Presidential Address Positivism, Naturalism and Pluralism in Consumer Research: Paradigms in Paradise

Richard J. Lutz, University of Florida
[ to cite ]:
Richard J. Lutz (1989) ,"Presidential Address Positivism, Naturalism and Pluralism in Consumer Research: Paradigms in Paradise", in NA - Advances in Consumer Research Volume 16, eds. Thomas K. Srull, Provo, UT : Association for Consumer Research, Pages: 1-8.

Advances in Consumer Research Volume 16, 1989      Pages 1-8



Richard J. Lutz, University of Florida

[I am grateful for comments on a draft of this paper by Paul Anderson. David Brinberg, David Mick, and Alan Sawyer.]

I stand before you today with the charge of saying something meaningful. I have considered, and discarded, topics such as the state of the Association, about which I care deeply; consumer response to advertising and product trial, which represents my most central research arena, and intergenerational transfer of consumer preferences and shopping behaviors, representing an emerging interest of mine. Instead, I have chosen to speak to you, in as brief and clear a fashion as possible, about issues emanating from the philosophy of science, or science studies, as it is more fashionably labeled. I speak to you, not as a practicing philosopher, but as a practicing consumer researcher. The goal of my remarks is to begin to bridge the gap between the level of philosophical discourse that characterizes the science studies literature and the so-called "workbench" level of activities that occupy consumer researchers in their day-to-day pursuit of knowledge.

I have chosen this topic for discussion, not because I am an expert (or anywhere close to it) on these issues, but because it represents an area that has demanded much of my attention over the past several months. As the editor of the Journal of Consumer Research, I have been faced with decisions regarding numerous manuscripts reporting on research efforts emanating from a research tradition that has been variously labeled postpositivism, interpretivism, postmodernism, and naturalism. [Hereafter, I shall use the term "naturalism" to refer to this paradigm, since I regard it the most descriptive of the various labels. As noted by Hudson and Ozanne (1988), investigation of phenomena in natural settings is the hallmark of this form of inquiry.] I have accepted some of these papers, and rejected others, based largely on the inputs of trusted reviewers skilled in the methods attendant to this tradition. At the same time, I have been forced, though quite willingly and with great interest, to learn more about this general form of inquiry. My intent today is to share with you some of what I think I have learned, and to sketch out a tentative blueprint for the future of our discipline, which I fervently hope will achieve a balance between the canons of positivism, and the dicta of naturalism, to forge a more eclectic view of the scientific enterprise which I will label, for present purposes, pluralism .

Hence, the title of my remarks today is "Positivism, Naturalism and Pluralism in Consumer Research: Paradigms in Paradise," the subtitle giving an explicit nod to the exotic surroundings in which these (hopefully not too) pedestrian thoughts are offered for your consideration.

In their analysis of the first fourteen ACR Presidential Addresses, Spiggle and Goodwin (1988) noted that not one had addressed the emergence of an alternative paradigm in consumer research. Even before I read their paper in my pre-speech evaluation apprehension frenzy, I had already decided to redress that deficiency. The field of consumer research is r..cst certainly experiencing what Kuhn (1970) identified as a paradigm shift. A significant number of our membership has rejected the tenets of positivism and has turned instead to naturalism as the guiding paradigm for knowledge generation. Ethnography, literary criticism, historicism, auto-driving, and other methods have become increasingly common in our field, bringing with them a bewildering array of criteria by which such research endeavors are to be judged. How is one to assess the "contribution to knowledge" arising from paradigms that differ widely in their ontological, axiological, and epistemological assumptions? [As discussed by Kuhn, a paradigm is a set of assumptions that is adhered to closely by researchers working in a particular venue. Kuhn's original discussion of the concept was remarkably fluid, with at least 22 different meanings attributable to the term in his introduction of it. Nevertheless, it has endured as one of the most powerful (and elusive!) concepts in the science studies literature.] [To paraphrase Karl Malden, "What is an editor to do?!"]

