannot explain why John Wayne and Datril were incompatible. Schudson's treatment of James Garner as a celebrity endorser is germane. Schudson (1984, p. 212) suggeststhat there is something mysterious about the advertisementsin which Garnerappears.
Garner does not play himself, the person, nor does he play a particular fictive character. Instead, he plays what I would call the generalized James Garner role, the type for which James Garneris alwayscast-handsome, gentle, bumbling, endearing, a combination of

312

THE JOURNAL OF CONSUMER RESEARCH

Bret Maverick from "Maverick" and Jim Rockford from "The RockfordFiles."

This thoughtful observation spells real trouble for the source models. If the celebrity endorserrepresents not himself but his stage persona, the issues of expertness and trustworthinesscan hardlyapply. After all, it hardlymakes sense to impute credibility to a fictional character.But even if we force the issue and insist that fictional characters can, somehow, be credible, Schudson's observation tells us this is a special, rolespecific credibility. It is no longer a simple matter of the willingness and ability to make valid assertions. The issue of source attractiveness is problematical in a differentway. According to Schudson's account, Garner's attractiveness consists in the "endearing," "gentle," and "bumbling" qualities of his stage persona. We must add to this perhaps the most salient trait of Garner'sstage persona, his claim to being the foremost representative of a particular category of American male. As a prototype of the category, Garner is a member of the largerpantheon of actors that helps define this gender category in America (e.g., James Stewart, John Wayne, Cary Grant, Sylvester Stallone). When we observe Garner from this point of view, we see that his attractivenessdepends not on his qualities as a person or even on his qualities as a famous person, but on the qualities he has created in his stage persona. For communications purposes,the celebrity is a composite of his fictional roles. This means that when consumers respond to Garner's "attractiveness," they are, in fact, respondingto a very particular set of meanings. They are identifying with a bundle of symbolic propertiescreated for, and by, Garnerin the television programs"Maverick"and "The Rockford Files." This is not "identification" in the ordinary sense. The source models do not capture and illuminate what is going on here. Audience response to James Garneris more complicated and interesting than the source models allow. Garner is persuasive as a communicator not only because he is "attractive," but also because he is made up of certain meanings the consumer finds compelling and useful. Garner succeeds as an endorserfor Mazda because he represents a bundle of meanings about maturity, Americanness, confidence, masculinity, intelligence, and good humor. It is here that we begin to uncover the real insufficiency of the source models as an account of celebrity endorsement. The source attractiveness model can tell us that consumers will identify with Garner, but it cannot tell us why-nor can it contend with the meanings contained in Garner's persona. Still more important, the model does not allow us to make sense of the meanings contained in a celebrity endorser once they are determined. The source model can tell

us only that a celebrity is attractive, not what attractive is. The implications of this insufficiency are powerful. First, the source models do not allow us to understand the appeal of any particular celebrity. This makes it impossible to understandwhy a celebrity like Garner should be persuasive for some products but not for others. The source models prevent us from identifying the matches and mismatches. We are left unable to assess how Garner'simage interacts with different products and creative themes. Second, the source models will never allow us to discriminate between celebrities in any useful way. Certainly they allow us to say that James Garner is, perhaps, more credible than Alan Alda. But it does not allow us to say how Garner and Alda differ from a symbolic or communications point of view. The source models might tell us only that Michael J. Fox, Tony Danza, and Don Johnson differ in their degree of attractiveness. This is problematical because we understand that their differences go much deeper than this. Hypothetically, the source models might tell us that Cybil Shepherd,Bea Arthur,and Joan Collins are equally credible. But we know this sameness masks profound and thoroughgoing differences. In short, the source models tell us about degrees of attractiveness and credibility when what we need to know about is kinds of attractivenessand credibility. Both the internal and external evidence contain anomalies that demonstrate the insufficiency of the source models. If we are to understand the endorsement process, we must build better, more sophisticated models. We especially must come to terms with the meanings contained in the celebrity and give an account of how these meanings serve the endorsement process. The remainderof the article is designed to suggest such an account.

CULTURAL MEANING AND THE CELEBRITY ENDORSER
The effectiveness of the endorser depends, in part,-upon the meanings he or she brings to the endorsement process. The number and variety of the meanings contained in celebrities are very large. Distinctions of status, class, gender, and age, as well as personality and lifestyle types, are represented in the pool of available celebrities, putting an extraordinarily various and subtle pallet of meanings at the disposal of the marketingsystem. For example, class and status are representedby the likes of Peter Jennings and John Forsytheas patrician men and CatherineDeneuve and Audrey Hepburn as regal women. The distinction of new wealth is contained in genteel versions such as Pierce Brosnan and Diane Sawyer or in more grasping versions such as Larry Hagman and Joan Collins. The upper middle class is represented by Tim Matheson and Shelley

