Varieties of “Sociological Enlightenment”: Critical Arts-Based Inquiry versus German Reconstructive Social Research

Abstract

This article is a methodological reflection on recent developments in qualitative research. It discusses the methodology of critical arts-based qualitative inquiry (CAI).Since the 1980s and in postmodern, poststructural or postpositivist approaches CAI is seen as a renewal of qualitative research. Though, apart from special discourses in Cultural Studies, CAI seems to be ignored by German sociology. I will focus on the theoretical-methodological fundamentals of these approaches with respect to a politics of interpretation, and discuss these basics from the perspective of German qualitative-reconstructive research traditions. Although CAI can be a source of productive irritations, the differences suggest contrasting concepts of a “sociological enlightenment.”

Qualitative social research has lately been witnessing the emergence of a wealth of critical approaches that rely on arts-based processes for data collection, analysis, and presentation (Geimer, 2011). Cases in point are, without any claim to comprehensiveness, the research programs of evocative ethnography (Ellis, 1997), autoethnography (Ellis & Bochner, 2003), performative autoethnography (Spry, 2011), performance ethnography (Denzin, 2003; Hamera, 2011) as well as interpretive ethnography (Denzin, 1997), performative social science (Roberts, 2008), critical-creative methodologies (Horsfall & Titchen 2009), creative analytical practice ethnography (Richardson & St. Pierre, 2005), and critical arts-based inquiry (Finley, 2011; Gwyther & Possamai-Inesedy, 2009).

This emergence of novel perspectives is also a consequence of the achievements of qualitative approaches. In a context of growing mutual competition, qualitative approaches are under substantive pressure to innovate (Travers, 2009) and, therefore, tend to produce a host of new labels. Not all of them, however, rely on arts-based processes or are committed to the outlook of a critique of ideology. For example, Leon Anderson (2006) calls for an “analytic autoethnography” as different from “evocative autoethnography,” arguing that the former does not embrace the postmodern turn that is crucial for the latter: “the turn toward blurred genres of writing, a heightened self-reflexivity in ethnographic research, an increased focus on emotion in the social sciences, and the postmodern skepticism regarding generalization of knowledge claims” (Anderson, 2006, p. 373; for a critical comment, see Ellis & Bochner, 2006, pp. 435–436).

My concern in the following is exclusively with those cases of critical arts-based inquiry (CAI) that have emerged in the context of Anglo-American qualitative research, which differ substantially from current positions in Germany that have come up at about the same time, that is, since the 1980s. My purpose in contrasting these two lines is not to clarify the causes of their differences but to compare the underlying methodological principles. In doing so, differences will be brought to light—primarily concerning the CAI “politics of interpretation” (Denzin, 1992) and “paradox of immediacy” (Gurevitch, 2002)—which, while they are a source of productive irritations in German qualitative research, suggest fundamentally different modes of “sociological enlightenment” (Luhmann, 2005) that seem more or less incompatible and hard to overcome, if at all.

CAI Fundamentals: A Politics of Interpretation

CAI is essentially a response to the emergence of “postmodern informed interactionism” (Fontana, 2005, p. 242), or the “literary postmodern turn” (Denzin, 2010, p. 35), and can be most concisely summarized as a reversal of classical ethnography (Moser, 2006). CAI focuses not on other cultures and their customs but focuses especially on cultures within the researchers' own societies. The task is not to describe and classify cultures in an as objective and nonjudgmental way as possible, but rather to analyze them in their interaction with the researcher's subjectivity. Thus, in contrast to Bronislaw Malinowski (1967, 2010) who, in an effort to come to terms with the experiences made on his stay abroad, kept a journal during his study of Melanesian culture but chose not to integrate these experiences into his actual research report, practitioners of CAI are specifically interested in the process of grappling with the self when confronted with the other (in their own culture).

Individual practitioners of CAI may differ in many respects, but they all agree on one point, namely that the purpose of this “postmodern ethnography” (Reed-Danahay, 2002, p. 423) is not to describe and adequately understand social reality but to change and improve it (Hamera, 2011, pp. 318–319; Spry, 2001, p. 710). Consequently, within the CAI community, the quality of research is essentially assessed by its ability to intervene in and transform everyday action structures: “Useful works offer interpretations persons can use to change their everyday worlds” (Denzin, 2010, p. 49). Similarly, Stacey Holman Jones (2005) argues that the main concern of CAI is with “creating space for dialogue and debate that instigates and shapes social change” (p.763), while Ellis, Adams, and Bochner (2011) contend that this type of work enables “participants and readers to feel validated and/or better able to cope with or want to change their circumstances” (p. 27).

