Interpretation and Reinterpretation in Theory Construction and Application: the Views of Some Prominent Consumer Behavior Theorists

Elizabeth C. Hirschman, Rutgers University
ABSTRACT - Six prominent consumer behavior theorists -James Bettman, James Engel, Morris Holbrook, John Howard, Everett Rogers and Jagdish Sheth -- were queried on their views concerning (1) other authors' interpretations of their model, (2) whose property a scientific theory is, (3) the importance of preserving the originator's intentions when applying a theory, and (4) whether they relied on processes of subjective interpretation or objective truth-testing in constructing their theory. Their answers shed some light on the current debate over what is science.
[ to cite ]:
Elizabeth C. Hirschman (1989) ,"Interpretation and Reinterpretation in Theory Construction and Application: the Views of Some Prominent Consumer Behavior Theorists", in NA - Advances in Consumer Research Volume 16, eds. Thomas K. Srull, Provo, UT : Association for Consumer Research, Pages: 653-658.

Advances in Consumer Research Volume 16, 1989      Pages 653-658


Elizabeth C. Hirschman, Rutgers University

[The author gratefully acknowledges the cooperation of James Bettman, James Engel, Morris Holbrook, John Howard, Everett Rogers and Jagdish Sheth in conducting this inquiry.]


Six prominent consumer behavior theorists -James Bettman, James Engel, Morris Holbrook, John Howard, Everett Rogers and Jagdish Sheth -- were queried on their views concerning (1) other authors' interpretations of their model, (2) whose property a scientific theory is, (3) the importance of preserving the originator's intentions when applying a theory, and (4) whether they relied on processes of subjective interpretation or objective truth-testing in constructing their theory. Their answers shed some light on the current debate over what is science.


Recently a debate has surfaced in consumer research over the role of interpretive inquiry in a social science (cf, Anderson 1986; Cooper 1987; Calder and Tybout 1987; Holbrook 1987; Kernan 1987). Although the present author's personal views on this debate are known (cf, Hirschman 1985, 1986), I believed it might be worthwhile to query some prominent, practicing consumer behavior theorists as to their own views on the use of interpretive processes in theory construction and application. To that end, I wrote to James Bettman, James Engel, Morris Holbrook, John Howard, Everett Rogers and Jagdish Sheth and asked them to respond to four questions:

1. In general, how do you feel about other authors' interpretations of your model? Are there instances where you believe other authors have abused the model by misinterpreting your intentions? Are there instances where you believe other authors have made constructive reinterpretations of your model (i.e., have put it to desirable uses you did not envision)?

2. Whose property is a scientific model once it has appeared in the literature? Does it belong to the creator or to the scientific community at large? Why?

3. In general, do you believe that researchers applying a model developed by someone else should take steps to insure that their interpretation is consistent with the developer's intentions? Why or why not? If so, what might those steps be?

4. In constructing your model do you believe you relied more on a process of subjective interpretation or on a process of objective truth-testing?

Interpretation, Misinterpretation, and Reinterpretation

The first question asked the theorists to comment on the processes of interpretation, misinterpretation and reinterpretation as applied to their theories by other researchers.

Rogers reported that he was generally satisfied with other researcher's interpretations of his models of the diffusion and adoption processes for innovations (Rogers 1962; Rogers and Shoemaker 1971).

"Other's interpretations [of the innovation diffusion and adoption models] are mainly correct.'

Howard elaborated on how other colleagues at Columbia had extended his early model (Howard 1963) of consumer problem solving:

I was immediately pleased when John Farley suggested that he would mathematically model the simplified version that had emerged. It emerged as I designed the data collection procedures for the instant breakfast study that General Foods had generously agreed to finance. I was never troubled by how others interpreted the model. I find others' criticisms helpful; there was always so much to be done to improve it that criticisms never bothered me. There were a number of instances where others have made constructive reinterpretations of the model. For example, John Farley and Donald Lehmann put it to uses, such as in developing their "meta-marketing," that I never dreamed of. So did Michel Laroche in a number of ways add to it.

