1 Religion and Science in Discursive Perspective
This book engages in a debate that cuts deep into contemporary European values and self-images. How can we adequately describe the place of religion in European societies after the long process of what is usually referred to as ‘secularization’? In the following chapters, I want to convince you that discursive analyses are particularly suitable in finding answers to this question. They also provide a solution to critical issues in the study of religion more generally. Because the academic study of religion is itself part of the process of what I will describe as scientification of religion, it is helpful to look at the scholarly construction of religion and its discursive entanglements with other cultural systems and academic disciplines; such an approach also contributes to the process of self-reflection that our discipline needs at the beginning of the twenty-first century.
For some time now, the academic study of religion has experienced fundamental challenges. Although established as an independent discipline at European universities more than one hundred years ago, the academic study of religion is still wrestling with severe problems of identity and legitimization. The reasons for this challenging situation are partly related to developments in the academic landscape that have influenced many ‘small disciplines’ in the second half of the twentieth century; but they are also related to the fact that religion has played a very special role in the scientific, political, and cultural debates of the past two hundred and fifty years. Unlike ‘law,’ ‘health,’ ‘economy,’ and other concepts that appear more innocent at first glance, the concept of ‘religion’ is charged with difficulties that have thrown its study into contestation. The study of religion is particularly challenged in regard to its link to theology and thus to confessional or experiential approaches to religion, its link to colonial agendas that imposed a Eurocentric view on ‘non-Western’ cultures, as well as the tendencies in influential parts of the discipline to essentialize religion as something sui generis. One of the most important theoretical and methodological questions today is whether the discipline can respond to these fundamental challenges in a way that takes these critiques seriously and is able to transform the study of religion into an academic discipline that operates within a rigorous and self-reflective interpretational framework. Given the ubiquitous presence of religion in the global cultural worlds of the twenty-first century, there should be no doubt that we need experts who are trained to scrutinize the history and present appearance of religion in a sound academic way.1
I argue that the notion of ‘discourse’ is of particular value if we are to establish such a self-reflective academic discipline. Although basic considerations about a discursive study of religion were made already in the 1980s (Kippenberg 1983; Lincoln 1989; see also Kippenberg 1992 and Lincoln 2005 [1996]), these suggestions have not been picked up in a more general way as an attempt to build a serious referential framework for a (self-)critical study of religion. Recently, we have witnessed a rich discussion in neighboring disciplines, mainly in sociology and historiography, about the usefulness of discursive analysis. German and French scholars, in particular, have readdressed the theories of discourse, which were developed by Michel Foucault and others a generation earlier, and made them fruitful for the study of cultural phenomena in the twenty-first century. The academic study of religion has only sporadically taken notice of these new approaches to the study of discourse that break down the boundaries between academic disciplines—and even between the humanities and the natural sciences—in a most productive and promising way. More recent publications show that there is a growing interest in discourse theory, including its application to the study of religion, though most contributions still base themselves on linguistic and textual analyses of discourse.2
Systematically applying discursive approaches is an important contribution to a discussion that is specific to the academic study of religion. It helps to resolve some of the conceptual problems in the field, as mentioned above, and tries to provide a coherent analytical framework for an academic study of religion that is capable of countering the theoretical challenges the discipline has to face.
Two directions of scholarly thinking, both reaching back to the first half of the twentieth century, are of special—and underestimated—importance for a new understanding of discursive approaches. For one, the contributions from the sociology of knowledge should be more seriously incorporated in our theoretical framework; furthermore, the historical analysis of discourse is something that scholars of religion need if they want to retain their strong basis in historical research. In the following, I will introduce these lines of thought that are closely related to discussions in sociology and historiography. I will then clarify the most important terms that constitute a discursive study of religion and will suggest a definition of religious discourse as a clearly demarcated object of study. Discursive study of religion provides a research perspective rather than a single method to study religion. Nevertheless, this perspective has implications for the concrete scholarly work of designing a research project, putting together a corpus of data, and interpreting this data with the use of appropriate methods. I will explain this research strategy with reference to the themes that the present book engages.