In attempting to resolve this rather imposing question, I have been particularly influenced by two works, Brinberg and McGrath's incisive treatise Validity and the Research Process, and Paul Anderson's exposition of a view of science he calls critical relativism. The former has assisted me immeasurably in framing the problem, while the latter has offered me a mechanism for beginning to solve the problem as I have framed it.

Brinberg and McGrath's (1985) core contribution was their Validity Network Schema (VNS), depicted visually in Exhibit 1. The VNS portrays the research enterprise as consisting necessarily of three domains: the conceptual, the substantive, and the methodological. Each domain is further decomposed into three levels, i.e., elements, relations, and embedding systems, as shown in Exhibit 2. My interpretation of the VNS is that the embedding system level within each domain represents paradigm-like aspects of that domain, as suggested by the use of the term "conceptual paradigm" by Brinberg and McGrath. According to Brinberg and McGrath, most of the Kuhnian construals of paradigm entail conceptual paradigms (e.g., behaviorism), though some references implicate methodological paradigms (e.g., the role of introspection). Brinberg and McGrath, in a self-proclaimed Procrustean venture, introduce the notion of a substantive paradigm (i.e., a set of guiding axiological assumptions regarding what is worth scientific study).



For clarity, and representative of a partial reinterpretation of Brinberg and McGrath's partitioning, I have re-labeled the relation and embedding system levels of the methodological and substantive domains (See Exhibit 3). My reinterpretation of the substantive domain entails treating the relations level as a substantive system rather than simply patterns among selected elements in the substantive domain. Hence, I am suggesting that the relations level be regarded as somewhat more inclusive than is depicted by Brinberg and McGrath. Any pattern of relationships, including highly complex and potentially exhaustive ones, would be the unit of analysis at that level. In turn, the substantive domain's embedding system level would consist of a set of substantive paradigms, each of which is characterized by its substantive content and, importantly, by guiding axiological assumptions that designate what is "worth" scientific study. Thus, the substantive embedding system level takes on the paradigm-like feature of encompassing a "worldview" that implicitly governs researcher choices in that venue. Note that multiple substantive paradigms may exist in a general area of inquiry, reflecting different emphases or foci of researcher activity.



In the methodological domain, my reinterpretation is similar in structure. I move research strategies to the level of relations and introduce the notion of a methodological paradigm at the embedding system level. The logic for this shift is analogous to that underlying the shift in the substantive domain. Research strategies encompass, at a higher level of abstraction and in more complexity, the relations among methods. Research strategies are dependent upon the perceived relations among various methods and hence represent the methodological equivalent of the substantive system discussed previously. Furthermore, multiple research strategies can exist within any given methodological paradigm. As with substantive paradigms, methodological paradigms carry with them certain assumptions, in this case epistemological in nature, regarding proper modes of knowledge production.

In order to put some meat on these rather abstract bones, let me offer two sets of examples. First, one of "normal science" (Kuhn 1970) as currently practiced in consumer research. This is reflective of the dominant positivist paradigm in our field. As shown in Exhibit 4, a prototypical positivistic consumer research endeavor would consist of elements from each of the three domains in the VNS. For example, the attitude construct might be used to explain brand choices generated in the context of a laboratory experiment. Each of these elements is representative of a broader array of elements and relations. Moving down the Substantive Domain column, for instance, one sees that the brand choice element is but one of many in a marketer-consumer system of exchange. Such marketer-consumer relations are embedded in an economic system paradigm. Similar hierarchical organizations exist within the Conceptual and Methodological Domain columns as well.

Note that the entries in each column of the Embedding System row are representative of paradigms (with a small "p", if you will) within each domain. In reality, a Paradigm (with a capital "P") in the way I am construing it, embodies the entire Embedding System level. That is, a Paradigm consists of a worldview that spans the conceptual, substantive and methodological domains.

As acknowledged by Brinberg and McGrath, though a paradigm shift may originate in any one of the three domains, the impact is eventually an necessarily felt in all three domains. Hence, a paradigm shift, in VNS terms, consists of adopting a new set of ontological, axiological, and epistemological assumptions that collectively comprise the "worldview" characterizing the new paradigm.