He was similarly sanguine concerning others' interpretations of the later Howard and Sheth model; especially noting the constructive extensions made by Bettman:

I am happy with others' interpretations. No one seems to have 'abused' the model. Clearly there were improvements. For example, Jim Bettman made a number of improvements between his point of view and the model. He had insights that we did not have.

Sheth, Howard's coauthor on the Theory of Buyer Behavior (1969), provided a detailed description of misinterpretations as well as constructive reinterpretations of their model:

Interpretation of the Howard-Sheth theory of buyer behavior by other authors is partially accurate. The most common misinterpretations or errors of omission are as follows:

a. The model is based on S-R paradigm of learning whereas, it is in fact, based on cognitive learning processes.

b. It relies on the concept of habit or routinized response behavior rather than on the more fundamental concept of simplification and complication.

c. It is limited to frequently purchased consumer goods, and therefore, not applicable to infrequent purchases or industrial buying behavior.

d. The model is limited to brand choice behavior and therefore, cannot explain product class choice.

e. The brand choice is a direct outcome of consumer's attitudes and intention even though the theory clearly discusses the concept of inhibitors which intervene between intentions and choice.

f. The major error of omission is not to recognize the theory as a contingent theory. There are several exogenous variables discussed in the theory of ceteris pariba. These include importance of purchase, time pressure, social class, culture, personality, etc. which also influence choice behavior.

A number of other authors have made constructive reinterpretations and extensions of our theory. These include:

a. Extending the theory from an individual to a unit such as family or organization buying behavior.

b. Reconstructing the theory for product class choices.

c. Extending the concept of novelty, curiosity and exploratory behavior as a major research issue in consumer behavior.

d. Acceptance of satisfaction/dissatisfaction as a discrepancy between expectations and experience and building alternative models around consumer satisfaction/dissatisfaction.

e. Further research into several exogeneous variables, especially importance of purchase or involvement and time pressure.

Regarding misinterpretations and constructive reinterpretations of the Engel-Kollat-Blackwell model of consumer decision making (1968), Engel wrote:

Generally speaking I am comfortable with the way others have interpreted our model. We never tried to set that up as anything definitive or as any kind of theory in the sense that I believe Howard and Sheth first did. We viewed it always as a pedagogical tool, and that's why it's changed. Also, it has proved useful to people in the applied world as they have attempted to see which variables affect decisions. Generally speaking people have interpreted us correctly, I think. I suppose constructive reinterpretations have been many. I haven't sat down and tried to think about this too seriously, but I am sure people would juggle things around and modify the order and functional flow. I don't remember that that has been done often, but I know people do disagree with us about the specific details

Bettman was generally positive about others' interpretations of his theory of information processing and consumer choice (Bettman 1969).

I feel that most of the interpretations I have seen have been reasonable. Perhaps the propositions in the book helped with that. I'm drawing a blank on constructive reinterpretations, but I'm sure these have been done

Holbrook was gratified by the extensions of the experimental perspective (Holbrook and Hirschman 1982) along several novel avenues:

I usually feel flattered when others choose to (re)interpret my work. Seeing the ways in which the experiential and hedonic [Hirschman and Holbrook 1982] papers have been cited as support for various new aspects of consumer research has helped me to reinterpret the significance and implications of what we were doing. I cannot think of anyone who has misapplied or distorted our views, but I have been gratified by some creative extensions such as those by Puto, Wells, Batra, Ahtola and several others. At the moment, the most fruitful extensions appear to have moved in the directions of (1) emotions and (2) esthetics, but I hope that others will follow

Perhaps the most striking feature of these statements is the generally positive attitude exhibited by the original authors to others' interpretations. With the partial exception of Sheth, they seem to view most modifications and extensions of their ideas by others as desirable, constructive, and even necessary activities of scientific progress. In general, the original authors do not seem to express a proprietary or possessive attitude toward their ideas. This theme was explored more directly by the next question.