CELEBRITY ENDORSER: CULTURAL FOUNDATIONS

313

Long, the middle class by John Ritter and Christie Brinkley, and the lower middle class by Patrick Swayze and Suzanne Somers. Culturalcategoriesof gender and age are also represented in the celebrity endorser. One extreme representation of maleness is establishedby the likes of Sylvester Stallone, the other by the likes of Dick Cavett. Ranging between them are Hulk Hogan, Arnold Schwarzenegger,Brian Bozworth, Fred Dryer, Tony Danza, Stacey Keach, Paul Newman, Patrick Duffy, Timothy Hutton, Bob Newhart, Tony Randall, and Jeremy Brett. For women, Loni Anderson and Sigourney Weaverrepresentthe extremes of the continuum of their gender. Ranging between them are Cheryl Ladd, Victoria Principal, Cheryl Tiegs, Pam Dawber, Kate Jackson, and Jane Seymour. Age categories range from the militantly youthful Pee Wee Herman to the prematurely ancient Danny DeVito, or from the callow youth representedin Judge Reinhold to the wisdom gained by age in E.G. Marshall. These age categories are subject to change, as Gary Coleman and Angela Lansburyhave recently demonstrated. In addition to these demographiccategories,the celebrity world also contains a range of personality types. The curmudgeon is represented by Edward Asner, the rake by John Larroquette,the irritableincompetent by John Cleese, the bewildered alien by Bronson Pinchet, the good hearteddimwit by Woody Harrelson, the irrepressiblyimpudent by Tracey Ulmann, the indiscriminately jolly by Ed McMahon, the irascible by David Letterman, the blandly agreeable by Gary Collins, and so on. The celebrity world also offers a range of lifestyle types. The quintessential yuppie is perhaps Ed Olin of"Thirtysomething." The stereotypic young professional woman was once Mary Tyler Moore and is now, perhaps, Pam Dawber. The perfect Dad was once Robert Young and now may be Michael Gross or Bill Cosby. The perfect princess is representedby Delta Burke of "Designing Women," the working class hero by Carroll O'Connor, the man of wisdom and experienceby David Brinkley or CharlesKerault. Here, too, the range and depth of representation is extensive. This review oversimplifies celebrity meanings. Even the most heavily stereotyped celebrity represents not a single meaning, but an interconnected set of meanings. Cher offers a useful case in point. It is possible to locate her on all the dimensions noted. She is low to middle class in her status meaning, located toward the "hot" end of the gender continuum, and clearlyyouthful in attitude if not age. The personality is extrovertedand outspoken, the lifestyle open, freewheeling, and alternative. But, plainly, none of these dimensions by itself captures the meanings with which Cher is charged or, more importantly, the essential configurationof meanings she bringsto the en-

dorsement process. For this, it is necessaryto characterize the whole person. Cher is hip, risk taking, individualistic, sensual, sexual, expressive, irreverent, and liberated. It is this larger package of meanings playing off one another that defines Cher. These meanings enter into the endorsement process when Cher speaks, for instance, for Baly-Matrix health clubs or her own perfume. These, then, are some of the meanings contained in the celebrity world. They are reviewed here in a cursory, undocumented way. An exact assessment of these meanings awaits empirical study and theoretical development. But enough has been said to indicate that the celebrity world is something richer and more complicated than a collection of merely credible or attractiveindividuals. It is, I would argue, precisely the meanings of the celebrity that makes him or her so useful to the endorsement process. For an endorsement succeeds when an association is fashioned between the cultural meanings of the celebrity world, on the one hand, and the endorsed product, on the other. Not all endorsements succeed in this transfer. Indeed, some are too unsophisticated even to undertakeit. But the best endorsements take their power and their efficacy precisely from this: the successful transferof meaning. For example, James Garner's endorsement of Mazda succeeds when a transfertakes place between his persona and the Mazda line. It succeeds when the qualities of maturity, Americanness, confidence, good humor, and a certain kind of maleness are made the qualities of the Mazda vehicle. The endorsement succeeds, in other words, when the properties of the man are made the propertiesof the car.

MEANING TRANSFER: THE GENERAL PROCESS
Celebrityendorsement is, in fact, a special instance of a more general process of meaning transfer.I have described this general process elsewhere in some detail (McCracken 1986, 1988) and review it only briefly here. According to this model, there is a conventional path for the movement of cultural meaning in consumer societies. Meaning begins as something resident in the culturally constituted world (McCracken 1988, pp. 72-73), in the physical and social world constituted by the categories and principles of the prevailing culture. Meaning then moves to consumer goods and finally to the life of the consumer. Several instruments facilitate this transfer. The movement of meanings from the culturally constituted world to consumer goods is accomplished by advertising and the fashion system. The movement of meanings from consumer goods to the individual consumer is accomplished through the efforts of the consumer. Thus does meaning circulate in the consumer society.

314

THE JOURNAL OF CONSUMER RESEARCH

Advertising serves as an instrument of meaning transferin a deceptively simple manner. The transfer process begins when the advertiseridentifies the cultural meaningsintended for the product (i.e., the type of gender, status, age, lifestyle, time, and place meanings). Or, more technically, the advertiserdetermines which of the "categories"and "principles" of culture pertain (McCracken 1988, pp. 73-77). In the language of current advertising practice, the advertiser decides what he or she wishes the product to say. Once this choice has been made, the advertisersurveys the culturally constituted world for the objects, persons, and contexts that already contain and give voice to these mednings. These elements enable the advertiserto bringthe selected cultural meanings into the advertisementin visible, concrete form. However, the advertiser must portray the elements and the product with consummate care and skill. This care is necessary for two reasons. First, elements come chargedwith more meanings than are wanted for the product so the advertiser must evoke some, but not all, of the meaningsof the elements. Second, elements and product must be presentedin such a way that the similarity between them suggests itself irresistibly to and the viewer. This precise combination of eleme&nts product set the stage for the transferof meaning from the product to the consumer. Imprecise or unsophisticated combinations discourageit. Note that there is no necessary or motivated relationship between the meanings and the product. It is not the case that chocolates can be given only certain meanings while tennis racquetscan be given only others. Any product can carry virtually any meaning. Certainly,goods lend themselves to particularmeanings (e.g., chocolates and social sentiment), but advertising is such a powerful mechanism of meaning transfer that virtually any product can be made to take virtuallyany meaning. This propertyof meaning transferis still another reason for taking special care in the selection of certain meanings. The transferprocess must be carefully controlled. Which meanings are chosen for the product will depend on the marketingplan and the sophistication of client, account executive, research group, and creative team. How well meanings are representedin and manipulatedby the advertisementwill depend in particular on the creativedirectorand his or her staff.But the final act of meaning transfer is performed by the consumer, who must glimpse in a moment of recognition an essential similarity between the elements and the product in the ad. The consumer suddenly "sees" that the cultural meanings contained in the people, objects, and contexts of the advertisement are also contained in the product. Well-crafted advertisements enable this essentially metaphoric transference. Badly crafted advertisementsdo not. Once meanings have been moved into goods, they must also be moved into consumers. Consumers

