Community and Canon: a Foundation For Mature Interpretive Research

Val Larsen, Truman State University
Newell D. Wright, James Madison University
ABSTRACT - This article argues that though its achievements are obvious, interpretive research is built upon an unstable epistemological foundation, the "method of authority." It suggests that interpretive research can be fully credible only if it is grounded in the evolving consensus of an informed interpretive community. And this informed consensus can develop only if an essential element of research infrastructure is establishedCa canon of texts that can be interpreted and reinterpreted by the research community. The article models the kind of reinterpretation it urges by taking a second look at Mr. Ed’s Elephant Museum and Roadside Attraction.
[ to cite ]:
Val Larsen and Newell D. Wright (1997) ,"Community and Canon: a Foundation For Mature Interpretive Research", in NA - Advances in Consumer Research Volume 24, eds. Merrie Brucks and Deborah J. MacInnis, Provo, UT : Association for Consumer Research, Pages: 310-314.

Advances in Consumer Research Volume 24, 1997      Pages 310-314

COMMUNITY AND CANON: A FOUNDATION FOR MATURE INTERPRETIVE RESEARCH

Val Larsen, Truman State University

Newell D. Wright, James Madison University

ABSTRACT -

This article argues that though its achievements are obvious, interpretive research is built upon an unstable epistemological foundation, the "method of authority." It suggests that interpretive research can be fully credible only if it is grounded in the evolving consensus of an informed interpretive community. And this informed consensus can develop only if an essential element of research infrastructure is establishedCa canon of texts that can be interpreted and reinterpreted by the research community. The article models the kind of reinterpretation it urges by taking a second look at Mr. Ed’s Elephant Museum and Roadside Attraction.

In recent years, interpretive research has so enriched our understanding of consumer behavior that, in our view, the value of thick descriptions of consumer behavior is firmly established. But its achievements notwithstanding, interpretive research is built, we believe, upon an unstable epistemological foundationCwhat C.S. Pierce (1934) has called the "method of authority." In this paper we argue that interpretive research can be fully credible only if it is grounded in the evolving consensus of an informed interpretive community. And this informed consensus can develop only if an essential element of research infrastructure is establishedCa canon of texts that can be interpreted and reinterpreted by the research community.

CRITIQUE OF CURRENT FOUNDATIONS OF INTERPRETIVIS

The major weaknesses of the current approach to interpretive research are apparent in the pioneering articles that helped make naturalistic inquiry (Lincoln and Guba 1985) possible in consumer research, e.g., Hirschman (1986), Wallendorf and Belk (1989). We focus our critique on Wallendorf and Belk’s "Assessing Trustworthiness in Naturalistic Consumer Research," but what we say generally applies to other founding documents as well. The problems begin with the central metaphor of these papers: the researcher as instrument. If we take this metaphor seriously, then the credibility of each interpretivist study hangs entirely upon the trustworthiness of the researcher. If the researcher-instrument is not a reliable and valid measure, then the study has little value. The authority of the researcher thus becomes the paramount epistemological problem of interpretive researchCin our view, an insoluble problem.

Building upon Lincoln and Guba (1985), Wallendorf and Belk (1989) attempt to address this problem by prescribing a methodological fixCa set of procedures that may help establish the trustworthiness of the researcher. Some of these procedures are designed to ensure that the researcher adequately immerses him or herself in the data. These include injunctions to triangulate across informants, to analyze negative cases, and to engage informants over a prolonged period of time. Other proceduresCinternal team audits, external peer audits, and member checksCare designed to intersubjectively certify the trustworthiness of the researcher. These procedures permit the researcher to say, in effect, "you should trust my interpretation because other team members, an external peer, and my informants say that it is trustworthy." In other words, the authority of a small group supplants that of the individual researcher. The epistemological foundation remains the same, however: Pierce’s dubious "method of authority."

Wallendorf and Belk are too sophisticated and experienced to suppose that procedures such as the ones they prescribe can ensure the overall quality of interpretive research. Apparently somewhat uncomfortable with their methodological solution, they state at the beginning of their paper that no set of prescriptions can guarantee novel and interesting insights. Indeed, they say that "a list of 'How-to’ steps is likely to negate the spontaneity and serendipity that guide good research. It is our personal experience that such ideas are much more likely to emerge from playfulness and openness than from mechanistic procedures" (p. 69). And yet, after conceding, finally, that "merely following the procedures we outline here would likely produce a rather boring output," they devote the bulk of their paper to developing the list of how-to steps that would produce that boring output, steps that have since become, in some measure, standard practice.

