Volume 6, No. 2, Art. 33 – May 2005
The Interpretive Turn: History, Memory, and Storage in Qualitative Research1)
Abstract: This article reviews the field of qualitative inquiry, identifying three conceptual breaks: the "orthodox consensus" of positivism which conceives the social world as a collection of external facts and attempts to eliminate bias and subjectivity; post-positivist philosophy of science, which concedes that objective observation of pure data is impossible but nevertheless tries to establish criteria of "good" research practice; and the interpretive turn, which rehabilitates subjectivity and views data collection as a mutual construction of meaning where the researcher is engaged in "double hermeneutics" (GIDDENS). The interpretive turn has implications for history, memory, and storage of data. However, while recognizing the interactionist and contextual nature of data collection, the article points out that it is counterproductive to overemphasize its implications, as some postmodern strands of interpretive research do. Drawing on the hermeneutic notions of objectivation and the hermeneutic circle, it is argued that interpretive research data consist of objectivations, and therefore lend themselves to storage and future revision, newly emerging questions, and (re-)interpretation. Furthermore, data storage allows for data access by non-specialists, including the subjects of the research. Archiving consequently potentially contributes to empowerment, feedback and dialogue.
Key words: interpretive turn, qualitative data, history, memory, archiving, hermeneutics, dialogue, empowerment
Table of Contents
1. Conceptual Breaks within Qualitative Research
2. The Turn towards Interpretation
3. History, Memory, Storage
Qualitative research in the social sciences constitutes by no means a homogeneous, unified field. This text will first unpack the notion of qualitative methods to explore some of the conceptual breaks within this field. Second, and more importantly for the present volume, specific methodological implications of these breaks for the collection, storage, and re-analysis of qualitative data will be examined. 
Qualitative research can be defined in general terms as "multimethod in focus, involving an interpretive, naturalistic approach to its subject matter (...) Qualitative researchers study things in their natural settings, attempting to make sense of, or interpret, phenomena in terms of the meanings people bring to them" (DENZIN & LINCOLN 1998, p.3). Beyond this general orientation, qualitative research does not involve an a priori set of research techniques that researchers apply. Following authors such as LEVI-STRAUSS (1963), BECKER (1989), or DENZIN and LINCOLN (1998), the practice of qualitative research can be more usefully conceptualized as a form of bricolage: a putting-together of a set of research practices that aim to provide a solution to a concrete problem. In other words, the choice of qualitative research techniques depends on the research question that is being asked: it is problem-driven rather than method-driven. 
Beyond these general features, there are profound disagreements amongst qualitative researchers on core issues such as the nature of qualitative research or its fundamental theoretical assumptions. Such disagreements reflect a number of conceptual breaks within the field of qualitative research. Any attempt to present an exhaustive history of qualitative research is in itself contentious, and the aim of the present text is not to do so. Rather, I will concentrate on three major conceptual breaks within the field, in order to explore their implications for the issues of history and memory, and consequently for the more practical concerns of data collection, data storage (archiving), and secondary analysis of qualitative data. It is important to point out that although these three breaks have emerged successively, my intention is not to suggest a cumulative view of historical changes in the field where one perspective is thought to neatly replace the previous one. Indeed, all three perspectives are co-present within the domain of current qualitative research. 
I will argue that, at first sight at least, interpretive-qualitative perspectives in particular involve a potential distrust of the feasibility of secondary data analysis, and therefore of the usefulness of data storage. I will therefore concentrate in more detail on the concerns raised by interpretive perspectives in the social sciences, and will address some of these central concerns. In its early decades from the early 1900s to World War II, the field of qualitative research reflected the prevalence of positivist perspectives within the social sciences more generally. Qualitative researchers such as MALINOWSKI or the members of the Chicago School aimed to produce "objective" accounts of their observations of society, and to offer valid, reliable descriptions. Positivist conceptions of science traditionally attribute a negative role to researcher subjectivity. What DURKHEIM termed the researcher's "pre-notions," i.e. her pre-existing ideas or prejudices, are regarded with suspicion. Following the Cartesian ideal of methodic doubting, the subjectivity of the researcher is seen as a bias which obscures the accurate view of reality, whereas the object of study, social reality, is conceptualized as an external object. For positivist social scientists the concept of subjectivity, unlike objectivity, is used pejoratively. It is seen as a source of bias that needs to be eliminated from the research process. The research topic—the social problem under investigation—is treated as similarly independent from the researcher. It is "out there" and treated as "given." Consequently, provided the elimination of bias has been successful, the process of data collection and the nature of the collected data are seen as relatively unproblematic. Since the produced data are seen as objective observations of external reality and independent from the researcher, data storage and the re-analysis of data are in principle similarly unproblematic (even though they tend to be rare in practice). 