The Discursive Construction of Knowledge
One problem of the notion of ‘discourse’ is the fact that the term is used in many, and often conflicting, ways. It has further added to the confusion that many scholars do not clearly define what they mean when they use the term ‘discourse.’ This is not the place to provide an overview of the many different usages of the term since the nineteenth century (for such an overview see Keller 2011a, 97–177; Keller 2011b, 13–58; see also Landwehr 2009, 60–90). Rather, I want to highlight the crucial contributions that come from the sociology of knowledge and from historiographical approaches. Although they acknowledge the importance of language in the study of discourse, both approaches move beyond classical linguistic analysis (in the field of social linguistics ‘discourse’ refers to the more minute and specific patterns of speech in the everyday sense) and include the materiality of discursive structures. This is of particular importance for the study of religion.
Since the 1960s, ‘knowledge’ has been an important dimension in sociological and discursive theory. This is true for the influential contributions of Peter L. Berger, Thomas Luckmann, and Alfred Schütz on the social construction of reality and knowledge (particularly Berger and Luckmann 1966; Schütz and Luckmann 1979–1984; applied to the study of religion and media in the overview of Krüger 2012, 11–162), but also for Michel Foucault’s interest in the structures that produce shared knowledge in a given societal and historical situation. Foucault put particular emphasis on the power-structures that distinguish approved from non-approved knowledge (Foucault 1980), a focus that puts his work in a Marxist line of interpretation that is still visible in some recent, post-socialist or post-Marxist approaches to discourse theory, such as those of Ernesto Laclau and Chantal Mouffe, who emphasize the primacy of politics and the importance to solve urgent problems of our time. Jacob Torfing builds on this approach when he defines a discourse “as a relational totality of signifying sequences that determine the identity of the social elements, but never succeed in totalizing and exhausting the play of meaning” (1999, 87). Sociologists of knowledge still include the dimension of power and the importance of politics in their analysis, but not necessarily as the main and determining dimension of discursive practice or as expression of the need to radically reform plural democracies (in the way Laclau and Mouffe would have it; see Jørgensen and Phillips 2002, 24–59; Honneth and Saar 2002; on the reception of Laclau and Mouffe in German-speaking sociology see Stäheli 1995).
Combining ideas about the social construction of reality with Foucault’s understanding of discourse, this approach argues that everything we perceive, experience, and feel, but also the way we act, is structurally intertwined with socially constructed forms of approved and objectified knowledge (I summarize Keller 2011b, 58–59; see also Landwehr 2009, 91–93). We do not have an unmediated access to the world an sich, even though the ‘robustness’ of its material quality limits the spectrum of interpretation. Knowledge of the world is not a neutral understanding but the cultural response to symbolic systems that are provided by the social environment. These symbolic systems are typically produced, legitimized, communicated, and transformed as discourses. Discourse analysis, from the perspective of the sociology of knowledge, aims at reconstructing the processes of social construction, objectification, communication, and legitimization of meaning structures. What is regarded as legitimate knowledge in a given society is generated on the level of institutions, organizations, or collective actors.
There is a close parallel between this approach to the social organization of knowledge and the poststructuralist positions that have reshaped postcolonial and gender studies since the 1980s.3 If you allow me an excursus into gender studies, reference must be made to Joan Wallach Scott, who already in 1988 criticized the binary construction of the sex-gender division and insisted on an examination of that binary opposition itself (1988, 40). That brings her to a redefinition of ‘gender’: “My definition of gender has two parts and several subsets. They are interrelated but must be analytically distinct. The core of the definition rests on an integral connection between two propositions: gender is a constitutive element of social relationships based on perceived differences between the sexes, and gender is a primary way of signifying relationships of power” (ibid., 42).