Why did this happen? Why did Wallendorf and Belk focus on the relatively unimportant how-to list rather than on the openness and play that they consider more important? For two reasons, we believe. First, they were boxed in by the unfortunate researcher-instrument metaphor. If the researcher must mediate our encounter with the data, then the issue of researcher trustworthiness (reliability) must be addressed. Second, and more importantly, they were compelled by the rhetorical context at the timeCby the dominance of positivism and the skepticism of positivistic researchersCto develop a methodology that might warrant interpretive findings as the experimental method warrants positivistic findings. The new interpretivist paradigm had to be sold in a language and mode of thought that had positivistic analogsCand that meant an emphasis on method (Holt 1991). If Hirschman (1986), Wallendorf and Belk (1989), and others had not been persuasive, if subsequent interpretivist research had not proven to be deeply insightful, there would have beenno occasion to write this paper.

Let us restate, then, our assessment of our current problem. The researcher-instrument metaphor was probably a rhetorical necessity early on. But it makes the validity of interpretivist research depend upon the personal authority of the researcher. It thus grounds interpretivist research in the dubious method of authority. (Holt [1991] has persuasively argued that the methods Wallendorf and Belk propose do nothing to change this.) It also builds upon a category error, for a researcher is almost infinitely more complex than a thermometer, a Likert scale, or even a particle accelerator. Unlike, say, a tape recorder, a researcher can never be a trustworthy instrument that reliably mediates between a complex data set and other researchers. So if we take this metaphor seriously, the validity of interpretive research will always be in question, and interpretive research will always be perceived as being second rate (Holt 1991). Fortunately, interpretive research has been accepted in some measure by the positivistically inclined ACR/JCR community, so it may now be possible to move beyond the initial spade work and establish this research tradition upon its proper foundation.

PROPOSED NEW FOUNDATIONS FOR INTERPRETIVIS

Our thesis, once again, is that interpretivist research can be fully credible only if it is grounded in the evolving consensus of an interpretive community, an evolving consensus that is possible only if the discipline has a canon. Thus, in our view two elements, among others, are essential in a mature interpretivism, and one is notably neither essential nor desirable.

The first essential element is an informed and critical interpretive community (Fish 1980; Wittgenstein 1969), a group of researchers who are acquainted with the data and who share a sufficient number of assumptions that they can agree on which interpretations are valid, which invalid. In a proper interpretive community of this kind, as group members evaluate and criticize each other’s interpretations, an evolving consensus emerges that does not rest upon the authority of any one or two members, for all group members read the texts that are being interpreted and judge for themselves whether a proposed interpretation is valid. Each intersubjectively certifies or decertifies a proposed reading. Consequently, if a finding is broadly judged to be valid, it is certified not by one or two or several members of the discipline but rather by the discipline at large. We expand on this point below.

The second essential element is a canon, a body of publicly available texts (interviews, ads, pictures) that the discipline regards as being a suitable object of study. This infrastructural element is essential because it makes possible unmediated access to the data that are being interpreted. If researchers are to judge for themselves how well an interpretation fits the data it purports to explain, they must have access to and be familiar with the data. A canon makes this possible and, thus, makes it possible to dispense with the method of authority. We discuss below what we mean by a canon.

The element that is not essential or even desirable in a mature interpretivism is an emphasis on method (Holt 1991). In a mature interpretive disciplineCthe humanities, for instanceCno one knows or cares how researchers arrive at a given interpretation. The only thing that matters is how well the interpretation fits with and/or accounts for the data that are being interpreted (Thompson 1990). Thus, in humanities courses, students are generally urged to avoid intellectual autobiographyCaccounts of their discovery process (i.e., method sections)Cand simply cut to the chase, the patterns they have discovered in the text. Their interpretation is evaluated on how well it fits the data, not on the procedures or thought processes they went through to produce it. Commenting on this point in Truth and Method (1975), a book that woud have been more aptly entitled Truth or Method, Gadamer argues that method is the enemy of truth, that interpretive truth emerges from openness and play, not from method. In saying this, he basically affirms Wallendorf and Belk’s (1989) pregnant aside on research creativity while calling into question their extended methodological bow to positivism. In his view and ours, method is not an appropriate foundation for validity in interpretive research.