This attitude is essentially a response to the crisis of the ethnographic way of presenting reality (Winter, 2009). Following the “writing culture” debate (Clifford & Marcus, 1986) and its critique of the fact that in social science research reports authors tend to be omniscient and invisible, CAI contends that the researchers' standpoint is not a factor that is part of the setting and should be methodically controlled for during the inquiry but that the researchers' subjectivity as such should be included as an object of research (Anderson, 2006, p. 384). Moreover, since research is supposed to be inherently incapable of producing generalizable knowledge—“the only generalization is that there is no generalization” (Lincoln & Guba, 1985, p. 110)—the research process is invested with a political purpose, and a linkage between interpretation and politics is postulated (Denzin & Lincoln, 2011 p. 10–11): “Ethnography like art is always political” (Denzin, 2000, p. 403), or “all of research is political” (Finley, 2011, p. 437).

Given the assumption that there is no way to methodically control for an understanding of the other, specific requirements are laid out for the researcher's personality. Researchers need to open up to the unstable and situated nature of their making sense of and giving meaning to the material collected, and make themselves as vulnerable as they can possibly be in order to nevertheless allow for those unlikely moments of an understanding of the other to happen. Thus, Norman Denzin (2000) writes:

I seek … an existential ethnography, a vulnerable ethnography which shows us how to act morally, in solidarity, with passion, with dignity … This ethnography moves from my biography to the biographies of others, to those rare moments when our lives connect. (p. 402)

Since chances for an authentic understanding of the other to occur are confined to a few rare moments, these moments are in fact saddled with a moral and political obligation. Researchers must use these moments in a responsible way, that is, not treat the people studied as providers of information for scientific purposes (such as publications, careers, etc.) that are unlikely to concern the participants but make sure that they come out strengthened, and confirmed in their own agenda. In this sense, Carolyn Ellis and Arthur Bochner (2006) observe, “It needs the researcher to be vulnerable and intimate. Intimacy is a way of being, a mode of caring, and it shouldn't be used as a vehicle to produce distanced theorizing” (p. 433). This self-understanding entails a moral obligation to use the rare and intimate moments of understanding to primarily address situations of crisis or distress that are of vital importance for and shape the existence of the people studied (Canella & Lincoln, 2011).

Many proponents of CAI refer to “epiphanies” (Ellis et al., 2011, p. 6) or “epiphanic moments” (Denzin, 1992, p. 83, 2000, p. 402) in this respect—that is, to crises that significantly impact a person's biography and constitute a turning point in their lives. Often epiphanies represent “remembered moments that are perceived as being especially important … or existential crises that need to be addressed” (Ellis, Adams, & Bochner, 2010, p. 346). Such moments and crises are often tied to experiences of exclusion, discrimination, marginalization, and so forth that should be brought to light: “This ethnography attempts to better understand the conditions of oppression and commodification that operate in the culture, seeking to make these ways of the world more visible to others” (Denzin, 2000, p. 402). Thus, research is invested not only with a political but also (depending on its orientation) with a therapeutic function that, in a context of “epistemological intimacy” (Smith, 2005), is seen as the “healing from emotional scars of the past” (p. 53), as Chang (2008) put it. Accordingly, Pelias (2004, p. 1) argues that research should be driven by a “methodology of the heart.” CAI perspectives vary in this respect, with autoethnography stressing the therapeutic aspects of research and performance ethnographies being more strongly committed to interventions in terms of a politics of everyday life, or micropolitics (for a review of the various overlaps between perspectives, see Fontana, 2005). Fundamentally, however, all CAI variations subscribe to a self-understanding as an “advocational ethnography that operates from a compassionate and lionhearted will to usurp and resist injustice” (Jackson, 1993; Spry, 2001, p. 499).

The problem of representation is eluded, as Denzin (1997) points out, by adopting a political and moral stance and by redefining key research issues. Research no longer seeks to offer adequate reflections or constructions of reality or a detailed mapping of experiences but rather to “bypass the representational problem by invoking an epistemology of emotion, moving the reader to feel the feelings of the other” (Denzin, 1997, p. 228). Works of qualitative research, then, are not only expected to have a political and/or therapeutic impact on the self as well as on the persons studied but are expected, in addition, to rely on arts-based performative practices (such as performances, stories, poems, plays) to present their findings. The purpose is to enable readers and/or audiences to empathize with those studied and to understand their existential distress and crises—they have to “be moved emotionally and critically. Such movement does not occur without literary craft” (Spry, 2001, p. 714). In their work “Autoethnography,” Ellis and Bochner (2006) observe very personal experiences and represent “struggle, passion, embodied life … Autoethnography wants the reader to care, to feel, to empathize, and to do something, to act” (p. 433). As a consequence, the perspective of the impersonal observer is to be abandoned in favor of “the embrace of intimate involvement, engagement” (Ellis and Bochner, p. 433) in order to convince research audiences of the need for intervention and change.