To Whom Does a Scientific Model Belong?

Rogers' response advocated universality: "It belongs to everybody--for them to make of it what they will." As did that of Howard:

A scientific model, once developed, belongs to anyone who wishes to extend it. It belongs to the scientific community at large to the extent that further development is contributed.

Engel further noted that one purpose of theory construction is to stimulate extensions of the theory by others:

In my opinion the people who put it forth in a sense lose control of it. After all, we put it out for other people to evaluate. Others then should feel free to take it and move with it from there. So I would say it belongs to the scientific community at large.

This same theme was reiterated by Bettman:

The community at large. The whole idea is to put forward ideas and hope that they lead to something of further interest. One would hope that others would see things missed by the originator or that the model would be heuristic in stimulating advances by others.

Finally, Holbrook viewed theories as belonging to humankind, in general:

I would say that an idea, once published or otherwise communicated, belongs to humanity. Depending on how you define 'scientific', I might want to expand the view of the relevant community of scholars. Specifically, I would want to include all aspects of the humanities in the relevant community All ideas deserve to be shared and considered in our efforts toward scholarship.

Thus, these original authors were unanimous in their belief that theories, once made public, belonged to the scientific community at large. In essence, research ideas were viewed as the fuel for generating new knowledge and, therefore, their open availability was deemed a powerful stimulus to the advancement of a field of inquiry.

The Significance of Authorial Intention

However, despite the view that scientific theories are universal property, it still may be viewed as appropriate scientific etiquette to seek approval from the originator of a particular model or concept before applying an interpretation. The authors were queried directly on whether other scientists should attempt to guarantee consistency of their application with the author's intention.

To which Rogers responded:

"Yes, generally. But the appliers may improve it through their applications. However, Howard took an opposing view: I do not feel that researchers applying a model by someone else should take steps to insure that their interpretation is consistent with the developer's intention. It is far more important that they be able to discern opportunities for improvement the developer missed. This is so, because the developer often has an implicit logic that he is not aware of until later. The applier can often smoke these out by discussion with the original builder, however.

Engel viewed the original authors' intentions as having some primacy as a starting point for interpretive applications:

Yes, I do think that researchers attempting to apply a model developed by someone else should first be sure that their interpretation is consistent with the developer's intentions. How can they apply it, if they don't fully understand? But I have no quarrel if they choose to modify it or change it from there. But at least let's start on the same base.

To Bettman, a distinction should be made between the original model and reinterpretations, with the burden of distinguishing between the two falling upon later researchers: .

I don't think it is necessary that the interpretation be consistent as long as the researcher indicates how it is and is not. I worry about operationalizing constructs differently than the originator intended; this may mean that a different construct is being used. A new interpretation may sometimes be useful in leading in directions unanticipated by the originator. A model may be useful in elucidating phenomena not envisioned originally. However, the investigator should be very clear about what is being done. I don't think one should present a reinterpretation as if it were the original model.

Finally, Holbrook noted that several contextual factors, in addition to authorial intention, could influence the validity of an interpretation:

I do not believe that fidelity to the author's intentions is an adequate criterion for validity in interpretation. Rather, the pre-occupation with authorial intent, while it may help to counteract an unhealthy tendency in some quarters to ignore the author entirely, appears much too narrow in light of all the other factors that impinge on meaning, e.g., the historical situation, the cultural context, the reader's response. I certainly would not go as far as the view that the author's intention is irrelevant, but I would argue against the false, positivistic premise that interpretation is a project of recovering the author's meaning.