must take possession of these meanings and put them to work in the construction of their notions of the self and the world. They must craft and shape these meanings to fulfill the strategies of meaning manipulation with which they have constructed their lives. Consumers are constantly finding gender, class, age, lifestyle, time, and place meanings in their possessions, and using these meanings to fashion aspects of the self. They are constantly taking possession of cultural principles in consumer goods that help define and fashion the home, the family, and other aspects of the world in which they live. Consumers turn to their goods not only as bundles of utility with which to serve functions and satisfy needs, but also as bundles of meaning with which to fashion who they are and the world in which they live (Belk 1988). When this is done, the movement of meaning is complete. The meaning that began in the culturally constituted world has finally come to rest in the life and experience of the consumer. The cultural circuit is complete. Thus, in general terms, do culture and consumption interact to create a system of meaning movement in contemporary societies. I have given just one account of this process (McCracken 1986, 1988). Readers may wish to consult other accounts of this meaning process (Adams 1973; Holman 1980; Levy 1959, 1981; Mick 1986; Prown 1980; Stern 1988; Wallendorf and Arnould 1988), how it is used (Ames 1982; Appadurai 1986; Belk 1988; Csikszentmihalyi and Rochberg-Halton 1981; Solomon 1983), how it enters into the marketing system (Douglas and Isherwood 1978; Gottdiener 1985; Hirschman and Holbrook 1981), and how it might best be studied (Prown 1982; Sherry 1989; Umiker-Sebeok and Levy 1987).

MEANING TRANSFER: THE CELEBRITY ENDORSER'S CONTRIBUTION
Celebrity endorsement plays a crucial part in the meaning transferprocessjust described.As the Figure shows, the meaning that begins in the dramatic roles of the celebrity comes, in Stage 1, to reside in the celebrities themselves. In Stage 2, this meaning is transferred when the celebrity enters into an advertisement with a product. Some of the meanings of the celebrity are now the meanings of the product. In the final stage, the meaning moves from the product to the consumer. Celebrity endorsement makes a very particularcontribution to each of these three stages.

Stage 1
Endorsement gives the ad access to a special category of person from the culturally constituted world. It makes available individuals charged with detailed and powerful meanings. Celebrities are, in this re-

CELEBRITY ENDORSER: CULTURAL FOUNDATIONS
FIGURE MEANINGMOVEMENTAND THE ENDORSEMENT PROCESS

315

Culture

Endorsement

Consumption

objectso

persons context celebrity

ceert

rdc

product

SIconsumer

role 1 2

Stage 1
Key:
=

Stage 2 path of meaning movement

Stage 3

D] = stage of meaning movement

spect, very different from the anonymous models (or anonymous actors) who are normally used to bring meanings to the ad. Celebrities deliver meanings of extra subtlety, depth, and power. The- contrast between celebrities and models is worth developing. It is clear enough that advertisements can undertake meaning transfer without the aid of celebrities. Anonymous actors and models are chargedwith meaning, and, obviously, they are available at a fraction of the cost. Indeed, for most advertising purposes,the meanings that can be "imported" throughan anonymous model are perfectly sufficient. The question, then, is why celebrities should be used for an ad. How does the celebrity "add value" to the meaning transfer process? What special powers and properties does the celebrity bring to the advertisement, to the product, and, ultimately, the consumer? Anonymous models offer demographic information, such as distinctions of gender, age, and status, but these useful meanings are relatively imprecise and blunt. Celebrities offer all these meanings with special precision. Furthermore, celebrities offer a range of personality and lifestyle meanings that the model cannot provide. Finally, celebrities offer configurationsof meaning that models can never possess. No mere model could bring to Baly-Matrixthe properties that Cher delivers, nor could any model have summoned the impatient, time-tested integrity John Houseman gave the Smith Barney line "We make money the old-fashioned way, we earn it." Only a man playing Houseman's roles in the way Houseman played them could empower the slogan as Houseman did. Celebrities have particular configurations of meanings that cannot be found elsewhere.

In addition, celebrities are more powerful media than anonymous models and actors. Even when they deliver meanings that can be found elsewhere, they deliver them more powerfully. Celebrities evoke the meanings in their persona with greatervividness and clarity. Models and actors are, after all, merely "borrowing" or acting out the meanings they bring to the ad. The celebrity, however, speaks with meanings of long acquaintance. Celebrities"own" their meanings because they have created them on the public stage by dint of intense and repeated performance.Audrey Hepburn delivers "elegance" much more vividly than even the most elegant model. She does so because she has enacted and absorbed this elegance by performingit on stage and screen. Celebritiesdraw these powerful meanings from the roles they assume in their television, movie, military, athletic, and other careers. Indeed, these careers act very much like large ads, as Stage 1 of the Figure shows. Each new dramatic role brings the celebrity into contact with a rangeof objects, persons, and contexts. Out of these objects, persons, and contexts are transferredmeanings that then reside in the celebrity. When the celebrity brings these meanings into an ad, they are, in a sense, merely passing along meanings with which they have been chargedby another meaning transfer process. Or, to put this another way, the meaning that the celebrity endorsement gives to the product was generated in distant movie performances, political campaigns, or athletic achievements. Interestingly, celebrities appear largely unaware of their part in the meaning transfer process. Nowhere is this better illustratedthan in their concern for type-