The close links between sociologies of knowledge and poststructuralist understandings of gender become fully visible when we look at Barbara Hey’s variation of Scott’s definition of gender, now using the German word Geschlecht (translatable as both sex and gender) as an analytical category that transgresses the binary construction of ‘sex’ and ‘gender’:
Geschlecht is knowledge of the societal relations between women and men and as such never absolute or persistent, but always dependent on context; it is controversial and an instrument of, as well as a result of, power relations. Knowledge as a way to order the world is inseparable from societal organization. Consequently, Geschlecht is the societal organization of gender differences [Geschlechterdifferenz]. But this does neither mean that it mirrors constant, natural differences nor that it enforces them. Rather, Geschlecht provides different meanings for these distinctions in historical, cultural, and social regard. Viewed from this perspective, the “sex/gender” distinction is misplaced (Hey 1994, 19–20; unless noted otherwise, all translations of quotations are mine).
We can learn a lot from Hey’s definition of Geschlecht when it comes to the study of religion. But at this point, let me emphasize that the notion of knowledge here does not refer to an objective truth of the world4 but to the social communication, attribution, and legitimization of what is accepted in a given society as knowledge. This knowledge can be explicit, but also implicit or tacit. An example of tacit knowledge would be the societal consensus in Europe that “democracy is better than dictatorship,” that “magic does not work,” or that “astrology is not scientific.” Implicit or tacit knowledge is, generally, not tested or challenged (or even thematized) by agents in a given society; what is more, such knowledge can change significantly from one society to another and from one historical period to another. That is why historical analysis of discourse addresses not only the explicitly available forms of knowledge (for instance, in the natural sciences) but particularly the ‘self-evident knowledge,’ the truth that is not formalized but generally accepted (see Busse 1987, 40–41). This brings us to the historical dimension of discourse analysis.
Historical Analysis of Discourse
Throughout his work, Michel Foucault was interested in the genealogy, or ‘archaeology, ’ of discursive structures, which naturally implies an historical dimension in his analysis of discourse (Bieder 1998; Bublitz 1999; see Busse, Hermanns, and Teubert 1994). Therefore, it is astonishing that Foucauldian approaches have only rarely been adopted in the study of religion, arguably a discipline that has a strong historiographical focus.5 One reason for this may be that the notion of ‘history of religion’ is to a large extent associated with Eliadean phenomenology of religion (as a critique see McCutcheon 1997), which led to a general disregard of the category ‘history’ in the study of religion, particularly in the United States (see Kippenberg 2001; von Stuckrad 2003b; see also Lincoln 2005 [1996]). I will address Mircea Eliade’s impact on twentieth-century attributions of meaning to ‘religion’ repeatedly in this book.
The so-called linguistic turn (programmatically Rorty 1967) has had far-reaching impact on all fields of cultural research. That our knowledge of the world is constituted in language and linguistic structures and that the scholar is also an author whose narrative account does not provide a privileged access to truth was famously argued for historiography by Hayden White (1973) and for anthropology by Clifford Geertz (1988). To be sure, large parts of historical scholarship, including scholars of religion, shunned the consequences of this reflective critique (see Vann 1998). Even today, “sources are still read as ‘documents’ of a past reality—perhaps they are read better, more diligently and critically, but nevertheless as medium with sufficient transparency” (Sarasin 2003, 32). That historical meaning is generated in communicative processes is only insufficiently acknowledged (examples of this acknowledgment include Koselleck 2004 [1979] and Rüsen 1997). Even fewer scholars include the category of ‘discourse’ in their historical analysis or make the argument that historical meaning is not ‘reconstructed’ from the ‘facts’ and ‘sources’ in a hermeneutical process of understanding (Verstehen) but discursively generated. This is exactly what Michel Foucault wanted to show in his critical reflection on our presupposition that historical truth is attainable in our accounts of it. Since Foucault, “discourse analysis can be understood as the attempt of scrutinizing the formal conditions that steer the production of meaning” (Sarasin 2003, 33; similarly Stäheli 2000, 73).
In close reference to Foucault’s work and in conversation with structuralist approaches, several forms of discourse analysis have emerged (Maingueneau 1991, 15, distinguishes seven for the French academic discussion; see also Bublitz et al. 1999; Bublitz 2003; Mills 2004, 1–25), some of them closer to linguistic and textual analyses than others. What they have in common is the argument that there is no ‘thing’ in the world that determines what is being said but that the meanings of things are generated by the chain of signifiers that the speaker is introducing.