Critique and Consensus within an Interpretive Community

In this section we discuss somewhat more extensively why mature interpretive research requires informed criticism within an interpretive community and, thus, requires multiple interpretations of a text. Unlike positivism, which depends upon multiplicative corroboration, interpretivism depends upon structural corroboration (Pepper 1970), complex judgments about how fact agrees with fact. Since the object of study in interpretive research is usually a multifaceted text, the number of potentially interlocking facts embedded in the text is generally very large. An adequate interpretation is one that accounts for and interrelates all elements in the text that are relevant to a given interpretation. Thus, "interpretation is inescapably descriptive.... If our interpretation is persuasive, it is so because our description enables others to see the aspects [of the text that] we are pointing out" and that they otherwise would not have seen (Dasenbrock 1986, p. 1035).

The potentially infinite complexity of a text notwithstanding, interpretivist researchers are often certain, both as individuals and as a group, that they have grasped the essential meaning or importance of a text. Building upon insights of Wittgenstein, Dasenbrock (1986) has persuasively argued that researchers attain and progressively deepen their certitude because shared assumptions create a conceptual space within which the range of possible meanings is limited. Within those conceptual limits, the aspect density, the descriptive power of various interpretations of the text can be compared.

For our purposes, the crucial point is that detail-oriented criticism from colleagues is the engine that leads to deeper levels of individual and group confidence in an interpretation. When presenting at a conference an interpretation of a publicly available text, "the speaker ... is always aware (or else is soon made aware) that his paper has slighted some aspects of the work under discussion." The speaker and others develop new levels of insight and depths of certitude in their reading by dealing with these objections. "Most of us" Dasenbrock (1986, p. 1033) says "emerge most of the time from such a discussion or from reading criticism about a work with a more complex perspective on it than we had before [and] an impression of coming to a deeper understanding of the work." The proper warrant of the value of interpretive research is this deep and shared confidence that we have grasped the meaning of the text, confidence grounded in a thorough engagement with textual details that is driven by the close critical scrutiny of other informed readers.

Dasenbrock tells us how certitude about and consensus on the meaning of a text are achieved in fully mature interpretive disciplines such as literary studies. Is similar certitude and consensus possible for us in consumer research? Not at present, because we lack the infrastructure that is necessary if our interpretations are to be intersubjectively ratified by and deepened through the criticism of other researchers. At our conferences, the person making a presentation is often the only one in the room who has read the interviews that are being interpreted. This researcher often provides proof-text quotations that support his or her interpretation. But no one else is in a position to challenge and deepen the interpretation by citing other passages that call the interpretation into question or otherwise compel a more careful and nuanced reading. We will now discuss the infrastructure that would make this kin of critique and deeper insight possible.

Canons and Canon Formation

Let us return to what we earlier said was the second essential element of a mature interpretivism, a canon, for it is a canon that can make possible unmediated access to data and the informed criticism that leads to validated interpretive research findings. For our purposes, the term canon has two relevant meanings. Defined broadly, a canon is the complete body of publicly available texts that a discipline regards as appropriate objects of study (Fowler 1982). In this broad sense, the literary canon includes both the classic novels of Jane Austen and the latest novels of obscure science fiction writers. This sense of the term is logically prior but theoretically less interesting than a second meaning. Defined more narrowly, a canon is the subset of texts in this larger collection that have been judged by a discipline to be of special interest and worth. Texts enter this more restricted canon on the nomination of researchers whose interpretations reveal in them important dimensions of meaning. As the interests of researchers change over time, the membership of this more restricted canon changes as well, though certain particularly rich texts tend to be retained (Smith 1984). In this narrower sense, the literary canon includes Jane Austen’s work but not the latest science fiction potboilers.