316

THE JOURNAL OF CONSUMER RESEARCH

casting. Actors say they dislike being cast repeatedly in the same role, claiming that typecastinglimits their career and creative options. What they do not see is that their careers,their art form, and the endorsement process all depend upon typecasting. The North American movie demands the participation of actors chargedwith meanings; it needs actors to bring their own meanings with them to a part. These meanings simplify the movie's expository task and give it substance and direction. Bill Murray,Sylvester Stallone, MorganFairchild, and Joanne Woodward all carry meanings with them from role to role. One of the troubles with unknown actors and actressesis precisely that they are unable to carrymeaning into the role they are asked to play. There are a few exceptions to this pattern. A few actors and actressesare "rinsed" of meaning between roles and, as a result, bring a new persona to each new film. Only an actress of the calibre of Meryl Streep is capable of earning (and exploiting) this privilege. For most, a mild case of typecasting is not just the consequence but actually the very cause of their participation in the Hollywood system. More to the present point, it is precisely this typecasting that makes celebrities so useful to the endorsement process. It is the accumulated meanings of celebrities that make them so potent a source of significance. Meryl Streep has limited value as a celebrity end'orserbecause she is largely free of accumulated meanings. The same may well be true of George C. Scott. Without typecasting, actors are unable to bring clear and unambiguous meanings to the products they endorse. Without typecasting, they have no meanings to give the transferprocess.

Stage 2
Ideally, the choice of particularcelebrities is based on the meanings they epitomize and on a sophisticated marketing plan. In the best of all possible worlds, the marketing/advertisingfirm firstwould determine the symbolic properties sought for the product (having determined which symbolic properties are in fact sought by the consumer). It would then consult a roster of celebrities and the meanings they make available, and, taking into account budget and availability constraints, would choose the celebrity who best representsthe appropriatesymbolic properties. At present, no roster exists, so advertising firms are forced to rely on a very general renderingof what meanings are available to them in the celebrity world and where these meanings are located. Once the celebrity is chosen, an advertising campaign must then identify and deliver these meanings to the product. It must capture all the meanings it wishes to obtain from the celebrity and leave no salient meanings untapped. Furthermore,it must capture only the meanings it wishes to obtain from the

celebrity. All celebrities will encompass in their range of cultural significance some meanings that are not sought for the product. Care must be taken to see that these unwanted meanings are kept out of the evoked set. This will be accomplished by filling the advertisement with people, objects, contexts, and copy that have the same meanings as the celebrity. These elements cue, by the principle of redundancy, the consumer to the salient message. They help select the exact set of meanings that are sought from the celebrity. The ad will sometimes operate on the meanings of the celebrity, and may even modestly help transform them. It is interestingto note, for instance, how novel treatments of Paulina Porizkova's appearandein the current Estee Lauder "White Linen" campaign "repositions" her beauty and redefines its meaning (Wells 1989, p. 72). This campaign very deliberately makes Porizkova'sbeauty more classic, more elegant, and therefore more appropriateto the Estee Lauder "White Linen" product line. In other words, an advertising campaign can sometimes have the effect of a new dramaticrole, bringingthe celebrity into contact with symbolic materials that change the meanings contained in their persona. Celebrities have been known to exploit this effectby choosing their endorsement to tune their image. Typically, however, the ad is not trying to transform the meanings of the celebrity. In most circumstances, it seeks only to transfer them. Finally, the ad must be designed to suggest the essential similarity between the celebrity and the product so that the consumer will be able to take the last step in the meaning transfer process. In a perfect world, copy testing is then used to judge whether indeed the ad succeeds in this regard. When assurance is forthcoming, the second stage of transfer is complete and the ad is put before the consumer. The consumer suddenly "sees" the similarity between the celebrity and the product, and is preparedto accept that the meanings in the celebrity (by dint of long and fond acquaintance) are in the product. If all has gone smoothly, the propertiesof James Garnerare now the propertiesof Mazda.

Stage 3
Let us now consider celebrity endorsement in its final stage of meaning transfer.How does the process of celebrity endorsement help consumers get meanings out of the product into their lives? How, in other words, does an endorsement by James Garner help the properties of the Mazda become the propertiesof the consumer? Consumers are constantly canvassing the object world for goods with useful meanings. They use them to furnish certain aspects of the self and the world. The object world, as we have seen, gives them access