Arts-based ways of presenting research findings by personal stories, so-called “Mysteries” (Denzin, 2010, p. 58; Ulmer, 1989), or by plays, poems, or performances are expected to go beyond the mere reproduction of the words and attitudes of those studied. Rather, they are supposed to capture these words and attitudes and subject them to a montage or bricolage (Denzin, 2006, p. 423; Denzin & Lincoln, 2011) that reworks and transforms them in ways that make them touch and move the audience:

This social science inserts itself into the world in an empowering way. It uses the words and stories that individuals tell to fashion performance texts that imagine new worlds, worlds where humans can become who they wish to be, free of prejudice, repression, and discrimination. (Denzin, 2003, p. 105)

In doing so, the boundaries between the social sciences and the arts are deliberately blurred (Horsfall & Titchen, 2009, p. 151), and their respective modes of representation are made to converge: “The mode of storytelling is akin to the novel or biography and thus fractures the boundaries that normally separate social science from literature” (Ellis & Bochner, 2003, p. 217). In this sense, arts-based practices of data collection and analysis are “a mode of inquiry and a methodology for social activism” (Finley, 2011, p. 436).

A seminal work in this respect is Clifford's (1981) “ethnographic surrealism” that “attacks the familiar, provoking the irruption of otherness—the unexpected” (p. 562). Drawing on the ideas of the literary and artistic avant-garde of 20th-century modernism, ethnography that subscribes to this line of thought seeks to produce “action that incessantly insinuates, interrupts, interrogates, antagonizes, and decenters powerful master discourses” (Conquergood, 1995, p. 138). Hamera (2011) conceives of this utopian aspect of CAI as hope placed in the power of poesis—that it “will productively intervene in our understanding of the world” (p. 327).

Assessment of CAI accomplishments relies on criteria that are pragmatic and address its artistic potential to affect participants' emotions and move them to political action (e.g., Denzin, 2010, p. 49; Jackson, 1993; Spry, 2001). Thus, in order to be considered valid, texts need to be “lifelike, believable and possible” (Ellis & Bochner, 2003, p. 229). But, because artistic language is not immune to such poetical clichés as surrealists like Breton sought to explore in their écriture automatique, works of critical and artistic research should take care to avoid them. Accordingly, a key aspect of reliability is the communicative validation that can be obtained by having participants read their own texts and discuss them with the researcher. This also serves to analyze the personal relationship between the researcher and the people studied (Lincoln, 1995, p. 283), and if the latter feel that they are adequately represented in the texts, these may be considered to be reliable. Conditions for generalizability—in terms of an artistic effectiveness—are fulfilled when readers feel that a text reflects their own experiences or allows them to intensely relive the experiences of others.

One might question the rejection of traditional generalization by drawing on theoretical saturation in terms of the grounded theory or on “theoretical generalization” (Smaling, 2003, p.6). That would imply that the theory is formed intrinsically during the research process. But, this is not true for the methodological framework that facilitates the research process in the first place. Thus, as I will discuss later, theoretical concepts shaping methodological frameworks (like habitus, biography or discourse) ineluctably structure research processes. Although It is for that reason that some German qualitative research traditions refer to theory as methodology (see below). Critical research theory also concentrates on “seeing what frames our seeing” (Lather, 1993, p. 675). Nevertheless, for Patti Lather, who already demanded a “catalytic validity” of research in the sense of CAI in 1986 (Lather, 1986, p. 67), theory has the same function as art in critical, arts-based approaches: Both allow for “resist[ing] the hold of the real … implode[ing] controlling codes … generat[ing] new locally determined forms of understanding … bring[ing] ethics and epistemology together” (Lather, 1993, 685–686). Then, theory (like art) is used to produce a “disciplined messiness” (Lather, 2010, p.10), which refers to social complexity in texts that hold “possibilities for surprise” (p. 809), as Mirka Koro-Ljungberg (2012) puts it. Lather uses Derrida and Foucault (Lather, 2006) as well as Baudrillard (see also Koro-Ljungberg, 2013), and Deleuze (see also Brkich & Barko, 2012) for theorizing as a form of art in order to overcome a “correspondence model of truth” (Lather, 1993, p. 675; for a critical stance on the more or less creative use of European philosophy see Grossberg, 2014).

In such a “post-qualitative research” (Lather & St. Pierre, 2013) experimental writing (see also Richardson & St. Pierre, 2005) is favored. In contrast to that, German qualitative research developed approaches that build on a “correspondence model of methodology,” stressing on a form of “disciplined inquiry.” Theories, therefore, openly guide (via the shaping of methodological principles) research processes in favor of second-order observations that abstain from intervention but offer other options of critique and a different form of enlightenment via the representation of first-order observations.