Positivism remained for a long time the "orthodox consensus" in social science methodology; this applied equally to qualitative research. Today, the orthodox consensus is dead. It has been shattered over the past two decades as a result of attacks from various quarters. Central to the crusade against the orthodox consensus was the rehabilitation of subjectivity. Indeed, the hegemony of positivist methods has been defeated by what RABINOW and SULLIVAN (1987) have aptly termed the interpretive turn in the social sciences, that is, the growing influence of disciplines such as ethnomethodology, phenomenology and hermeneutics. In addition to the interpretive critique of positivist and objectivist perspectives, other attacks have come from within a post-positivist philosophy of science, in particular by authors such as POPPER, KUHN, LAKATOS, FEYERABEND, HESSE and BHASKAR on the one hand; KOYRE, CAVAILLES, CANGUILHELM and BACHELARD on the other. Post-positivist philosophy of science is still concerned with problems such as verification or prediction, issues that were central to the traditional model of natural science as represented by CARNAP's logical positivism. Its proponents nevertheless clearly accept that scientific inquiry is largely of a hermeneutic nature. Interpretation and explanation, objectivity and subjectivity, cannot be clearly separated within natural science research. As BACHELARD stressed long before KUHN, HANSON or FEYERABEND, the objective observation of pure data is impossible. All scientific data are already interpreted at the same time as they are being observed. 
Critical developments within the natural sciences have also contributed to the decline of positivist views of science and to the revaluation of subjectivity within scientific research. The logical-positivist model of natural science is currently considered outdated by most natural scientists. Somewhat ironically, the strongest remnants of positivism are now possibly to be found in the social rather than in the natural sciences. Insofar as these approaches claim to follow natural science "scientificity," they seem to be based upon a rather primitive and obsolete view of natural science methodology. 
A number of post-positivist perspectives emerged within the field of qualitative research from the post-War period up to the 1970s. Whereas positivist research aims to offer "objective" accounts of reality, post-positivist perspectives recognize the flawed nature of all methods, and therefore the impossibility of ever fully achieving this aim. The aim of authors such as GLASER and STRAUSS, CORBIN, MILES and HUBERMAN (and more generally, the grounded theory school) was to increase the "scientific" nature of social science research by developing as much as possible the equivalence to quantitative criteria of "good" research practice. They shared a common emphasis on attempts to formalize qualitative methods through the use of "quasi-statistics" and software packages. From the point of view of data analysis, the concern with the elimination of bias and the systematic nature of the analysis was further pursued. Similar to positivist perspectives, post-positivist methods involve a relatively unproblematic view of data storage and re-analysis, putting the central concern on the elimination of bias during the process of data collection. 
For the past three decades, there has been a multiplication of new perspectives that have been loosely grouped together under the label of "the interpretive turn in the social sciences" such as hermeneutics, ethnomethodology, symbolic interactionism, dramaturgical analysis, poststructuralism, and discourse theory. Interpretive approaches share a common emphasis on the analysis of constructions of meaning, of the ways people make sense of their everyday activities and surroundings. In contrast to positivist and post-positivist perspectives, subjectivity is seen as a crucial and positive component of research in interpretive approaches.2) Indeed, interpretive perspectives do not define social reality as an exterior object. The social world is seen instead as a subjectively lived construct. Interpretive perspectives consequently abandon claims to objectivity to emphasize instead the reflexive nature of the research process and the subjective nature of constructions of meaning, both by the research subjects and by the researcher. This emphasis on and indeed revalorization of subjectivity involves a critical view of "data" (not a term that most interpretive researchers like to use). Indeed, for interpretive researchers, data are not "given" observations of external social facts that are independent from the researcher. Rather, interpretive approaches recognize the constructed nature of "data," which are seen instead as the outcome of a reflexive research process wherein the "social problems" under investigation are themselves treated not as "given" but as socially constructed. Symbolic interactionists furthermore emphasize the interactive nature of data collection. As symbolic interactionists point out, data collection cannot be adequately considered as a passive extraction of information from participants by the researcher, but rather as a mutual construction of meaning during the data collection—for example, in interview situations. Consequently, a number of potential question marks arise as to the feasibility or indeed the desirability of secondary data analysis (and therefore of data storage), which the next section will present. 