CELEBRITY ENDORSER: CULTURAL FOUNDATIONS

317

to workable ideas of gender, class, age, personality, and lifestyle, in addition to cultural principles of great number and variety. The material world of consumer goods offers a vast inventory of possible selves and thinkable worlds. Consumers are constantly rummaging here. We know that this final stage of the transferprocess is complicated and sometimes difficult. It is not enough for the consumer merely to own an object to take possession of its meanings, or to incorporate these meanings into the self. The meanings of the object do not merely lift off the object and enter into the consumer's concept of self and world. There is, in other words, no automatic transfer of meaning nor any automatic transformation of the self. The consumer must claim the meanings and then work with them. We have some general sense that rituals play an important part in this process. Consumers must claim, exchange, care for, and use the consumer good to appropriate its meanings (Cheal 1988; McCracken 1988; Rook 1985). We know that they must select and combine these meanings in a process of experimentation (Belk 1988; Wallendorf and Arnould 1988). But this process is still very much terra incognito from a scholarly point of view. Of all of the topics in the culture and consumer behavior portfolio, this one is the most neglected. A cultural understandingof celebrity endorsement illuminates this little known terrain. Celebritiesplay a role in the final stage of meaning transferbecause they have createdthe self. They have done so publicly, in the first stage of the meaning transfer process, out of bits and pieces of each role in their careers. All the world has watched them take shape. From darkened theaters, consumers have looked on as celebrities have selected and combined the meanings contained in the objects, people, and events around them. The self so created is almost always attractive and accomplished. Celebrities build selves well. The constructed self makes the celebrity a kind of exemplary, inspirational figure to the consumer. Consumers are themselves constantly moving symbolic propertiesout of consumer goods into their lives to construct aspects of self and world. Not surprisingly, they admire individuals who have accomplished this task and accomplished it well. Celebrities are proof that the process works. Celebrities have been where the consumer is going. They have done in Stage 1 what the consumer is now laboring to do in Stage 3 of the meaning transferprocess. Or, to put this another way, consumers are all laboring to perform their own Stage 1 construction of the self out of the meanings supplied by previous and present roles and the meanings accessible to them there. But this is more than just a formal parallel between celebritiesand consumers in Stages 1 and 3. The consumer does not revere the celebrity merely because

the celebrity has done what the consumer wants to do, but also because the celebrity actually supplies certain meanings to the consumer. Celebrities create a self out of the elements put at their disposal in dramatical roles, fashioning cultural meanings into a practicable form. When they enter the endorsement process, they make these meanings available in material form to the consumer. Consumersare gratefulfor these meanings and keen to build a self from them. The celebrity is supplying not just an example of selfcreation, but the very stuffwith which this difficultact is undertaken. Let us return to the James Garner example once more. Consumers have watched James Garner fashion what Schudson calls the "fictive self" out of the objects, events, and contexts of his screen life. Garner has given them a dramatic example of the very act in which consumers are themselves engaged. Furthermore, Garnerhas put useful and interestingmeanings at their disposal. He has given them a vivid, well-organized and "performable"self. This film persona (and its successor in the television world, Thomas Magnum) offers a self that is capable but occasionally incompetent, forthrightbut unassuming, and almost always the master of his fate (except when conspired against by a comically or otherwise imperfectly malevolent force). The Garner self is diverse, balanced, and, most of all, workable. But there is a second, in some ways more interesting, way in which celebrities play the role of a "super consumer." This occurs when the film persona of the celebrity consists not merely in the presentation of an interesting film persona but actually in the creation of a self that is new and innovative. Most film stars bring to the screen a self, cut whole cloth, from the standard American personality inventory. But there are a few who have undertakena much more difficult and creative innovation in which personality elements are created or dramaticallyreconfigured. In this highly creative mode, the celebrity becomes a kind of experiment in self-construction. This makes the celebrity very powerful indeed. He or she has become an inventor of a new self the consumer can use. A good example of such an act of self-invention is the character portrayed by Bruce Willis in the series, "Moonlighting." Willis invented a version of maleness, a way of interacting with others (not the least of whom was a female superordinate),and a posture toward the world that holds enormous appeal for certain consumers. In his creation of David Addison, Willis put useful meanings at the disposal of the consumer. In a sense, he product-testednotions of the self for a group of consumers who are themselves engaged in an act of experimentation. Consumers perceive that Willis is working on a self and is within "spitting distance" (as his screen persona might say) of accomplishing this self. Consumersengagedin a similar process are grateful for both the example and the point-

318

THE JOURNAL OF CONSUMER RESEARCH

ers. It is preciselythis exemplarystatus, and the novel meanings contained in Willis's persona, that make him a compelling endorserfor Seagram'swine coolers (Gabor, Thorton, and Wiener 1987). The celebrityworld is, to this extent, a realm of experimentation in which actors sometimes do more than merely play out cultural categories and principles. Sometimes they also engage in innovation, as when Bea Arthur creates new notions of the elderly, the stars of "Family Ties" work out certain notions of the family, and rock and roll stars invent and reinvent the adolescent self. This experimentation makes the celebrity an especially potent source of meaning for the marketing system and a guide to the process of self-invention in which all consumers are engaged. Celebritiesserve the final stage of meaning transfer because they are "super consumers" of a kind. They are exemplary figures because they are seen to have created the clear, coherent, and powerful selves that everyone seeks. They are compelling partners to the meaning transfer process because they demonstrate so vividly the process by which these meanings can be assembled and some of the novel shapes into which they can be assembled. But who really needs the meanings created by celebrities?We know that there are certain groups especially keen on using them. Solomon (1983) observed that anyone undergoingany sort of role change or status mobility is especially dependent on the meanings of their possessions. McCracken(1987b) has tried to show the importance of this meaning to those who are moving from one age categoryto another. O'Guinn et al. (1985) have pointed out that those who are newly arrived to a culture are also heavily indebted to the meanings contained in the consumer society and the celebrity world. But it has also been asserted that everyone in a modern, developed society is underspecified in this sense. As Belk (1984) and Sahlins (1976) have argued, modern Western selves are deliberately left blank so that the individual may exercise the right of choice. Also pertinent is the relative collapse of institutions that once supplied the self with meaning and definition (e.g., the family, the church, the community). Workingtogether, individualism and alienation have conspired to give individuals new freedom to define matters of gender, class, age, personality, and lifestyle. The freedomto choose is now also an obligation to decide, and this makes us especially eagerconsumers of the symbolic meanings contained in celebrities and the goods they endorse. This, in broad detail, suggests how celebrity endorsement operates as a process of meaning transfer. We have reviewedeach of the three stages of this process, consideringin turn how meaning moves into the persona of the celebrity, how it then moves from the celebrity into the product, and, finally, how it moves from the product into the consumer. Celebrities are,

by this account, key players in the meanings transfer process.