The Fundamentals of German Reconstructive Social Research: First-Order and Second-Order Constructions

Turning to the more recent traditions of German qualitative research, my concern in the following is more specifically with “elaborated approaches” (p. 197) as defined by Reichertz (2007) and represented, for instance, by conversation analysis (Konversationsanalyse), objective hermeneutics (Objektive Hermeneutik), hermeneutic sociology of knowledge (Hermeneutische Wissenssoziologie), biographical research (Biografieforschung), or the documentary method (Dokumentarische Methode), all of which are understood—again following Reichertz (2007) as well as the definitions given in many textbooks (e.g., Bohnsack, 2008; Flick 2009; Przyborski & Wohlrab-Sahr, 2010, among others)—to be distinct from content analysis and qualitative ad hoc processes. The defining feature of elaborated approaches, and the feature that allows us to differentiate between them, is their reliance on action theory premises for working out the methodological fundamentals that underlie their processes of data collection and analysis. Hermeneutic sociology of knowledge (Flick, 2009; Hitzler, Reichertz, & Schroer, 1999), for instance, builds on Alfred Schutz's social phenomenology. The documentary method (Bohnsack, 2014) is rooted in Mannheim's sociology of knowledge. Last, conversational analysis (Bergmann, 2000) draws on Garfinkel's ethnomethodology. This anchoring in fundamental theories is accompanied by a number of basic assumptions that are described, for instance, as basic vocabularies (Basisvokabulare, Reckwitz, 2004, p. 4), guiding assumptions of observation (beobachtungsleitende Annahmen, Kalthoff, 2008, p. 12), or formal-sociological or meta-theoretical categories (formal-soziologische or metatheoretische Kategorien, Bohnsack, 2008, p. 15). These assumptions defy empirical verification. Rather, they are what make the research process possible in the first place.

Researchers who draw on objective hermeneutics (OH), the documentary method (DM), hermeneutical sociology of knowledge (HSK), or conversational analysis (CA) to analyze protocols of social actions will assume, in terms of theoretical fundamentals, that there are interpretive patterns (OH), orientation patterns (DM), interpretations and typifications (HSK), or conversational practices (CA) and that these patterns structure everyday action. As a consequence, the established postulate that qualitative research needs to adapt to its objects has only limited validity. Instead, ontology is not only understood as what “is presented [emphasis added] as the reality of the social world” (Hollis, 1995, p. 22) but as a means for actors to produce the social world. In the context of phenomenological sociology, this was already pointed out by sociologist Alfred Schutz (1962) who distinguished between “constructs of the second degree” (p. 6) from such of the first degree, that is, between the (first order) constructions proposed by everyday actors and the (second order) constructions proposed by researchers as a result of their reworking of these first-order constructions (Flick, 2009, p. 75).

Given that elaborated approaches rely on (action theory-driven) reconstructions of first-order constructions, they may be defined as reconstructive approaches (Bohnsack, 2014, p. 220). ¹ How qualitative research draws on those first-order constructions, and which assumptions are made concerning the practical knowledge and skills of everyday actors, defines the (meta-theoretical) difference between elaborated, reconstructive approaches.

Rather than simple constructions proposed by the researchers, the empirical points of reference of elaborated, reconstructive approaches are, thus, understood to be reconstructions of the constructions that are produced by everyday actors. They are conceptualized as “real-life guidelines” (Bergmann, 2000, p. 533) that are supposed to be effective in everyday life. In this sense, the methodologies of reconstructive research are grounded in ontological assumptions as to how people orient themselves in the social world. These assumptions vary depending on the main differences between the above-mentioned (meta-) theories from which they are derived. As Maxwell (2012) notes, social science research programs share an (often tacit) realism (Beetz, 2010, p. 13) that (implicitly) legitimates the research activity of reconstructing structures of meaning: “Meaning and culture are real… . mental properties and processes are just as real as physical ones, also they are understood using a different language and conceptual framework” (Maxwell, 2012, p. 15).