The thing that is meant, the referent, is as referent of a certain linguistic sign not prior to language; rather, it is the system of signs that ultimately creates it as social reality from the “chaotic variety” [chaotische Mannigfaltigkeit] (Kant) of all possible things in the world: “It is the world of words that generates the world of things” [Jacques Lacan]. Something else is fundamental for discourse analysis: This is not about the abstruse question whether there is more than texts; it is about how the non-linguistic things gain their meaning. No discourse, no grid of classification, however familiar it may appear, has ever been derived ‘from the things themselves’; it is the other way round and discourse and classification generate the order of things. […] Even though practices, gestures, and objects are themselves no longer constituted in language, they are relevant in the social world only because meaning has been discursively attributed to them (Sarasin 2003, 36; see also Busse 1987, 23; on the related concept of ‘empty signifier’ see Laclau 1994).
Or, in Achim Landwehr’s apt remark, “at the bottom [Grund] of realities and discourses there is no other fundament than their own historicity. Hence, the shortest possible definition of the function of discourses must be: discourses generate realities” (2009, 92). We can understand the working of discursive structures only if we know their genealogy and formation. And only through comparison—in diachronic or synchronic perspective—we can see the historicity and even singularity of discourses (see also Scott 2007, 8). There are no discourses that emerge ‘naturally’ or that are dictated by the working of some abstract reality; historical and comparative analysis of how social communicational structures attribute meaning to the world and organize explicit and implicit knowledge is the basis of discursive approaches.
Discourses typically lead to a “shortage of possible notions [Verknappung von Aussagemöglichkeiten] (we cannot say everything at all times)” (Landwehr 2009, 92). Thus, the historicity of knowledge should not be misunderstood as arbitrariness. What a group of people in a given situation regards as accepted knowledge is by no means arbitrary; it is the result of discursive formations that critical scholarship can reconstruct and interpret.
This is even true for knowledge that is legitimized by empirical methods in the ‘exact sciences’ and thus based on what is seen as ‘hard facts.’ Reconstructing the conditions of knowledge in the natural sciences is the goal of historical epistemology, a division within the history of science that is closely related to discursive approaches. The historicity of knowledge in the natural sciences was famously discussed by Ludwik Fleck (Fleck 1935). Edmund Husserl, Gaston Bachelard, Georges Canguilhem, and others have contributed to this debate and helped us to understand that it is not ‘nature’ that formulates natural laws but that ‘facts’ are produced in communicative and social processes (overview in Rheinberger 2006, 21–72; see also Ashmore 1989; Ashmore, Myers and Potter 1995; Latour 2010). Under the label of ‘discursive constructionism,’ Jonathan Potter and Alexa Hepburn summarize what is at stake here:
Discursive constructionism (DC) is most distinctive in its foregrounding of the epistemic position of both the researcher and what is researched (texts or conversations). It studies a world of descriptions, claims, reports, allegations, and assertions as parts of human practices, and it works to keep these as the central topic of research rather than trying to move beyond them to the objects or events that seem to be the topic of such discourse. It is radically constructionist in that it is skeptical of any guarantee beyond local and contingent texts, claims, arguments, demonstrations, exercises of logic, procedures of empiricism, and so on. In this sense it can be described as antifoundationalist and poststructuralist. It takes seriously the work in rhetoric and the sociology of scientific knowledge that highlights the contingent, normative, and constructive work that goes into, say, logical demonstrations, mathematical proofs, or experimental replications (Potter and Hepburn 2008, 275).