As the previous section outlines, the different perspectives associated with the interpretive turn in qualitative research consider the social world not as a collection of external "facts," but as a subjectively experienced construct. Whereas objectivist and causal perspectives privilege the notion of external social facts, interpretive approaches are in contrast articulated around the notion of the social and cultural world as a milieu of meaning (HERMAN 1988, p.45): DILTHEY's Geisteswelt, HUSSERL's Lebenswelt, HEIDEGGER's Umwelt, SCHUTZ's everyday world or common-sense world, WITTGENSTEIN's form of life. From this perspective, following WEBER's view of social analysis, the goal of the social sciences lies in the interpretive understanding of the subjective meaning of social practices and of cultural artifacts, within a lifeworld that the researcher is embedded in. It follows that the study of social reality as an "external object" is a methodological impossibility. 
Hermeneutic versions of the interpretive turn have most systematically addressed methodological issues relevant to data collection, data storage and re-analysis. The term hermeneutics is generally used to refer to the interpretation of the meaning of cultural objects (texts, documents) and social practices. The different strands of hermeneutics share the idea that the subjective meaning of action and words is grasped through the operation of interpretive understanding, although there is a lack of agreement around the conceptualization of Verstehen (see, for e.g., DALLMAYR & McCARTHY 1977). In DILTHEY's earlier writings, similar to SCHLEIERMACHER's, the process of understanding is based on the method of Erlebnis, consisting of the attempt to reconstitute the mental states of others. In other words, Verstehen is thought of as the empathic re-living (Nacherleben) of the original life experience on the basis of our own subjective experience. This reconstitution is, however, doomed to remain incomplete. As GUSDORF (1988, p.235) points out, between the original life experience and its reconstitution, there is a fatal loss of authenticity. In later writings, DILTHEY came to see Verstehen as the location of a text within its objective framework of meaning, including cultural, historical and linguistic context—a view which is currently prevalent. 
Subsequent theorists, in particular HEIDEGGER's student GADAMER, conceptualized Verstehen in relation to the interpreter's historical context, as well as to that of the author and her original audience. For GADAMER, the point is not to interpret an action through the motivations of which the individual has subjective consciousness, but rather to concentrate the analysis on the form of life within which the action unfolds and acquires meaning. From this angle, there is a fundamental unity between subjective experience, language, and the historical horizon of meaning. Given that meaning is embedded in a specific historical and cultural context, the meaning of the object of research is irreducible to the cultural meanings that envelop the interpreter. From this viewpoint, hermeneutics is provoked by the encounter with incommensurable discourses. This idea is radicalized in WINCH's claim that all language games (forms of life) are unique (WINCH 1958). 
Hermeneutics shares with analytic philosophy an emphasis on the linguistic nature of subjectivity. HEIDEGGER, GADAMER, RICOEUR, as well as the WITTGENSTEIN of Philosophical Investigations emphasize first, that it is only through the intermediary of a conceptual framework, which is of a linguistic nature, that reality is apprehended. Second, it is only through language, through shared meanings, that social agents give meaning to their actions. It follows that the nature of meaningful action is fundamentally social (see DALLMAYR & McCARTHY 1977, p.7; GIDDENS 1976, pp.52ff; WINCH 1958). The social nature of meaningful action is stressed in the WITTGENSTEINian notion of language game. This concept emphasizes that human action is embedded in social conventions, which derive from common sense. Insofar as language games are embedded in concrete forms of life, WITTGENSTEIN stresses both the linguistic structure of subjectivity and its insertion in a concrete world. HEIDEGGER conceptualizes subjectivity similarly as the embeddedness in language and a concrete historic moment. HEIDEGGERian ontology considers "In-der-Welt-sein" (being-in-the-world) as constitutive of human existence. Being cannot be dissociated from being-in-the-world. This strong emphasis on the historicity of subjectivity is an important move away from previously dominant views on subjectivity as "pure" consciousness. As HEIDEGGER and GADAMER point out, subjectivity is embedded in a concrete, historical world. In contrast to HUSSERL for example, HEIDEGGER and GADAMER believe that subjectivity accesses truth and knowledge through, not despite, its anchorage in the world. 
GADAMER interprets the fundamental embeddedness in the world as an insertion in tradition. The individual belongs to tradition, and therefore acquires the traditional meanings that are transmitted. "That is why," he argues, "the prejudices of the individual, far more than his judgments, constitute the historical reality of his being" (GADAMER 1989, pp.276ff). In relation to truth and knowledge, traditional meanings are not a constraint upon understanding but, on the contrary, a precondition thereof. It is only because the interpreter has a certain pre-understanding of a text, based upon "constructive" prejudices, that she can develop an interpretation. GADAMER (1989, pp.298ff) therefore calls the prejudices by which we understand "true," as opposed to "false," ones by which we misunderstand. For prejudices to have a positive role, they have to be made explicit and transparent. In GADAMER's terminology, it is necessary to make explicit the "fore-meanings." Pre-understanding is one of the conditions of understanding meaning, and prejudices facilitate understanding. Within GADAMER's work there is thus a revaluation of prejudice as a condition of understanding. 