In opposition to “postpositivist, constructivist, and postmodern” (Maxwell & Mittapalli, 2010, p. 148) as well as critical and arts-based approaches, Maxwell advocates a perspective of critical realism (see also Danermark, Ekström, Jakobsen, Karlsson, 2002), which, in spite of a constructivist epistemology, acknowledges the need for an ontological positing and assumes that there are different valid perspectives on the world. However, it holds that both the “people we study as well as ourselves are part of the world that we want to understand, and that our understanding of these perspectives can be more or less correct” (Maxwell & Mittapalli, 2010, p. 157). In this respect, the elaborated approaches of qualitative research in Germany, too, share a common feature because they all involve methods of a controlled understanding of the other and use these methods to intersubjectively verify, depending on the specific paradigms that determine the aspects emphasized by the respective methodologies, the reliability of the reconstructions, as proposed by the researchers, of the perspectives of those studied (Bohnsack, 2005; Reichertz, 2007). Moreover, different methodologies tend to favor very similar forms of controlling for the observer's position—such as bracketing the validity aspect (“Einklammerung des Geltungscharakters,” Mannheim), suspending natural attitudes (Schutz), or practicing ethnomethodological indifference (Garfinkel). The intent is to make sure that research activities are not informed by value-based attitudes or any categories other than those in agreement with meta-theoretical assumptions.

Different strategies of methodologically controlling for the observer's position are also at the bottom of the tendency shown by schools of reconstructive research in Germany to conceive of themselves as distinct traditions and to seal themselves off from each other. Methodologies and methods are used as a vehicle to reproduce and relegitimate the underlying meta-theoretical principles. So, there is good reason to raise the issue of how the fundamentals of qualitative research can be systematically broadened, for “[a] methodological relativization of scientific background assumptions … does not necessarily mean that there is theoretical arbitrariness. This relativization is especially useful for counteracting fundamentalist positions” (Beetz, 2010, pp. 36–37). As Bohnsack (2008, pp. 27ff and 187ff) points out, the question has to be answered primarily through the practice of interpretation because researchers may get to feel that the material indeed suggests novel categories and basic concepts.

A Critical Comparison of Methodological Fundamentals

Typical Points of Criticism of CAI

CAI does not differentiate between epistemological and ontological fundamentals. Rather, methodology and methods (epistemology) and core categories or basic vocabularies (ontology) are intertwined in their own works and separated in the deconstruction of other works. Therefore, “showing how the ‘facts’ and ‘truths’ that had been ‘found’ were inextricably linked to the vocabulary and paradigms employed by the researchers” (Ellis et al., 2010, p. 345) is considered a critical deconstruction of the history of qualitative research. In the German reconstructive tradition of qualitative research, in contrast, this is part of the research routine (Bohnsack, 2008, p. 29, see also Willis, 1980). Far from qualifying as an adequate point of criticism of social research, it is understood to be the crucial basis for interpretations to be intersubjectively verified. In other words, a dialogue about whether or not interpretations of a particular set of data are correct is possible precisely because methodologies differ in their respective aspect-orientations and perspectives. From the perspective of current programs of reconstructive research in the German tradition, critical arts-based approaches typically lack this dimension of verifiability. From the perspective of CAI, such verifiability means to systematically narrow, close, and control (discipline) the potential meaning of texts and actions.

Unsurprisingly, however, no contribution from German research was needed for a critique of the provocative CAI mode of doing research to arise. On the contrary, discussions in the US seem to be much more entrenched and acrimonious and are handled in so rigid a way that in some journals, for instance, the editors make a point of publicly stating their fundamental disapproval of critical arts-based inquiry (e.g., Qualitative Health Research ² ). Abrasive polemics most notably target the self-referential nature of CAI, denouncing it as a “postmodern but asocial theory of knowledge that argues the impossibility of knowing anything beyond the self” (Gans, 1999, p. 542; see also Geertz, 1988). CAI is further criticized for reducing the political to the personal and for losing sight, in doing so, of macro-sociological issues (Atkinson, 2004, p. 109). Furthermore, in the reception of postmodern or post-structuralist theories, there seems to be a tendency to trivialize the issues at stake. Stäheli (2000) already cautioned against a certain kind of misunderstanding, that is, that the assumption that “the concept of power … always includes, as an abstract … necessity, the possibility of resistance, in the sense, for instance, that any practice of power will imply its own negation” (p. 52) that only needs to be given full reign. Similarly, Reckwitz (2001) criticized certain variations of post-structuralism for encouraging “the dramatizing of the permanent changeability of identities” (p. 34). In this sense, critical arts-based approaches ignore the inertia of transformation processes. In terms of an argument developed by Bourdieu and Wacquant (1996, p. 130), they obliterate the hysteresis of a habitus and instead focus on the possibility of negation and the emergence of novel forms of “subjectivation” that can be activated once the appropriate conditions for interaction are created.