These responses are illuminating in several respects. First, (with the partial exception of Holbrook) they each suggest some notion of authorial intention, purpose, or personal logic that is intrinsic to the integrity of the theory, itself. This supports the premise (cf, Hirschman 1985) that social narratives, including scientific texts, are constructed by people, and that although scientific theories are in one sense the 'universal' property of all scholars, they are also in another and perhaps equally important sense, the creative product of a particular scholar or scholars. Thus, those psychoanalysts who term themselves Freudian generally take care to communicate how they adhere to the tenets of Sigmund Freud and also to point out areas in which they may diverge from these principles. Similarly, as Anderson (1986) has recently described, marketing researchers who labeled their research as applying the Fishbein model of attitude structure were taken to task by other marketing researchers who viewed their applications as actually reinterpreting/altering the Fishbein model.

This leads to a second, deeper issue. The present responses suggest that scientific theories -even those deemed constructed on positivist premises -- are not independent of the subjective intentions, logics, and conceptions of their creators. The responses also imply that members of the scientific community, on an implicit level at least, recognize and accept the primacy of authorial intent as an inherent rule for applying the model (cf, Hirsch 1967, 1976); that is, scientific theories are not freestanding pieces of knowledge, but rather are the creative product and to some extent the intellectual property of the originating scientist. Thus, we are not free to do with them as we wish, unless we first seek some sort of approval from the author or clearly state at the outset of our inquiry that we are altering the creating scientist's theory in such-and-such a way.

The communal sharing of this implicit norm of scientific authorship becomes clear if we consider the ostracism that would surely befall a scientist who directly copied another's theory, and published it as his own with no acknowledgement of the earlier researcher's work. Thus, the narrative aspect of social science has an analog with other forms of textual construction; like the authors of plays, novels, historical treatises, and newspaper stories, scientific theorists also claim and are likely to be granted the authorial privileges of original intent and primacy of formulation.

Subjective Interpretation or Objective Truth-Testing

The extent to which these particular consumer behavior theorists perceived themselves to be creating their theories from processes of subjective interpretation or objective truth-testing was explored by the fourth question:

Rogers believed both processes played a role in his construction of the innovation diffusion model:

Approximately equal parts of each -- the diffusion model was based rather directly upon survey research results, but interpretation was also crucial in forming the model.

Howard also acknowledged the role of both in constructing his early theory of consumer problem-solving:

I relied on both interpretation and truth-testing. In the early years it was almost interpretation alone. There was even then, however, an intuitive sense of truth-testing, although not at all systematic. Later, as Farley, Lehmann and I worked together, truth-testing became a larger part of the enterprise. Still later, Farley and Lehmann focused upon their "meta-marketing" effort, which was heavily truth-testing.

He saw interpretation as playing a more dominant role in formulation of the later Howard and Sheth theory of buyer behavior:

I believe that I relied much on the process of interpretation in developing the model with Jagdish. I was continually looking for scientific support, but there was not much available at that time Discussions with Ray Bauer and reading Berlyne's material were immensely helpful, but overall these made up a small part of what was needed. Bill McGuire, then at Columbia, contributed much. - Johann Arndt's dissertation was very relevant.

Engel viewed both processes as relevant to development of the EKB model:

We tried to come up with a flow chart in 1968 and a model that represented the process as we understood it. I suppose there was the sense of objective truth testing in that [we believed] it should be a veridical model. But if you want to apply this in the formal sense of theory development and all the rigor that formal logic implies, I don't think we followed that to that degree of rigor. You might remember the book earlier on Metatheary by [Zaltman, Angelmar, and Pinson 1972]. I think they gave us pretty good marks in terms of theory development, if I remember right.

Bettman reported placing great emphasis on personal interpretation in constructing his theory of consumer information processing:

A process of interpretation - I used my own intuitions and my readings of many other theories, and tried to make a coherent 'story' out of it. In doing so, I found places where I needed to add new pieces to make the story work, e.g., memory phenomena and prior knowledge. I really did a lot of introspection about what things made sense to me. I wanted to have a model that fit my phenomenological experience of consumer choice.