Such a position sharply contrasts with positivist conceptions of science. As pointed out earlier, positivist approaches consider prejudice as a barrier to be eliminated, preventing the observer from seeing reality as it is. According to GADAMER, such an elimination of one's own horizon of meaning is an ontological impossibility. Understanding is achieved instead through what GADAMER calls the fusion of horizons, that of the text and of the interpreter. On this point, GADAMER seems close to SCHLEIERMACHER. For the latter, interpreter and author are each embedded in a precise Lebensmoment. It is impossible to make abstraction from this specific historical and cultural context. The access to truth therefore necessarily takes the form of a dialogue between two epochs rather than the repetition of a historical life experience (see GUSDORF 1988, p.324). The model of the dialogue involves the notion of "listening to the truth of the other." For GADAMER, this involves a more egalitarian model of knowledge, in contrast with the attitude of domination towards the research object which he ascribes to positivism. 
Such a dialogue takes the shape of "situating oneself" within the tradition that is under scrutiny. However, this "resituating oneself" is different from the Dilthean concept of Erlebnis, or lived experience. The reproduction of the original life experience is impossible, GADAMER believes. When we place ourselves in a situation, we nevertheless maintain our personal historical horizon. Given that this personal grid of interpretation is a fundamental part of our existence, it is impossible to dissociate subjectivity from its horizon of meaning. "But into this other situation," as GADAMER (1989, p.305) writes, "we must bring, precisely, ourselves. Only this is the full meaning of 'transposing ourselves'." While DILTHEY's Verstehen is never complete because new connections within the original context of meaning can always be uncovered, GADAMER's classic reformulation of hermeneutics posits that interpretations can never be final because different meanings are constantly projected upon human expressions and action.3) 
The turn towards interpretation within qualitative research has a number of implications for the ways in which we think about the nature of social science "data." First, as we have seen above, interpretive perspectives problematize the notion that researchers can directly capture lived experience, and recognize instead its mediation by the text as well as the reflexive nature of the research process. As researchers we engage in "double hermeneutics" (to use GIDDENS's term)—that is, we construct interpretations of interpretations. We try to interpretively "read" the meaning of cultural texts by writing in turn our own texts. Second, interpretive perspectives problematize the identity and the role of the researcher in this process. It is recognized how issues of gender, class and race, as well as the more immediate contextual conditions of the data collection, shape the research process and the nature of the data. 
The most radical views proposed by postmodern strands of interpretive research, especially, develop these critical issues into a fundamental critique of data analysis and data collection. At the level of data analysis, postmodern authors tend to take the biographical metaphor of social science analysis as "writing tales of the field" further to argue that social scientific accounts are, primarily, autobiographies. In other words, not only do we write stories about the social world that we observe, but we write primarily stories about ourselves observing the social world. Research accounts thus come to be seen as historically situated author's memoirs rather than memories—raising questions about the usefulness of storage of such individualized tales. 
At the level of data collection, if we overemphasize the interactionist and contextual nature of data collection, secondary analysis of qualitative data would similarly seem pointless. The interpretive turn in the social sciences importantly leads us to recognize the cultural and historical situatedness of constructions of meaning by social agents, and the interpretations that researchers develop of them. However, this stress on historical situatedness of the research could potentially imply deep skepticism towards the usefulness or even the very possibility of re-analyzing (from within a different social, cultural or historical context) such primary data. 
And yet, there is no need—indeed it is in my view counterproductive—to be too fundamentalist in this matter. It is possible to further draw on other contributions of hermeneutics itself, in particular, on the notions of objectivation and of the hermeneutic circle, to argue both for the possibilities and for the fruitfulness of successive re-readings of cultural texts or textual data. As RICOEUR (1986) points out, research data are both culturally and historically situated, and the product of a process of objectivation. Once collected, any research material acquires a certain autonomy from its original context of production as well as from its original author—a detachment which makes re-analysis of such data feasible. Furthermore, as GADAMER points out, the understanding of the (cultural and social) text is dependent on its pre-comprehension. This pre-understanding is, in turn, determined by its insertion in tradition which is itself in constant transformation. As a result, the movement of the whole to the parts and vice versa is less a means for achieving truth than a structural, ontological condition of understanding. "The circle of whole and part is not dissolved in perfect understanding but, on the contrary, is most fully realized" (GADAMER 1989, p.293). It follows from GADAMER's version of the hermeneutic circle that interpretation is both conditioned by tradition, and never final. A text can be the object of various successive interpretations in different historical periods. In order to achieve understanding it is necessary to continuously return to one's own prejudices, which are in turn constantly being transformed. As GUSDORF (1988, p.203) puts it, understanding implies turning in circles or spirals of understanding. The hermeneutic concept of the interpretive circle suggests that interpretation is an endless process while postmodern perspectives similarly emphasize the multiplicity and infinity of interpretations. 