But while the critical comments cited so far may well be justified, they all fail to go into the details of the basic paradox of CAI reliance on personal experiences, that is the “paradox of immediacy” (p. 405) as described by Gurevitch (2002)—which, among other things, allows researchers to deduce specific standards and quality factors that, so far, defy concrete criticism. The paradox, here, is that while artistic practices are deployed to reproduce experiences in a way that enables readers to immediately and intensely relive them, the artistic mode of reproduction suggests that there is something that typically evades experience—something that will only emerge in the future, as a result of the arts-based critical reflection of this very experience. In terms of “poetic sociology” (Gurevitch, 2002, p. 404), this is a concern that is midway “between the performative that highlights voice, autobiography, and play and a prophetic, critical or deconstructive mode, which is sensitive to the edges, to impassability and impossibility to voice, to tell to figure” (Gurevitch, 2002, p. 405). This “paradox of immediacy” has indeed been a source of fruitful irritations in qualitative social research in Germany—especially in the realm of Cultural Studies (Winter, 2009)—but, at the same time, alerts to the fact that the two lines of qualitative research rely on different, and partly incompatible, models of “sociological enlightenment” (soziologische Aufklärung, Luhmann, 2005; see also Baecker 1999).

Following Luhmann, I will in the present context use the concept of enlightenment in its broader—German—sense. In German, it may mean enlightenment when referring to a specific historical era, but also elucidation or clarification when applied to errors or general facts or even, as a verb (to resolve) to a criminal case. However, this broader use implies that even postmodern, anti-enlightenment projects (Denzin, 2000, p. 407) can be understood as a continuation of the historical project of enlightenment (d'Entrèves, 1999; Foucault, 1984; Gebhardt, Meißner, & Schröter, 2006), at least if one steers clear of the dilemma “of choosing between alternatives: either the defense of the principle of reason, progress, freedom etc., or their rejection as insufficient, the highlighting of their repressive character etc.” (Dolar, 1995, p. 264).

Applied Enlightenment Versus Detached Enlightenment: Artistically Processualized Critique of Ideology Versus Systematic Communication of Second-Order Observations

CAI fundamentals are akin to those underlying a pragmatic sociology of criticism as proposed by Boltanski (2010). Boltanski denounces the distinction between prereflexive and deliberate action and, thus, rejects the tenet that the theories participants hold about themselves are fundamentally unreliable because of their supposed quasi-inability to access their own motivations and orientations. Bourdieu, it will be recalled, describes this as the nontransparency of the self's habitus and, as a consequence, cautions social science against the “illusion of immediate knowledge” (Bourdieu, Chamboredon, & Passeron, 1991, p. 13). Sociology as a craft “always presupposes a break with the real and with the configurations that it offers to perception” (Bourdieu et al., 1991, p. 13). A former disciple of Bourdieu, Boltanski has since rejected this principle. Drawing on pragmatism and ethnomethodology, Boltanski (2010, p. 41) instead centers on the reflexivity that is inherent in the practices of everyday criticism. CAI also centers on the potential of everyday criticism but differs from Boltanski in assuming that in situations of crisis (epiphanies), everyday actors are indeed able to operate the break with common sense (Turner, 1989). These situations, therefore, are particularly well suited for revealing potentialities of criticism, and arts-based methods of inquiry and presentation are employed as a means to keep this critical potential alive. This, in my view, is a key CAI accomplishment: seeking to capture situations and moments (in the researchers' lives and/or the lives of others) where the actors themselves operate the break with common sense and get to critically address the foundations as well as the contingency and the conditions of their subjectivity. The German tradition of qualitative-reconstructive research would therefore be well advised not to leave the exploration of critical life events (Filipp, 2010) more or less to psychology, but to intensify its efforts to detect, in the crises experienced by those studied, the reflexive potential of everyday actors, and to see it as a force that is capable of stimulating social change.

In contrast, there seems to be no way of overcoming the antagonism between the German reconstructive approaches and the CAI commitment to micro-political intervention in the biographies and interpretational structures of both the persons studied and the audience, with the goal of improving their condition and/or boosting their reflexive potential. In terms of Luhmann's (2005) broader concept of enlightenment, this basic aspiration of CIA can be described as a form of “practical” or “applied enlightenment” (angewandte Aufklärung) that is oriented to immediate intervention in everyday practice. In the German tradition, in comparison, the critical potential of established research programs is framed in terms of a more theoretical and amoral enlightenment, or “detached enlightenment” ³ (abgeklärte Aufklärung, Luhmann, 2005, p. 85), which abstains from intervention and the goal of producing a different and a better world. Still, research might use the results obtained by “detached enlightenment,” as I will show in the following, to disrupt processes of cultural self-affirmation by revealing latencies (Baecker 1999; Gebhardt et al., 2006, p. 279). ⁴

Luhmann (2005 developed the position of detached enlightenment primarily in his critical analysis of the Frankfurt School (p. 85) which, not unlike CAI (and although it did not offer a specific approach for qualitative research in Germany), was also committed to a politics of interpretation: “Epistemology is in itself ethics, and ethics is epistemology” (Lather, 1993; Marcuse, 1964, p. 125). Luhmann (1990, p. 510) most notably rejects the hierarchization based on superior insight” that is associated with this stance: Critical theory had “adopted attitudes that suggested superior insight. It acted as a competing describer with impeccable moral impulses and superior perspicacity” (Luhmann, 1991, p. 148). Its descriptions, framed in terms of a critique of ideology, did not go beyond the level of first-order constructions as defined by Schutz (1962, p. 6) and competed with other forms of everyday self-interpretation and world-ordering: “Their perspective was that of a first-order observer of the world. They offered a competing description of society” (Luhmann, 1991, p. 148).