In a similar vein, Hans-Jörg Rheinberger, former director of the Max Planck Institute for the History of Science in Berlin and one of the most influential authors in the field, defines “the concept of epistemology, with reference to the French usage of the term, as the reflection on the historical conditions under which, and the means with which things are made into objects that start up the process of scientific inquiry [Prozess der wissenschaftlichen Erkenntnisgewinnung] and that keep this process going” (Rheinberger 2007, 11, italics original). Rheinberger exemplifies this with the history and epistemology of experimentation in the life sciences. One collaborative research project, on “A Cultural History of Heredity,” aims at “studying the juridical, medical, cultural, technical, and scientific practices and procedures in which knowledge of heredity became materially entrenched in different ways and by which it unfolded its often unprecedented effects over a period of several centuries” (Müller-Wille and Rheinberger 2007, ix). These discourses generate “epistemic spaces” in which shared knowledge is established and legitimized.6
Put into the language of discourse theory, we can say that the practices and procedures in the natural sciences are a materialization of a discourse on, in this case, heredity. The discursive materializations, in their turn, stabilize and legitimize the discursive assumptions that have made them possible. By so doing, discursive structures steer the attribution of meaning to things and establish shared assumptions about accepted and unaccepted knowledge. The example makes clear that discourse analysis breaks down the borders between the natural sciences and the social or cultural sciences. Despite their different methodologies to produce accepted knowledge, the natural sciences are no less discursively structured and thus socially steered than the humanities. As will be demonstrated in this book, discursive approaches help us to overcome the conceptual boundaries between the natural sciences and the humanities in our attempt to understand the communal production of ‘academic knowledge.’ It will also become clear that academic producers of meaning form a discourse community with non-academic authors.
Consequently, discourse analysis argues that our knowledge is not about ‘the world out there’ (even if the existence of ‘a world out there’ is not denied) and that we should adopt a relativist rather than a realist position in the philosophical debate that is linked to these epistemological and ontological issues. The relativist position has led to many, often highly polemical objections. Derek Edwards, Malcolm Ashmore, and Jonathan Potter call the most prominent rejection the “Death and Furniture” response:
‘Death’ and ‘Furniture’ are emblems for two very common (predictable, even) objections to relativism. When relativists talk about the social construction of reality, truth, cognition, scientific knowledge, technical capacity, social structure and so on, their realist opponents sooner or later start hitting the furniture, invoking the Holocaust, talking about rocks, guns, killings, human misery, tables and chairs. The force of these objections is to introduce a bottom line, a bedrock of reality that places limits on what may be treated as epistemologically constructed or deconstructible. There are two related kinds of moves: Furniture (tables, rocks, stones, etc.—the reality that cannot be denied) and Death (misery, genocide, poverty, power—the reality that should not be denied) (Edwards, Ashmore, and Potter 1995, 26, emphasis original; see also Potter and Hepburn 2008, 287–288; Nikander 2008, 413; more generally Parker 1998).
What is at stake is not “a lack of concern with that which may exist beyond discourse” (Benavides 2010, 210, as a critique of my position) but an acknowledgment of the difference between something that simply happens (often without being reported) and something that is made into a fact or event by discursive and communicative procedures. As Antonio Gramsci reminds us, without an inventory that organizes our knowledge about ourselves and our history, understanding is impossible. “The starting-point of critical elaboration is the consciousness of what one really is, and is ‘knowing thyself’ as a product of the historical process to date which has deposited in you an infinity of traces, without leaving an inventory. Such an inventory must therefore be made at the outset” (quoted from Forgacs 1988, 326).
Discursive Study of Religion: Concepts and Definitions
I have explained so far how recent discussions in the sociology of knowledge and the historical analysis of discourse have provided important new interpretational tools that make use of Foucauldian and poststructuralist approaches, at the same time adjusting them to their own needs and research interests. Now I turn to the question of how we can apply these considerations to the study of religion. I argue that a discursive study of religion is the most convincing form of analysis, if we want to avoid the traps and challenges that have confronted the study of religion during the twentieth century.
Before I discuss the most important concepts and definitions, it is necessary to point out that discourse analysis is not itself a method. Sometimes harking back to Ludwik Fleck’s notion of Denkstil (“thought style,” see Fleck 1935), many theorists of discourse agree that discourse analysis is a research perspective or research style that applies a spectrum of possible methods in order to answer its guiding research question.7 This is in contrast to those approaches that focus on linguistic analysis or what Norman Fairclough calls “textually oriented discourse analysis.”8 Within the more historically oriented discourse theory that I am advocating here, and which leans more heavily on Michel Foucault’s work, the methods that are considered useful can range from philological methods to quantitative and qualitative methods, content analysis, etc. (for the study of religion, see the overview in Engler and Stausberg 2011). However, even if discourse analysis is not a specific method, it follows certain steps and rules that have proven useful in concrete analytical work. As I will explain below, these steps consist of the demarcation of the discourse under scrutiny, the collection of relevant data, and the decision of which method would be most productive in collecting and interpreting the data. I will explain these steps with direct reference to the argument and material of the present book.