Note that on the VNS view, paradigm shifts entail changes at the embedding system level. Changes at the element or relation level are regarded merely as innovations within an existing paradigm (Kuhn's "normal science ") . For example, Jim Bettman's 1987 ACR Presidential Address offered innovation within the existing decision making paradigm. At the Element level Jim focused on the construct of adaptivity in decision making. Innovations at the Relation level were represented by specification of conditions moderating the nature and degree of consumer adaptation in the decision process. Such changes are evolutionary, while changes at the Embedding System level, i.e., paradigm shifts, are more revolutionary.

Consider, for example, aspects of the Naturalist paradigm depicted in Exhibit 5. The substantive domain was represented prominently in Russ Belk's passionate call for macro consumer behavior research in his 1986 ACR Presidential Address and contrasts sharply with the substantive domain discussed earlier. For instance, rather than focusing on the micro-level phenomenon of brand choice, the broader phenomenon of societal materialism is of interest. Al the relations level, an alternative "system" of consumption's relationship to societal well-being replaces the marketer-consumer system discussed previously. Finally, the substantive paradigm is one which prioritizes individual and societal quality of life rather than economic performance. Turning now to the conceptual domain, we can think in terms of the conceptual paradigm emerging from the Consumer Behavior Odyssey (Belk, Wallendorf and Sherry 1989). At the element level, the construct of sacredness is used to distinguish consumption objects of extreme personal significance to individuals and groups. The various processes by which objects attain or relinquish sacredness is a corresponding example at the relations level. Finally, the conceptual paradigm housing this analysis might best be characterized as the sociology of religion. This clearly represents a much different "worldview" of consumer behavior than that offered by the dominant psychological information processing paradigm.



Finally, consider the methodological domain, where, in my opinion, is where the real action is, because it is in this domain that questions of epistemology arise - i.e., how research is evaluated for its merits in generating new knowledge, or "truth." There are major and obvious differences between Positivism and Naturalism at the element and relations levels; ethnography and audits, both highly qualitative procedures, share few if any features with experimentation and its attendant control and manipulation. However, the differences at the embedding system level are even more profound. Whereas the Positivist aspires to causal explanation, the Naturalist eschews notions of linear causality and instead seeks Verstehen, a richly textured understanding of the phenomena of interest. Thus, the very roots of scientific inquiry are axiomatically different and affect epistemological judgments about the merits of various methods and strategies.

It is in the methodological domain that the differences between the Positivist and Naturalist paradigms are most striking, and the attempt at violent overthrow most apparent. For example, in contrasting Positivism and Naturalism, Lincoln and Guba (1985) resort to rather inflammatory rhetoric. Positivism is variously characterized as inadequate, repugnant, unfounded, untenable, and ignorant of humanness, although Russ Belk's (1987) charge that consumer researchers have wasted their time on the "dog-food level of things" (p.2) has something of the same flavor (no pun intended) in the substantive domain. Lincoln and Guba's somewhat more dispassionate summary of key axiomatic differences between Positivism and Naturalism is shown in Exhibit 6.

From the VNS perspective, an important (perhaps the most important?) difference between Positivism and Naturalism is the order in which the domains are considered in any inquiry, Conceptual-Methodological-Substantive versus typically Substantive Methodological-Conceptual, respectively. As discussed by Brinberg and McGrath, successive research choices in the domains set constraints on the range of possibilities in the remaining domains. The fact that the Positivist paradigm typically begins m the Conceptual domain, while the Naturalist paradigm ends in it and vice versa, underscores the very fundamental differences in worldview between these two paradigms. [This parallels a similar analysis of the academic practitioner debate in the field of marketing (Brinberg and Hirschman 1986).]

Having outlined the nature of the problem, i.e. a dominant paradigm and a strong challenger, where does this leave us as a field? As Sherry (forthcoming) has noted, at least two possibilities exist - complete schism or a re-integration into some sort of more pluralistic paradigm. [Hudson and Ozanne (1988, p.519) have identified some additional possibilities that I will not consider here.] However difficult the latter ma v be to achieve, the former in my opinion would be disastrous.