At present, consumer research has no canon in either the broad or the narrow sense of the term. We lack a canon in the broad sense not because there are no texts that are appropriate objects of study for consumer researchers. Over the past decade, researchers have collected thousands of pages of interviews with consumers on a wide range of important topics. We lack a canon because we have not established conventions that encourage and infrastructure that facilitates public availability of the important texts that already exist. Thus, if we want to become a mature interpretive discipline and put the method of authority behind us, our lack of data sharing conventions must be our first concern. And for the moment, resistance to data sharing seems to be firmly entrenched. Over the past year, we have contacted several researchers and asked for access to interviews that were the basis of JCR articles. Though the researchers we contacted are as genial and generous as the average consumer behaviorist, they denied access to the interviews. The conventions of our discipline do not encourage them to share their data. So for various reasons, none of them compelling if our conventions should change, they didn’t share them. And yet, data sharing has already proven beneficial to the discipline (Stern 1995).

If we can change our data sharing conventions, our second concern would be to establish the infrastructure that would make canonical texts publicly available. This should be relatively easy, given the growth of the World Wide Web, on which interviews, photographs, video and audio recordings can be stored and accessed electronically at very little cost. The Association for Consumer Research might be the organization best positioned to provide and maintain a web site that archives various canonical texts.

If changes in conventions and infrastructure led to the creation of a canon in the broader sense, a canon in the more important narrow sense would soon follow as certain texts were identified as being especially meaningful and important. The groundwork for the emergence of this subset has already been done. Previously published articles based on interview data implicitly nominate their source interviews for inclusion in this more selective consumer behavior canon (Frye 1971). If some of these interviews were selected for a second interpretation and then still other reinterpretations, they would eventually become standard texts, widely read and deeply understood on account of the various articles devoted to explicating them.

Virtual Canon in Consumer Research

While consumer research has no proper canon, published interpretive research has been grounded to varying degrees in a kind of virtual canon, for it has focused to a greater or lesser degree on phenomena that are familiar to other consumer researchers. Thus, various studies can be plotted on a continuum that reflects the degree to which the object of study is, in this limited sense, canonical. At the canonical end of this continuum, one might place Holbrook and Grayson’s (1986) article on the semiology of cinematic consumption. This study interprets a text, Out of Africa, that is delimited, tangible, and readily available to other researchers who may wish to assess the quality of the interpretation. Only slightly less available are texts Hirschman (1990) cites in her article on secular immortality: Connoisseur and Avenue, magazines targeted at the wealthy; Trump: The Art of the Deal, an autobiography; and Tom Wolfe’s novel The Bonfire of the Vanities. Also at the canonical end of the continuum is Belk, Wallendorf, and Sherry’s (1989) paper on sacred and profane consumption. This study also treats textsCinterviews, photos, videos and audio recordingsCthat are available, though not readily available, to other researchers. In his article on celebrity endorsers, McCracken (1989) explains why character actors may be more effective endorsers than actors who play a wider range of roles. This study interprets ads that are not available except, perhaps, in memory, but virtually all researchers can assess the validity of McCracken’s claims based on their daily experience of other ads, so in this respect, the analysis is rooted in a canon. Further toward the noncanonical end of the continuum would be Rook’s (1987) "The Buying Impulse." Other researchers may not know any compulsive consumers, and they have no access to the interviews on which this article was based. (Proof texts included in the article naturally don’t facilitate critique.) One might anchor the non-canonical end of the continuum with Hill and Stamey’s (1990) article on the homeless in America. Researchers have no access to the interviews on which this article was based (again, proof texts have little critical utility) and, in most cases, no firsthand knowledge of the plight of the homeless. They are, therefore, poorly positioned to judge the quality of the interpretations offered in this article.

In plotting these articles on our virtual canon continuum, we do not mean to suggest that articles on the non-canonical end of the continuum offer less valid interpretations than those at the canonical end. We assert only that researchers are ill equipped both individually and collectively to evaluate the quality of interpretations in the more non-canonical articles. And we have a vague sense that reviewers may be somewhat inhospitable to interpretive analyses of phenomena that fall outside their experience since they are not equipped to assess the value of these studies. This is unfortunate since compulsive consumption and consumption among the homeless are legitimate topics for consumer research. Presumably, a canon would open up more opportunities for studying abnormal consumer behaviors such as these because it would make data on these behaviors more widely available and would facilitate informed assessment of articles treating these phenomena.