Basic Concepts for Discourse Analysis
Let me now clarify the terms that are most relevant for our purpose here. Making use of the recent discussion that I have outlined above, I define ‘discourses’ as follows: Discourses are communicative structures that organize knowledge in a given community; they establish, stabilize, and legitimize systems of meaning and provide collectively shared orders of knowledge in an institutionalized social ensemble. Statements, utterances, and opinions about a specific topic, systematically organized and repeatedly observable, form a discourse. Hence, the concept of discourse refers to “the regularity of fields of statements, which regulate what can be thought, said, and done” (Stäheli 2000, 73). When it comes to the link between several discourses, we can conceptualize these as “intertextual and inter-discursive relationships between utterances, texts, genres and discourses, as well as extra-linguistic social/sociological variables, the history of an organization or institution, and situational frames” (Reisigl and Wodak 2010, 90).
Consequently, ‘discourse analysis’ addresses the relationship among communicational practices and the (re)production of systems of meaning, or orders of knowledge, the social agents that are involved, the rules, resources, and material conditions that underlie these processes, as well as their impact on social collectives (similarly Keller 2011b, 8).
‘Historical discourse analysis’ explores the development of discourses in changing sociopolitical and historical settings, thus providing means to reconstruct the genealogy of a discourse.
In addition to these fundamental terms, it is useful to introduce the concept of ‘dispositive,’ a term which was coined by Michel Foucault (le dispositif, often translated as ‘device,’ ‘deployment,’ or ‘apparatus’), but which recently has been defined more clearly in scholarly discussions. The concept moves beyond the analysis of discursive practices to include non-discursive practices and materializations, 9 tacit and implicit knowledge, as well as the relationship between these dimensions of social action. A dispositive is here understood as the material, practical, social, cognitive, or normative ‘infrastructure’ in which a discourse develops. This can include governmental decisions and laws, new technologies and media, museums, educational programs, television, or the healthcare system. Dispositive analysis examines how “assignments of meaning create reality” (Jäger and Meier 2010, 39; as a detailed introduction see Bührmann and Schneider 2008). The distinction between discourses and dispositives becomes clear when we consider examples. Television, the Internet, a governmental decision, and an institution such as the Nobel Prize are not ‘discourses’ in themselves (discourses on what?), but they provide the communicative infrastructures in which attribution of meaning becomes operative. That is why historical discourse analysis has to include these dimensions of communicative structure.
Discourses develop within cultural processes and dispositives. They form around specific topics, but many discourses also contain ‘strands’ from other discourses. For instance, the statement “The preamble of the future constitution of the European Union should refer to Christianity as Europe’s religious and philosophical roots” is linked to several discourses, particularly discourses on European identity, on constitutional law, on religion, on Christianity, and on philosophy. What we see here is that several discourses can be entangled and form a ‘discursive knot’ (Jäger and Meier 2010, 47). The notion of discursive knots reminds us of the fact that the borders of a discourse are flexible and dependent on scholarly definition, which means that discourses do not exist ‘out there.’ They have no ontological status other than being analytical categories that the analyst of cultural processes constructs to serve her or his interpretative goal.
This is entirely in line with Michel Foucault’s understanding. What I do in this book, namely looking at re-entanglements of discursive knots in historical perspective, resembles Foucault’s program of deconstructing and reconstructing analytical frameworks:
The […] purpose of such a description of the facts of discourse is that by freeing them of all the groupings that purport to be natural, immediate, universal unities, one is able to describe other unities, but this time by means of a group of controlled decisions. Providing one defines the conditions clearly, it might be legitimate to constitute, on the basis of correctly described relations, discursive groups that are not arbitrary, and yet remain invisible. […] [I]t is not therefore an interpretation of the facts of the statement that might reveal [the relations], but the analysis of their coexistence, their succession, their mutual functioning, their reciprocal determination, and their independent or correlative transformations (Foucault 2010 [1972], 29).