Paul Anderson's critical relativism appears to offer a philosophical foundation upon which a pluralistic paradigm for consumer research can be erected. In fact, in a recent critique of critical relativism, Siegel (1988) suggested that it might be relabeled critical pluralism, and Anderson (1988) in his rejoinder did not disagree. Building on Laudan's (1984) reticulated model of scientific rationality, one of Critical Relativism's central notions " that there exists no single 'scientific method.' Instead, disciplinary knowledge claims are viewed as contingent upon the particular beliefs, values, standards, methods, and cognitive aims of its practitioners...there are multiple scientific objectives and alternative methods for attaining these objectives... [knowledge] claims must be assessed in light of their unique modes of production and their methods of justification" (Anderson 1986, p. 156). To quote Anderson further, "A critical relativist demands to know a program's methodological, ontological, metaphysical, and axiological commitments before he or she is willing to grant epistemic authority to its knowledge products. .Most importantly, critical relativists want to know a program's realizable cognitive (and social) aims before they are willing to give it serious consideration" (p. 167).

One aspect of Critical Relativism with which I take issue is its emphasis on the sociological forces at work in knowledge production. I accept such factors as descriptive of scientists' behavior, but I do not share Anderson's conclusion that such considerations must enter into normative statements about the practice of science. Time and space do not allow me to pursue this disagreement here; let me simply indicate my desire to reduce the emphasis on sociological factors by accepting Siegel's label of Critical Pluralism.

What are the implications of Critical Pluralism for the day-to-day practice of consumer research? That is, how does the philosophical translate into the pragmatic? A whimsical subtitle for this concluding segment of my remarks might be "Notes from Paul Anderson's Workbench," though I am not at all certain that Paul would want to endorse all of them.

First, each of us needs to make a commitment to learn more about these science studies issues. We can no longer afford to be like a fish in water, not knowing he or she is wet. Our work, whether as researchers or reviewers, implicitly reflects a guiding paradigm. We can benefit as individual researchers by making those ontological, axiological and epistemological assumptions explicit, at least in our own minds. At a minimum, then, each of us needs to understand the current dominant paradigm in consumer research.

Then, we need to learn more about the alternatives to positivism. Just as consumers search for information in order to make informed purchases we owe it to ourselves to make informed choices about how we seek knowledge. Not the least of these choices resides in the substantive domain. What are we attempting to achieve through our research efforts? Whether it is lo assist practicing marketing managers in making advertising decisions or to address macro consumer behavior issues, we need to devote more serious effort to thinking about substantive domain issues. As Belk (1987) implied and Anderson asserted in his 1983 paper, a discipline achieves scientific respectability only when it is widely perceived as addressing questions of social significance.

Simultaneous with our own "consciousness-raising" (Mick 1988) regarding these issues, we owe it to our students to raise their consciousness as well. The terminal degrees we bestow are Ph.D.s, and we need to put more philosophy behind that Ph.

In our day-to-day research, we should strive to make more explicit our paradigmatic assumptions (cf., Hudson and Ozanne 1988). As I have attempted to demonstrate, knowledge claims are judged by dramatically different criteria, depending on the paradigm upon which they rest. Making the paradigm explicit helps guide research strategies throughout the research enterprise. For example, a positivist seeks to maximize causal explanation, whereas a naturalist strives for "thick description."

Ideally, the review process will be a flexible one in which the explicit nature of the paradigm governing the research makes it conceivable to identify reviewers who are both willing and able to adopt the criteria attendant to that worldview in judging the research. [Of course, this does not mean that a reviewer should uncritically accept a researcher's criteria as probative, especially if they do not serve to achieve the researcher's cognitive aims.] We need more Morris Holbrooks, scholars who can move easily between positivism and naturalism. One can almost envision a cover letter of submission that specifies the ontological, axiological and epistemological assumptions guiding the work (though this idea may move beyond the workbench to the cookbook! ).