THEORETICAL AND PRACTICAL BENEFITS OF A CANON

We have already suggested reasons why a consumer behavior canon would enhance the credibility and validity of interpretive consumer research and might facilitate research on abnormal behaviors. We discuss, now, other theoretical and practical benefits that a consumer behavior canon might generate.

Deeper Insights from Close Readings

In ther critical review of a set of interpretive studies, Calder and Tybout (1989) offer, among others that we find unpersuasive, one compelling criticism: they suggest that the findings of interpretive researchers often fit the data so loosely that they do not deliver the promised richness of the qualitative approach. They mention as an example the findings that place is important for a retail store, a finding that tells us nothing we don’t already know? Though interpretive research has generated some insights much deeper than this one, our broad sense is that it has not delivered in consumer research the kind of aspect-rich descriptions that it routinely produces in more mature interpretive disciplines such as the humanities. For comparison purposes, see Janice Radaway’s (1984) wonderfully nuanced analysis of the function of romance novels in the lives of the women who buy them, a paper that could have been published in a consumer behavior journal but was actually published in Daedalus, a journal in the humanities.

We attribute the dearth of compelling close readings in consumer research to our discipline lacking the two essential elements of mature interpretive research: a canon and a tradition of critically reinterpreting texts. The lack of a canon is important because researchers cannot engage a phenomenon in detail if they cannot assume their readers will be familiar with the textual details. Our loose, virtual canon does not provide the familiarity that is requisite if we are to have close readings, the kind of readings that most fully illuminate texts and disclose the patterned complexity of the phenomena reflected in them.

The lack of a tradition of critical reinterpretation also leads to comparatively shallow readings. The finest readings of cultural texts usually emerge over time through a process of critical point and counterpoint. Our one-shot tradition provides no opportunity for Holbrook and Grayson’s (1986) analysis of consumption in Out of Africa to be deepened by subsequent critical responses that challenge and modify the initial interpretation, and it provides even less for Hill and Stamey’s (1990) findings to be modified by a reinterpretation of their interviews with homeless people.

Expanded Data Availability, Lower Data Costs

A canon would permit us to expand the availability of data while lowering its cost and increasing our understanding of it. Interpretive research typically uses rich data. It is, therefore, wasteful to spend a great deal of time gathering the data, then subject it to only a single interpretation. Most any set of well conducted interviews can be profitably interpreted from multiple perspectives. But because we lack a canon in which data are shared, single interpretations are normal practice. A colleague of ours recently spend a great deal of time interviewing informants on their collecting hobbies even though she knew other researchers who already had a large set of interviews on this topic. It did not occur to her to use the existing data set or to those who had the data to make it available to her. And yet a critical reinterpretation that responded both to the existing data and to the previous interpretation of that data would probably have yielded deeper insights into collecting behaviors than another, one-shot interpretation of new data. While new data are always welcome, our discipline would certainly benefit if much more time were spent interpreting and reinterpreting the rich data we already have, much less laboriously gathering new data. Though reading the canon takes time, one can carefully read a number of existing interviews in the time it takes to set up, conduct, and transcribe one new one. Provided that the discipline is prepared to accept and publish reinterpretations, carefully reading the canon should prove to be a productive use of researcher time.

It is certain that consumer research should be open to novel reinterpretations of existing data. Judging from the history of literary criticism, we need not suppose that the person who gathers the data and is most deply immersed in it is best positioned to interpret it. Plato long ago pointed out in the Ion that poets are very often poor interpreters of their own poetry. The same may well be true of researchers who collect data. Time spent reading theory will sometimes lead to deeper insights into the meaning of a set of interviews than time spent living with or otherwise immersing oneself in the lives of the informants. But the researchers best equipped to interpret a given data set may not be best positioned to gather it, either because they are not proximate to the informants or because they are not skilled interviewers. A canon permits a division of labor and, thus, a more efficient use of various researchers, whether their skills most lie in data collection or data interpretation. Of course, the discipline must arrange to reward those who collect as well as those who interpret good data.

Broader Perspectives, Larger Samples

Paradoxically, a canon makes possible not only a deeper immersion in the details of consumer behavior (and, thus, closer readings) but also a much broader exploration of changes and continuities in consumer behavior across expanses of space and time. A researcher with access to a consumer behavior canon will likely be much better positioned to study the ways in which self-gift giving differs from the Midwest to California to New Zealand to Singapore than any researcher could be who must draw only upon self-generated data. And a sample drawn from a canon is certain to be larger than any self-generated sample since the existing data could be added to any new data, and the new data, would, in turn, expand the scope of the canon.