Luhmann (1990 challenges this attitude by putting forward his concept of a second-order observer (p. 86): Everyday actors are first-order observers who are not aware of the distinctions that guide their observation; scientific observers are second-order observers who observe the distinctions that underlie their observation. What is crucial here is that observing the distinctions that guide observation typically and necessarily supposes an attitude that abstains from criticism, or an amoral attitude (Gebhardt et al. 2006, p. 279; Luhmann, 1990, p. 87)—which is precisely why programs of qualitative-reconstructive research in the German tradition are committed to the suspension of natural attitudes as defined by Schutz, the bracketing of the validity aspect according to Mannheim, or the concept of ethnomethodological indifference.

CAI, in contrast, explicitly states its intention to confine itself to the first-order level. Accordingly, Denzin (2010) rejects “second-order concepts” in favor of “first-order textuality” that is concerned with “flesh-and-blood human beings [who are] talking to each other” (p. 92). Within this conceptual framework, researchers and everyday actors meet on an equal footing while inquiry results in more or less novel observations whose contingency, however, cannot be controlled for and whose genesis cannot be verified. Gebhardt et al. (2006) note in this respect: “When sociology (or theory in general) switches to second-order observations, that is, to observing the critics engaged in criticizing, it becomes evident that the normative standards are socially and historically situated, that they are contingent positings” (p. 279).

It could be maintained that self-reference and observation of the self's world of emotions may also be a second-order observational attitude. After all, even Luhmann (1991) points out that an “observation of observers … can be undertaken as either self-observation or observation of the other” (p. 149). If, however, the contingent nature of the researcher's observation is invariably referred to as a guarantee of the authoritative nature of criticism, or as a resource for a validation that is supposed to apply both when the self is concerned in person and when it is not, it cannot possibly be a systematically controlled second-order observation. In the final analysis, acknowledging and disclosing the fact that the self is concerned (or not concerned) results in criticism being immunized to criticism, for the artistic forms of presentation preclude any discussion of the how of observation; a result which, far from being a side-effect, is fully intended: “As the emphasis on performance implies, there is little attempt to enter the minds of other people; to argue about what a performance means for the other person” (Denzin, 2000, p. 404). Thus, the responses of CAI audiences are as strictly tied down to a first-order observational perspective as CAI researchers' actions are. The presentation of experiences (of the self and/or the other) by artistic means and with maximum intensity and the goal of voicing criticism is understood to be successful if audiences are impressed or stirred, or less successful if they are not. Because no communication about the how of observation is possible, audiences are confined to the alternative of approval or refusal. Seen this way, CAI is not only rather less likely to stimulate social dialogue but, with respect to the special (and of course restricted) premise that observational positions must be distinguishable, the CAI issue of applied enlightenment is also unscholarly.

Now, does this imply that there is no critical potential at all in second-order observations? In terms of Luhmann's (2005) “detached enlightenment,” this potential resides in the fact that second-order observations, while they are critically aware of their basic assumptions and of the provisional nature of the position adopted (the observation might have been conducted in a different way), are nevertheless able to highlight the contingencies and latencies of first-order observations. In doing so, processes of cultural self-affirmation and their pervasive “no alternatives” argument (without, however, evoking the possibility of a better world, and provided that findings are indeed perceived) can be disrupted. This critical potential is specifically found in those perspectives of qualitative inquiry that allow for a reconstruction of latent and implicit structures of meaning, for instance objective hermeneutics or the documentary method. I will illustrate this using the example of the latter.

The documentary method explicitly draws on a system-theory concept of the observer (Bohnsack, 2007, 2008, 2014). Accordingly, second-order observations focus on the how (Luhmann, 1990) of social actions and interactions. In the performative structure of action (Bohnsack, 2007), meanings are documented that the actors themselves cannot reflexively access: “The performative structure, or performativity, is the process structure that can be reconstructed by documentary interpretation, it is the modus operandi, or the habitus as defined by Bourdieu” (p. 204). This observer's position implies that everyday actors always know more than they actually realize. ⁵ The goal of the documentary method, then, is to reconstruct this implicit action-guiding knowledge (for instance, by analyzing everyday conversations, group discussions, interviews, paintings, or video-graphical material, etc., Bohnsack, 2008, 2010a, 2010b, 2014; Bohnsack, Pfaff, & Weller, 2010; Nohl, 2012; Przyborski & Slunecko, 2009; Przyborski & Wohlrab-Sahr, 2010).