The research questions that are applied to social science data reflect the historical embeddedness of the researcher, as GADAMER importantly points out. However, following the principle of the hermeneutic circle, our individual as well as collective research questions change over time—as do our "social problems" and policy priorities. The types of concerns that a given society considers important, worthy of research as well as of policy interventions, are neither "given" nor already "out there." They are socially and politically constructed, and therefore variable. For example, issues of gender equality or ecological concerns are currently important topics of public debate, social science research and public policy, whereas previously they went unacknowledged. The storage of qualitative data would allow for later researchers to go back to earlier studies on, for example, urban poverty, to find that the same data could provide important insights into processes of social and political change around gender as well—a concern that the initial researchers may not have had, reflecting their own historical embeddedness. Such re-analysis would be of interest not only to social historians, but also to current policymakers for whom insights into processes of social change over the medium and long term is crucial. 
Finally, to the extent that data storage also allows access to the data by non-specialists, including the subjects of the research themselves, the storage of qualitative data could also be seen as potentially empowering, allowing for feedback and dialogue. In this respect, it could be seen as a citizenship issue—similarly, an important ethical concern which most qualitative researchers share. 
1) I would like to thank Randy SEEPERSAD for stylistic improvements; Max BERGMAN, Thomas EBERLE and an anonymous reviewer for helpful suggestions; the IEPI, University of Lausanne and the Swiss National Science Foundation (Grant 61-66003.01) for institutional support. Some sections of this text have appeared previously in MOTTIER (2001). <back>
2) Arguably and more fundamentally, any research method could be considered to have an interpretive component. This text, however, reserves the term interpretive for those perspectives that are conventionally associated with the interpretive turn. <back>
3) Postmodern literary theory argues similarly against determinate meaning, with the claim that every decoding is another encoding. <back>
Becker, Howard (1989). Tricks of the Trade. Studies in Symbolic Interaction, 10, 481-490.
Dallmayr, Fred & Thomas McCarthy (Eds.) (1977). Understanding and Social Inquiry. Notre Dame: University of Notre Dame Press.
Denzin, Norman & Yvonna Lincoln (Eds.) (1998). Introduction: Entering the Field of Qualitative Research. In Norman Denzin & Yvonna Lincoln (Eds.), The Landscape of Qualitative Research. Theories and Issues (pp.1-34). Thousand Oaks: Sage.
Gadamer, Hans-Georg (1989). Truth and Method (2nd ed.). London: Sheed and Ward.
Giddens, Anthony (1976). New Rules of Sociological Method: A Positive Critique of Interpretative Sociologies. New York: Basic Books.
Gusdorf, Georges (1988). Les origines de l'herméneutique. Paris: Payot.
Herman, Jacques (1988). Les langages de la sociologie. Paris: P.U.F.
Lévi-Strauss, Claude (1963). La "pensée sauvage" et le structuralisme. Paris: Esprit.
Mottier, Véronique (2001). Hermeneutics and Interpretive Validity: Limits to Interpretation. In Jörg Blasius, Joop Hox, Edith de Leeuw & Peter Schmidt (Eds.), Social Science Methodology in the New Millenium [CD-Rom]. Opladen: Leske and Budrich.
Rabinow, Paul & William Sullivan (Eds.) (1987). Interpretive Social Science: A Second Look (2nd ed.). Berkeley: University of California Press.
Ricoeur, Paul (1971). The Model of the Text. Social Research, 38, 529-55.
Winch, Peter (1958). The Idea of a Social Science and its Relation to Philosophy. London: Routledge.
Present position: Professeure boursiere (Swiss National Science Foundation Research Professor), IEPI, University of Lausanne & Fellow of Jesus College, Cambridge.
Major research areas: interpretative methodologies; social theory; gender and sexuality; eugenics.
Institut d'Etudes Politiques et Internationales
Université de Lausanne BFSH2
Lausanne 1015, Switzerland
Mottier, Véronique (2005). The Interpretive Turn: History, Memory, and Storage in Qualitative Research [21 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 6(2), Art. 33, http://nbn-resolving.de/urn:nbn:de:0114-fqs0502330.