Holbrook stressed the interpretive nature of all scientific activity in his response:

I do not believe that there is any such thing as 'objective truth testing.' All scientific effort involves interpretation. A scientific theory influences the measures and methods used to test it in the same way that a holistic gestalt influences our reading of specific elements as part of the hermeneutic circle. Anyone claiming to purge interpretation from his work commits a giant positivistic fallacy. It follows that, of course, I have relied heavily on interpretation -as have all other consumer researchers.

Perhaps the most surprising and striking element running through all of these theorists' descriptions is the acknowledgement of the role of personal interpretation in constructing their respective theories. Let us examine some of the reasons given for the necessity of interpretive inputs to theory construction. Roger's explanation suggests that interpretive processes may be useful for creating a coherent story out of pre-existing survey data. That is, the data may already exist -- in this case, concerning adoption/diffusion processes -- but a narrative must be constructed which makes sense of them; i.e., the diffusion/adoption theory. Howard's description of the interpretive processes involved in creating the Problem Solving and Buyer Behavior theories suggests that interpretation was required because adequate empirical data were not yet available on the phenomena of interest to him. Thus, of necessity he relied almost solely upon subjective interpretive processes in forming the early versions of the theory. These were then subjected to increasingly sophisticated empirical tests as other researchers became involved in the project. Thus, Howard's description is the reverse of Roger's account, suggesting that interpretive processes may either precede or follow formal empirical data-gathering.

Engel's description suggests that the E-K-B model resulted from a collaborative interpretive process among the three originators. After its initial formulation, the model was subjected to more formal empirical verification, although not of a strict falsificationist nature.

Finally, of particular interest was the similarity of the theory construction descriptions provided by Bettman and Holbrook. Despite the recent characterization of Bettman as an objective analytical scientist and Holbrook as a highly subjective conceptual humanist (Hirschman 1985), both described their theory development as primarily a process of personal interpretation. Bettman even goes so far as to state that he attempted to construct a model that "fit my phenomenological experience of consumer choice," [This may account for the appearance of the several Consumer Jerry Baker (i.e., J.B. = Jim Bettman) examples in the Bettman (1979) book.] while Holbrook, more generally, argues that all consumer research requires interpretive processes and that, at a minimum, the model of the experiential consumer was constructed in an interpretive fashion.


The foregoing comments by some prominent consumer behavior theorists are instructive in several regards. They are especially enlightening in documenting the personal, constructive aspects of science -- those processes of intuitive, interpretive creativity that are the soul of research inquiry not only in the social sciences, but the natural sciences as well (cf, Cole 1987). Theories must originate somewhere, and that location inevitably and invariably is in the mind of an investigator. Whether the theory is germinated by attempting to create a coherent narrative from raw survey data (e.g., Everett Rogers and the innovation diffusion paradigm), or by formulating an initial causal framework from bits and pieces of prior theorization and phenomenological inference (e.g., James Bettman and John Howard), or by iconoclastically constructing a dialectical image of human nature (e.g., Morris Holbrook), theory will, as indeed it must, contain elements of interpretation. Interpretation is not anathema to science; rather it is the fulcrum by which progressive change is effected in science.

Without interpreting phenomena and interpreting the work of others, novel scientific concepts could never emerge, novel scientific perspectives could never be formulated, and new paths for scientific inquiry could never be embarked upon. Further, without reinterpretation, existing theories could never be applied to novel contexts and novel phenomena; science would soon fold in upon itself and stagnate.

Researchers observe events in the-world, read other scientists' accounts of these events, or acquire knowledge from both these sources. However, merely acquiring knowledge is not sufficient to construct a theory. A theory must tell the story of some event. That is, it must integrate knowledge about the event into a narrative format, one that usually describes a causal scenario or pattern. Thus theories are narratives in that they tell why or how something happens, as opposed to merely labeling or identifying it.