As for time, next to its epistemological benefits, perhaps the most important benefit of a canon would be the development of a new degree of historical consciousness in consumer behavior research. Fully mature interpretive disciplines such as those in the humanities have a deep understanding and appreciation of the importance of history, an understanding that the natural sciences have traditionally lacked for reasons Kuhn (1969) has laid out. Perhaps owing to the influence of positivism, consumer researchers have paid relatively little attention to historical developments that have undoubtedly had very important effects upon consumer behavior and that business practitioners, an important constituency, must deal with every day (Smith and Lux 1993). Interpretivist researchers have no reason to ignore history, but in the absence of a canon, they will have little opportunity to study changes in consumption across time since they will lack access to historical data. Such longitudinal interpretivist research as has been done confirms the importance of institutions that preserve data and make it publicly available. Belk and Pollay (1985), for instance, used data preserved in archives to produce a longitudinal study of advertising.

While regional and historical influences would probably receive much attention were a consumer behavior canon to be formed, the range of possible perspectives that could organize the canon and guide research extend well beyond time and space. Judging from other disciplines, a multiplicity of perspectives distributed up and down the ladder of abstraction could and would be used to select relevant data from a consumer behavior canon. These perspectives would likely include but not be limited to product classes (automobiles, shampoo, travel services), demographic groups (women, Hispanics), classes of behavior (binge eating, charitable giving), and conceptual frameworks (structuralism, hermeneutics). A canon would provide a data set rich enough to sustain well grounded work from all of these and many other perspectives.

A BRIEF EXAMPLE

We conclude with an example of the kind of textual reinterpretation that we have been urging. Our focus is Mr. E’s Elephant Museum and Roadside Attraction, the business in Gettysburg, Pennsylvania, that is profiled in the Marketing Science Institute video "Deep Meaning in Possessions." Offering a "teleological," trusting interpretation that is consistent with Mr. Ed’s own view, Belk, Wallendorf, and Sherry (1989) note that this business has two sides, the sacred Elephant Museum (from which items are never sold) and the profane Roadside Attraction (where Mr. Ed earns his living selling elephant paraphernalia and other souvenirs). We offer a suspicious, unmasking interpretation of this same business, what Ricoeur (1974) calls an "archaeological" hermeneutic, an interpretation that Mr. Ed would probably reject.

Derrida (1973) has pointed out that putatively balanced dichotomies such as rich/poor, mind/body, man/woman, good/bad are actually unbalanced, the first term being preferred or valorized. Thus, in the sacred/profane video and article, the sacred receives much more attention than the profane, for it is the more deeply meaningful and important of the two domains in the minds of those who have created this distinction. Derrida also suggests that by inverting the valorization, one deconstructs a reading and discloses new dimensions of meaning.

Our deconstructive reading builds upon the Marxist distinction between base and superstructure (Marx 1988), a dichotomy that quite neatly inverts the sacred/profane distinction. For Marx, the drive to acquire material wealth is fundamental; meaning making, including concepts of the sacred, subserve that fundamental drive. In this Marxist frame, the more important part of Mr. Ed’s business is the Roadside Attraction, the side that produces wealth. The Elephant Museum merely supports this base, perhaps by linking the business to Lincoln, a founder of the Republican party, certainly by attracting customers, indoctrinating them in the sacredness of souvenirs, and modeling for them behaviors that will help sustain the Roadside Attraction. It encourages consumers to prominently display, talk about, and never sell the souvenirs they have purchased. In other words, it fosters word-of-mouth promotion and erects an ideological barrier to competition, all the while lending dignity to an establishment that might otherwise be viewed as tourist trap full of kitsch junk.

Neither our skeptical, perhaps even mean-spirited discussion of this business, nor the more charitable sacred/profane analysis fully encompasses this phenomenon. But in offering a second interpretation, we have presumably thickened the description and increased the aspect density of our understanding, in other words, have produced a more adequate composite interpretation.

REFERENCES

Belk, Russell W. and Richard W. Pollay (1985) "Images of Ourselves: The Good Life in Twentieth Century Advertising" Journal of Consumer Research, 11 (March), 887-897.