This conceptual framework, first, enables researchers to identify contradictions, unnoticed on the subjective level, between the actors' subjective self-perception and the implicit knowledge structures documented (Geimer, 2012, 2014), or between the actors' reflexive and legitimating reference to common sense and their implicit knowledge structures (Bohnsack, 2012). Cases in point are couples that present themselves as gender-neutral and equal in an interview while observations of everyday practices or detailed accounts of everyday life suggest the contrary. Bringing to light these contradictions between self-presentation and habitus is what defines the critical potential of the documentary method and its commitment to the second-order observational attitude described by Luhmann (1990). Second, implicit knowledge structures can also be made accessible to the actors themselves, as documentary evaluation research attempts to do (Bohnsack, 2010c; Lamprecht, 2012). Qualitative-reconstructive research, thus too, has an interventional dimension. However, this dimension is actuated only when participants ask questions that elicit this kind of response. It is not presupposed in terms of a critique of ideology which, furthermore, tends to label participants a priori as “optimizable,” or “in need of help,” and their everyday practice per se as “improvable.”

Conclusion

Notwithstanding all the criticism that can indeed be leveled at CAI for its “failure” to differentiate between research, the arts, and politics, its insistence that everyday actors—not only researchers—have a potential of critical reflection and, thus, of operating the break with common sense—”dismantling the notion that the researcher is the only knower and expert on the lives and experiences of the participants” (Dixson, Chapman, & Hill, 2005, p. 17)—seems a promising opening. In this respect, in particular, reconstructive research programs in the German tradition that rely on the potential of second-order constructions to reconstruct first-order constructions could productively build on CAI. Arts-based processes and esthetic practices for data collection thus could be methodologically reflected and used in methodically controlled for ways that make it easier for participants to articulate experiences of crises and radical change, as compared to traditional research settings.

Furthermore, “detached enlightenment” (Luhmann, 2005), too, will reveal contingencies that have a potential to disrupt processes of self-affirmation (albeit with more or less sustainable results, and depending on the public visibility of research findings). This is true particularly for approaches (such as objective hermeneutics and the documentary method) that rely on second-order observations in order to reconstruct latent and implicit knowledge structures. These approaches observe how everyday actors observe and, therefore, are a variation of constructions of the second degree as defined by Schutz, focusing on the performative aspects of producing social reality (for a more detailed discussion see Bohnsack, 2014, p. 220). These reconstructive research programs have the critical potential to reflect, for instance, contradictions between action-guiding knowledge structures and certain forms of self-presentation, or to use evaluation studies to detect blind spots and bring them to the attention of those studied.

In contrast to CAI, sociological “detached enlightenment” (Luhmann, 2005) attempts to keep the systems of politics and of research apart. This is done precisely because researchers are well aware that processes of mutual influence are likely to occur, and hope to be able to decrease the chances for this to happen by framing research as an environment where second-order observations (whose methodological limitations and fundamentals have to be made clear) can be discussed in as unprejudiced a way as possible. As for practical and concrete ways of implementing “detached enlightenment,” a more in-depth discussion of the potentialities of public sociology (Burawoy, 2005; Hitzler, 2012; Scheffer & Schmidt, 2013) is needed. This discussion should be committed to “taking knowledge back to those from whom it came, making public issues out of private troubles, and thus regenerating sociology's moral fiber” (Burawoy, 2005, p. 5) but should not, for all that, depart from the established standards of professional sociology and the methodologies of second-order observation. Another definite potential for criticism is inherent in the intra- and extra-scientific communication of third-order observations. On this third-order level, the distinctions relied on by the research system or by specific studies, as well as the blind spots or the voices that are—or are not—included, and how this is achieved, can be systematically identified. In this respect, Burawoy (2005) evokes a critical sociology that is primarily self-critical and, thus, bound to “examine the foundations—both the explicit and the implicit, both normative and descriptive—of the research programs of professional sociology” (p. 10)

Finally, I would like to emphasize that my purpose in this article is by no means to dismiss the CAI mode of personally committed practical and applied enlightenment from the realm of research. Especially in the research community, differences should be appreciated as an asset that has a high intrinsic value because critical dialogue allows for standpoints to be specified, and multiple perspectives help to gain deeper insight into the processes that make up the social construction of reality.