Consumer behavior theorists construct stories about the behaviors of consumers. The several theorists whose views were examined here are individually and collectively excellent storytellers. Their theories have been recognized for the depth and

breadth of their impact on the field. [The group contains three ACR Fellows (Engel, Howard, and Rogers), three ACR Presidents (Bettman, Holbrook, and Sheth), and a JCR Editor (Bettman).] Yet, as they acknowledge, their narratives were each constructed in part or in whole from interpretive processes. Does this mean that their theories are 'just' stories -- stories on the same fictive plane as a myth or novel? Or that somehow they are the biased, distorted templates described recently by Calder and Tybout (1987)? Of course not! That a theory was constructed or applied using elements of interpretation means only that the scientist creating or applying it drew upon personal knowledge to make sense of the available data (Polanyi 1962). Scientists, as all humans, make sense of what they have experienced by interpreting it within pre-conceived textual structures (cf, Berger and Luckmann 1966; Ricoeur 1971).

Scientists differ from nonscientists in that they take care to document their reasoning while forming and altering their interpretations, and also to openly communicate those interpretations, in the form of theories, to other scientists for scrutiny. For doing this, they are acknowledged as the originators of the theory and, as described earlier, their interpretation of a phenomenon may serve as a template for others' interpretations and applications. Science grows and knowledge about consumer behavior expands as theoretical texts are generated, interpreted, and applied. Thus, to be a consumer researcher, one must be also an interpreter of consumer behavior.


Anderson, Paul F. (1986), "On Method in Consumer Research: A Critical Relativist Perspective," Journal of Consumer Research, Vol. 13, September, pp. 155-173.

Berger, Peter L. and Thomas Luckmann (1966), The Social Construction of Reality, New York: Doubleday and Company.

Bettman, James R. (1979), An Information Processing Theory of Consumer Choice, Reading, MA: Addison-Wesley.

Calder, Bobby J. and Alice M. Tybout, "What Consumer Research Is...", Journal of Consumer Research, Vol. 14, No. 1, (June), 136-140.

Cole, K.C., (1987), "A Theory of Everything," New York Times Magazine, October 18, pp. 20-28.

Cooper, Lee G., (1987), "Do We Need Critical Relativism?", Journal of Consumer Research, June, 126-127.

Engel, James L., David T. Kollat and Roger D. Blackwell, (1968), Consumer Behavior, Holt, Rinehart and Winston.

Hirsch, E.D., Jr. (1976), The Aims of Interpretation, London: University of Chicago Press.

Hirsch, E.D., Jr. (1967), Validity in Interpretation, New Haven: Yale University Press.

Hirschman, Elizabeth C., (1985), "Scientific Style and the Conduct of Consumer Research," Journal of Consumer Research, Vol. 12, (Sept.), pp. 225-239.

Hirschman, Elizabeth C. (1988), "Consumer Behavior Theories as Heroic Quest," Working Paper, New York University Graduate School of Business, New York City.

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

Holbrook, Morris B. (1987), "What is Consumer Research?", Journal of Consumer Research, June, 128-132.

Holbrook, Morris B. and Elizabeth C. Hirschman (1982), "The Experiential Aspects of Consumption: Consumer Fantasies, Feelings, and Fun," Journal of Consumer Research, September, Vol. 9, 132-140.

Howard, John A., (1963), Marketing Management: Analysis and Planning, rev. ed., Homewood, Ill.: Richard D. Irwin.

Howard, John A. and Jagdish N. Sheth, (1969), The Theory of Buyer Behavior, New York: John Wiley.

Kernan, Jerome B., (1987), "Chasing the Holy Grail," Journal of Consumer Research, June, 133-135.

Polanyai, Michael, (1962), Personal Knowledge: Towards a Post-Critical Philosophy, Chicago: University of Chicago Press.

Ricoeur, Paul, (1971), "The Model of the Text: Meaningful Action Considered as a Text," Social Research, Vol. 38, pp. 529-562.

Rogers, Everett M. (1962), Diffusion of Innovations, New York: The Free Press.

Rogers, Everett M. and F. Floyd Shoemaker, (1971), Communication of Innovations, The Free Press.