THE REAL CONSUMER SOCIETY
This discussion of the cultural aspects of celebrity endorsement enables us to address one or two larger issues in the field of consumer research. More precisely, it carrieslargerimplications for the debate that now rages about the nature of the consumer society. One party to this debate arguesthat the consumer society encourages low artistic standards, materialistic preoccupations, and an affection for the trivial and unimportant (e.g., Barnouw 1978; Ewen 1976; Lasch 1979; Pollay 1986). The consumer society has been declared the domain of the Philistine. The North American preoccupation with movie stars is a favorite target (Ewen 1988, pp. 92-100). North America is accused of having a trivially minded fascination with the affairsof the rich and famous. This fascination is cited as evidence of the depths to which popular culture is destined to fall, the bankruptcyof North American life, and the shallowness of the individual who lives therein. These arguments are, no doubt, very satisfying from a political and polemical point of view, but they do not reckon very well with the cultural realities of the celebrityworld as describedhere. These criticisms fail to see that Hollywood, the star system, and'celebrity endorsement are all profoundly cultural enterprises and that our fascination with celebrities reflects our involvement in the meaning transfersystem they accomplish. The celebrity world is one of the most potent sources of cultural meaning at the disposal of the marketing system and the individual consumer. It is, therefore, not at all surprising that we should care about celebrities and the lives they lead. It would be much more surprising if we were indifferent and somehow above an interest in the world of the stars. As all of us labor to fashion manageable selves, it is inevitable that we should cultivate a knowledge of this world. More plainly, North Americans are not "star-crazy,"but rather merely active consumers of the meanings that are made available by the celebrity world. There is indeed a delicate and thoroughgoing relationship between the culture, the entertainment industry, and the marketing system in modern North America. We are beginning to understand what this relationship is and how it works. We must hope that the first victims of this emerging understanding will be the glib assertions that characterizeNorth American society as thoughtlessly materialistic, and North American consumers as the narcissistic, simpleminded, manipulated playthings of the marketplace. Celebrity endorsement and the marketingsystem are cultural undertakingsin which meaning is constantly

CELEBRITY ENDORSER: CULTURAL FOUNDATIONS

319

in circulation. As we begin to render a more sophisticated account of how these systems work, we will begin to see that North American culture and commerce are more interesting and more sophisticated than its critics have guessed.

FUTURE RESEARCH
The cultural perspective suggests three avenues of research.The first of these is a thorough assessment of how meanings move in the celebrity world. We know that each role, event, or accomplishment in the careerof the celebrity changes the meanings of the celebrity, but we do not know precisely how this takes place. We do not know what the precise relationship is between the role, event, and accomplishment, on the one hand, and the celebrity, on the other. Nor do we know how meaning transfers from one to the other. We need a precise idea of how meanings come to exist in celebrities. The first avenue of research has a methodological component as well. We need an instrument that allows us to determine methodologically the meanings that adherein celebrities. We know that the meanings that exist in celebrities are extraordinarilynumerous and various, but we have yet to devise an instrument that allows us to detect and survey these meanings. Only some delicate combination of qualitative and quantitative methods will enable us to decipher the meanings of the celebrity world in the individual and the aggregate. Once an instrument is devised, certain crucial empirical work can be undertaken. We can determine the meanings that any individual celebrity brings to the endorsement process, and to survey the meanings that exist in the entire world of endorsement. A typography of the meanings contained throughout the celebrity worlds of sports, politics, business, art, the military, television, and Hollywood is possible and will give us a systematic sense of the meanings at the disposal of the endorsement system and the meaning transferprocess. The second avenue of research should concentrate on a more precise determination of how advertising accomplishes the transfer of meaning from celebrity to product. How do creative directors identify and catalog the symbolic properties contained in the celebrity world? What are the rhetorical and visual devices by which this celebrity meaning is transferred within the advertisement? What celebrity meanings are seen to work best with what products?What is the process by which consumers contribute to the meaning transferprocess?It is also relevantto ask here how a celebrity changes his or her stock of symbolic properties by participatingin an advertisement. The third avenue of research deals with how consumers appropriateand use the meanings that come to them as a result of endorsement. How does the con-

sumer take possession of this meaning?How does the consumer use the meaning of the celebrity in the construction of self and world?It may be that some consumers routinely canvass the symbolic meanings of one or several celebrities to take advantage of the experimentation taking place here. Do consumers set up long-term relationshipswith a single celebrity and systematically "download" all the new meanings this celebrity makes available through new roles and endorsements?Do consumers follow a variety of celebrities from whom they draw a variety of meanings? What happens to consumers when celebrities are transformedby disgraceor new fame?We must begin to chart what becomes of the cultural meanings after they leave the endorsement and enter into the life of the consumer. These are all questions that need to be answeredfor us to understandthe process of celebrity endorsement in fine detail. They are the researchopportunities the meaning transferperspectivebrings to light. The meaning transfer approach casts some doubt on the sufficiency of the source models' explanation of celebrity endorsement. But it does not prevent us from asking the questions that have been asked in this tradition. For instance, it is still possible to talk about the issue of credibility or other questions relevant to source research.It is still possible to see that some celebrities are more credible than others and that each celebrity is more credible for some promotional purpose than others. What the meaning transfer model does is shift the terms of this debate. When we consider credibility in this new context, we are no longer talking about the manner in which celebrities communicate information, but ratherthe manner in which they communicate meaning (McCracken 1987a). Credibility now turns on which meanings celebrities make available to endorsements and how well they transfer these meanings to the product. Examples of this cultural credibility are not hard to find. John Houseman was the compelling choice for the Smith Barneyadvertisement. The actor chosen to succeed him, Leo McKern, carries different meanings in a different configuration. He is, in a word, less credible. The Smith Barney slogan is changed, and diminished, as a result. The symbolic or cultural perspective (McCracken 1987a) allows for a new credibility measure of a different sort. There is, for instance, no longer any single kind of credibility. A celebritycan be extremely credible for certain meanings and not at all credible for others. Plainly, this aspect of credibility cannot be capturedby the theories and instrumentsconventionally used. Indeed, to investigate the celebrity endorsement from the symbolic or cultural point of view, a new set of questions and methodologies must be investigated.