Belk, Russell W., Melanie Wallendorf, and John F. Sherry (1989), "The Sacred and Profane in Consumer Behavior: Theodicy on the Odyssey," Journal of Consumer Research, 16 (June), 1-38.

Calder, Bobby J. and Alice M. Tybout (1989), "Interpretive, Qualitative, and Traditional Scientific Empirical Consumer Behavior Research," Interpretive Consumer Research, ed. Elizabeth C. Hirschman, Provo, UT: Association for Consumer Research, 199-208.

Dasenbrock, Reed Way (1986), "Accounting for the Changing Certainties of Interpretive Communities," Modern Language Notes, 101 (5), 1022-1041.

Derrida, Jacques (1973), Speech and Phenomena, Evanston, IL: Northwestern University Press.

Fish, Stanley (1980), Is There A Text In This Cass?" Cambridge: Harvard University Press.

Fowler, Alastair (1982), Kinds of Literature: An Introduction to the Theory of Genres and Modes, Cambridge: Harvard University Press.

Frye, Northrop (1971), The Anatomy of Criticism, Princeton: Princeton University Press.

Gadamer, Hans-Georg (1975), Truth and Method, New York: The Seabury Press.

Hill, Ronald Paul and Mark Stamey (1990), "The Homeless in America: An Examination of Possessions and Consumption Behaviors," Journal of Consumer Research, 17 (December), 303-321

Hirschman, Elizabeth C. (1986), "Humanistic Inquiry in Marketing Research: Philosophy, Method, and Criteria," Journal of Marketing Research, 23 (August), 237-249.

Hirschman, Elizabeth C. (1990), "Secular Immortality and the American Ideology of Influence," Journal of Consumer Research, 17 (June), 31-42.

Holbrook, Morris B. and Mark W. Grayson (1986), "The Semiology of Cinematic Consumption: Symbolic Consumer Behavior in Out of Africa," Journal of Consumer Research, 13 (December), 374-381.

Holt, Douglas B. (1991), "Rashomon Visits Consumer Behavior: An Interpretive Critique of Naturalistic Inquiry," Advances in Consumer Research, eds. Rebecca H. Holman and Michael R. Solomon, 18, 57-62.

Lincoln, Yvonna S. and Egon G. Guba (1985), Naturalistic Inquiry, Beverly Hills, CA: Sage Publications.

Marx, Karl (1988), Selected Writings, ed. David McLellan, Oxford: Oxford University Press.

McCracken, Grant (1989), "Who is the Celebrity Endorser? Cultural Foundations of the Endorsement Process," Journal of Consumer Research, 16 (December), 310-321.

Pierce, Charles Sanders (1934/1965), "The Fixation of Belief," in The Collected Papers of Charles Sanders Peirce, Cambridge: Harvard University Press, 5, 223-247.

Pepper, Stephen C. (1970), World Hypotheses: A Study in Evidence, Berkeley: University of California Press.

Rook, Dennis W. (1987), "The Buying Impulse," Journal of Consumer Research, 14 (September), 189-199.

Smith, Barbara Herrnstein (1984), "Contingencies of Value," in Canons, ed. Robert von Hallberg, Chicago: University of Chicago Press.

Smith, Ruth Ann and David S. Lux (1993), "Historical Method in Consumer Research: Developing Causal Explanations of Change," Journal of Consumer Research, 19 (March), 595-610.

Stern, Barbara B. (1995), "Consumer Myths: Frye’s Taxonomy and the Structural Analysis of Consumption Texts," Journal of Consumer Research, 22 (September), 165-185.

Thompson, Craig J. (1990), "Eureka! And Other Tests of Significance: A New Look at Evaluating Interpretive Research," in Advances in Consumer Research, eds. Marvin E. Goldberg, Gerald Gorn, and Richard W. Pollay, Provo, UT: Association for Consumer Research, 25-30.

Wallendorf, Melanie and Russell W. Belk (1989), "Assessing Trustworthiness in Naturalistic Consumer Research," in Interpretive Consumer Research, Provo, UT: Association for Consumer Research, 69-84.

Wittgenstein, Ludwig (1969), On Certainty, New York: Harper & Row.

----------------------------------------