Finally, we need to read and benefit from one another's work. Consumer research is a vital and vibrant field, due in large part to its diversity. We need to respect the legitimacy of different paradigms and draw on the eclectic insights offered by this variety of approaches. By cultivating a more sophisticated pluralistic view of the research enterprise, we can actually learn more from each other.

Spiggle and Goodwin speculated that the solidarity-producing function of the ACR Presidential Address has caused past presidents to ignore the present paradigmatic divisiveness in the field. I hope that my remarks today have not only recognized our current differences and addressed them head-on but also have begun to pave the road to re-integrating the two camps. In his 1981 ACR Presidential Address, 'Toward a Science of Consumer Behavior," Jerry Olson issued a plea for more tolerance of new theoretical perspectives. To this, I would add a plea for methodological and substantive domain tolerance as well. If we are successful in adjusting and enlarging our individual worldviews, the field of consumer research may take a giant leap toward achieving full fledged, intellectually pluralistic, disciplinary status. I encourage each of you to join with me in pursuit of that objective.




Anderson, Paul F. (1983), "Marketing, Scientific Progress, and Scientific Method," Journal of Marketing, 47 (Fall 1983), 18-31.

Anderson, Paul F. (1986), "On Method in Consumer Research: A Critical Relativist Perspective," Journal of Consumer Research, 13 (September) 155-73.

Anderson, Paul F. (1988), "Relative to What - That is the Question: A Reply to Siegel," Journal of Consumer Research, 15 (June) 133-7.

Belk, Russell W. (1987), "ACR Presidential Address: Happy Thought," in M. Wallendorf and P. Anderson (eds.), Advances in Consumer Research, Vol. XIV, Provo, UT: Association for Consumer Research, 14.

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

Bettman, James R. (1988), "Processes of Adaptivity in Decision Making," in M. J. Houston (ed.), Advances in Consumer Research, Vol. XV, Provo: UT, Association for Consumer Research, 1-4.

Brinberg, David, and Elizabeth C. Hirschman (1986), "Multiple Orientations for the Conduct of Marketing Research: An Analysis of the Academic/Practitioner Distinction," Journal of Marketing, 50 (October), 161 -73.

Brinberg, David and Joseph E. McGrath (1985), Validity and the Research Process. Beverly Hills, CA: Sage.

Hudson, Laurel Anderson, and Julie L. Ozanne (1988), "Alternative Ways of Seeking Knowledge in Consumer Research," Journal of Consumer Research, 14 (March), 508-21.

Kuhn, Thomas S. (1970), The Structure of Scientific Revolutions. Second Edition, enlarged. Chicago: University of Chicago Press.

Laudan, Larry (1984), Science an l Values, Berkeley, CA: University of California Press.

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

Mick, David Glen (1988), "The Future of Post-Positivistic Methods in Marketing: Naturalistic, Humanistic, and Interpretive Inquiry," in Proceedings, Summer Educators' Conference, Chicago: American Marketing Association, p.33.

Olson, Jerry C. (1982), "Presidential Address - 1981: Toward a Science of Consumer Behavior," in A. A Mitchell (ed.), Advances in Consumer Research, Vol. IX, Provo, UT: Association for Consumer Research, v-x.

Sherry, John F., Jr. (forthcoming), "Postmodern Alternatives: The Interpretive Turn in Consumer Research," in H. H. Kassarjian and T. S. Robertson (eds.), Handbook of Consumer Theory and Research, Englewood Cliffs, NJ: Prentice-Hall.

Siegel, Harvey (1988), "Relativism for Consumer Research? (Comments on Anderson)," Journal of Consumer Research, 15 (June), 129-32.

Spiggle, Susan, and Cathy Goodwin (1988), "Values and Issues in the Field of Consumer Research: A Content Analysis of ACR Presidential Addresses," in M. J. Houston (ed.), Advances in Consumer Research, Vol. XV, Provo, UT: Association for Consumer Research. 5 - 12.