CONCLUSION
A new perspective on the process of celebrity endorsement has been developed. It has suggested that

320

THE JOURNAL OF CONSUMER RESEARCH

the source models with which endorsement is now understood by practitioner and scholar are insufficient. The chief deficit of these models is that they ask us to accept that it is the attractivenessand credibility of the celebrity that make the endorsement work. Useful for certain purposes,this approachpreventsus from seeing that celebrities are in fact highly individualized and complex bundles of cultural meaning. It also prevents us from seeing that endorsement consists in the transferof these meanings from the celebrity to the product, and from the product to the consumer. The meaning transfermodel presented here is intended to demonstrate that the secret of the celebrity endorsement is largely cultural in nature, and that the study of the celebrity and endorsement is improved by a cultural perspective. [ReceivedNovember1988. Revised April 1989.]

REFERENCES
Adams, Marie Jeanne (1973), "StructuralAspects of a Village Art," American Anthropologist, 75 (February), 265-279. Ames, Kenneth L. (1982), "Meaningin Artifacts:Hall Furnishings in Victorian America," in Material Culture Studies in America, ed. Thomas J. Schlereth, Nashville, TN: The American Association for State and Local History, 206-221. Appadurai, Arjun, ed. (1986), The Social Life of Things, Cambridge,England:CambridgeUniversity Press. Atkin, Charlesand Martin Block (1983), "Effectivenessof CelebrityEndorsers,"JournalofAdvertisingResearch, 57-61. 23 (February/March), Baker, Michael J. and Gilbert A. Churchill (1977), "The Impact of PhysicallyAttractiveModels on Advertising Evaluations,"Journal ofMarketing Research, 14 (November), 538-555. Barnouw,Erik(1978), TheSponsor.Notes on a ModernPotentate,New York: OxfordUniversity Press. Belk, Russell W. (1984), "Cultural and Historical Difference in Concepts of the Self and Their Effectson Attitudes Toward Having and Giving," in Advances in ConsumerResearch, Vol. 11, ed. Thomas C. Kinnear, Provo, UT: Association for Consumer Research, 753760. (1988), "Possessions and the Extended Self," Journal of ConsumerResearch, 15 (September), 139-168. Cheal, David (1988), "The Post Modern Origin of Ritual," Journalfor the Theoryof Social Behavior, 18 (September), 269-290. Csikszentmihalyi, Mihaly- and Eugene Rochberg-Halton (1981), TheMeaning of Things,New York: Cambridge University Press. Debevec, Kathleen and Jerome B. Kernan (1984), "More Evidence on the Effects of a Presenter's Physical Attractiveness:Some Cognitive, Affectiveand Behavioral Consequences," in Advances in Consumer Research, Vol. 11, ed. Thomas C. Kinnear, Provo, UT: Association for ConsumerResearch, 127-132. Dholakia, Ruby Roy and Brian Sternthal (1977), "Highly Credible Sources: Persuasive Facilitators or Persua-

sive Liabilities?" Journal of Consumer Research, 3 (March),223-232. Douglas, Mary and Baron Isherwood (1978), The Worldof Goods: Towards an Anthropology of Consumption, New York: W.W. Norton. Ewen, Stuart(1976), Captainsof Consciousness,New York: McGraw-Hill. (1988), All Consuming Images, New York: Basic Books. Finn, David W. (1980), "The Validity of Using Consumer Input to Choose AdvertisingSpokesmen,"in Advances in ConsumerResearch, Vol. 7, ed. JerryC. Olson, Ann Arbor, MI: Association for Consumer Research, 776779. Friedman, Hershey H. and Linda Friedman (1979), "Endorser Effectivenessby Product Type," Journal ofAdvertisingResearch, 19 (October),63-71. , Michael J. Santeramo,and Anthony Traina(1978), "Correlatesof TrustworthinessFor Celebrities,"Journal of theAcademy ofMarketingScience, 6 (Fall), 291299. , SalvatoreTermini, and Robert Washington(1976), "The Effectiveness of Advertisements Utilizing Four Types of Endorsers,"Journal of Advertising, 5 (Summer), 22-24. Gabor, Andrea, Jeannye Thorton, and Daniel P. Wiener (1987), "Star turns that can turn star-crossed," U.S. News and WorldReport,(December 7), 57. Gottdiener, Mark(1985), "Hegemony and Mass Culture:A Semiotic Approach,"American Journal of Sociology, 90 (5), 979-1001. Hirschman,Elizabethand MorrisB. Holbrook, eds. (1981), Symbolic ConsumerBehavior,Ann Arbor,MI: Association for ConsumerResearch. Holman, Rebecca (1980), "Product Use as Communication: A Fresh Appraisalof a VenerableTopic," in Review of Marketing, eds. Ben M. Enis and Kenneth J. Roering, Chicago: American Marketing Association, 250-272. Hovland, Carl I. and Walter Weiss (1951-1952), "The Influence of Source Credibility on Communication Effectiveness,"Public Opinion Quarterly,15 (Winter), 635-650. , IrvingL. Janis, and Harold H. Kelley (1953), Communication and Persuasion, New Haven, CT: Yale University Press. Joseph, Benoy W. (1982), "The Credibilityof PhysicallyAttractiveCommunicators:A Review," JournalofAdvertising, 11 (July), 15-24. Kahle, Lynn R. and Pamela M. Homer (1985), "Physical Attractivenessof the CelebrityEndorser:A Social Adaptation Perspective,"Journal of ConsumerResearch, 11 (March),954-961. Kaikati, Jack G. (1987), "CelebrityAdvertising:A Review and Synthesis,"InternationalJournal ofAdvertising,6 (2), 93-105. Kamen, Joseph M., Abdul C. Azhari, and Judith R. Kragh (1975), "Whata SpokesmanDoes for a Sponsor,"Journal ofAdvertisingResearch, 15 (April), 17-24. Klebba, Joanne M. and Lynette S. Unger (1983), "The Impact of Negative and Positive Information on Source Credibility in a Field Setting," in Advances in Consumer Research, Vol. 10, eds. Richard P. Bagozzi and

CELEBRITY ENDORSER: CULTURAL FOUNDATIONS

321

Alice M. Tybout, Ann Arbor,MI:Association for Consumer Research, 11- 16. Kuhn, Thomas S. (1962), The Structureof Scientific Revolutions, Chicago:University of ChicagoPress. Lasch, Christopher(1979), The Cultureof Narcissism, New York:W.W. Norton. Levy, Sidney J. (1959), "Symbols for Sale," HarvardBusiness Review, 37 (July/August), 117-124. (1981), "Interpreting Consumer Mythology: A StructuralApproachto Consumer Behavior," Journal of Marketing,45 (Summer), 49-6 1. Marshall,Christy(1987), "It Seemed Like A Good Idea at the Time," Forbes, (December 28), 98-99. McCracken,Grant (1986), "Culture and Consumption: A Theoretical Account of the Structure and Movement of the CulturalMeaningof ConsumerGoods," Journal of ConsumerResearch, 13 (June), 71-84. (1987a), "Advertising: Meaningor Information?"in Advancesin ConsumerResearch,Vol. 14, eds. Paul Anderson and Melanie Wallendorf, Provo, UT: Association for ConsumerResearch, 121-124. (1987b), "Cultureand Consumption Among the Elderly: Three Research Objectives in an Emerging Field," Ageing and Society, 7 (June), 203-224. (1988), Cultureand Consumption,Bloomington: Indiana University Press. McGuire, William J. (1985), "Attitudes and Attitude Change," in Handbook of Social Psychology, Vol. 2, eds. GardnerLindzey and Elliot Aronson, New York: Random House, 233-346. Mick, David Glen (1986), "ConsumerResearchand Semiotics: Exploringthe Morphologyof Signs, Symbols, and Significance,"Journal of ConsumerResearch, 13 (September), 196-213. Motavalli, John (1988), "AdvertisingBlunders of the Rich and Famous,"Adweek,(January 11), 18-19. O'Guinn, Thomas C., Ronald J. Faber, and MarshallRice (1985), "PopularFilm and Television as ConsumerAcculturation Agents:America 1900 to Present," in Historical Perspective in Consumer Research: National and International Perspectives, eds. Chin Tiong Tan

and JagdishN. Sheth, Singapore:National University of Singapore,297-30 1. Pollay, Richard(1986), "The Distorted Mirror:Reflections on the Unintended Consequences of Advertising," Journal of Marketing,50 (April), 18-36. Prown, Jules D. (1980), "Style as Evidence," Winterthur Portfolio, 15 (Autumn), 197-210. (1982), "Mind In Matter:An Introductionto Material Culture Theory and Method," Winterthur Portfolio, 17 (Spring), 1-19. Rook, Dennis W. (1985), "The Ritual Dimension of Consumer Behavior," Journal of Consumer Research, 12 (December), 251-264. Sahlins, Marshall (1976), Culture and Practical Reason, N Chicago:University of Chicago Press. Schudson, Michael (1984), Advertising, The Uneasy Persuasion, New York: Basic Books. Sherry,John F., Jr. (1989), "PostmodernAlternatives:The Interpretive Turn in Consumer Research," in Handbook of Consumer Theory and Research, eds. Harold H. Kassarjian and Thomas Robertson, Englewood Cliffs:Prentice-Hall,in press. Solomon, Michael R. (1983), "The Role of Productsas Social Stimuli: A Symbolic Interactionism Perspective," Journal of ConsumerResearch, 10 (December), 319329. Stern,Barbara (1988), "MedievalAllegory:Roots of Advertising Strategyfor the Mass Market,"Journal of Marketing, 52 (July), 84-94. Sternthal, Brian, Ruby Roy Dholakia, and Clark Leavitt (1978), "The Persuasive Effect of Source Credibility: Tests of Cognitive Response," Journal of Consumer Research, 4 (March),252-260. Umiker-Sebeok, Jean, ed. (1988), Marketing and Semiotics, Bloomington: Indiana University Press. Wallendorf, Melanie and Eric Arnould (1988), "'My Favorite Things': A Cross-CulturalInquiry into Object Attachment, Possessiveness, and Social Linkage," Journal of ConsumerResearch, 14 (March),531-547. Wells, Linda (1989), "Face Value," New YorkTimesMagazine, Part 2 (February26), 46, 56, and 72.