My project is Foucauldian in its analytical strategy as well. When I ‘disentangle’ and reconstruct discursive knots that have been tied around the concepts of ‘religion’ and ‘science,’ I suggest new ‘unities’:
I […] will do no more than this: of course, I shall take as my starting-point whatever unities are already given (such as psychopathology, medicine, or political economy); but I shall make use of them just long enough to ask myself what unities they form; by what right they can claim a field that specifies them in space and a continuity that individualizes them in time; according to what laws they are formed; against the background of which discursive events they stand out; and whether they are not, in their accepted and quasi-institutional individuality, ultimately the surface effect of more firmly grounded unities. I shall accept the groupings that history suggests only to subject them at once to interrogation; to break them up and then to see whether they can be legitimately reformed; or whether other groupings should be made; to replace them in a more general space which, while dissipating their apparent familiarity, makes it possible to construct a theory of them (ibid., 26).
Consequences for a Discursive Study of Religion
In a discursive framework of analysis, it does not make any theoretical difference whether we study religion, politics, technology, cars, animals, music, masculinity, or any other topic of social and symbolic communication that is linked to an identifiable discourse. But for the study of religion as a specialized area of research, discursive approaches have implications that need to be made explicit. To begin with, religion completely loses its status of being something sui generis. Rather, discursive approaches study the very claim that “religion is sui generis” as part of a discourse on religion that has formed under identifiable historical circumstances and that has materialized in university institutions and scholarly programs, in turn stabilizing and legitimizing the attributed meaning of religion as sui generis.10 We can historicize the discourse on sui generis religion; what is more, we can scrutinize the dispositives and discursive knots, which characterize this discourse and maintain the construction of meaning, until at some point in the historical development other discourses determine the socially communicated knowledge about religion.
Discursive approaches provide a solution to another problem as well. It is no longer necessary—in fact, it would be counterproductive—to apply a generic definition of religion (see also von Stuckrad 2010b, 165 – 167). Definitions of religion are statements and utterances that attribute meaning to things and that provide orders of knowledge. As contributions to a discourse on religion, these definitions are objects of discursive analysis rather than its tools.
Regarding the term ‘religion’ as an empty signifier that can be activated with definitions, meanings, and communicational practices does not compromise the clarity of the object nor the scholarly rigor of the study of religion. It only moves the obligation to define our objects from the level of communicational practices to the level of discursive reflection. I make this distinction visible in a change of typeface: ‘religion’ refers to contributions to a discourse on religion, while ‘RELI GION’ refers to the discourse itself.11 After this clarification, we can go a step further and define RELIGION simply as follows: RELIGION is the societal organization of knowledge about religion. In the same vein, SCIENCE would be defined as the societal organization of knowledge about science.
The definitions of ‘discourse analysis’ and ‘historical discourse analysis,’ as given above, pertain to the discursive study of religion as well, only that now the analysis is directed to the discourse on religion (i.e., to RELIGION), the dispositives that serve as the infrastructure of RELIGION, and possible entanglements with other discourses.12 RELIGION produces meanings and orders of knowledge that materialize in concrete practices and institutions; these orders interact with non-discursive practices and dispositives such as the organization of higher education or the appearance of new media and technology. Moreover, as a discursive constellation RELIGION is entangled with other discursive constellations, which could be defined as LAW, SCIENCE, SPIRITUALITY, MAGIC, HEALTH, ECONOMY, HEREDITY, or any other discourse that may be of interest for scholarly analysis. Dispositives and discursive knots are subject to change, which means that RELIGION is fully historicized and open to intercultural comparison.
Methodological Implications
As noted above, we should not consider the study of discourse itself to be a method. Rather, as a research perspective, or research style, the study of discourse is an interpretative endeavor and thus a hermeneutical strategy to scrutinize the organization of knowledge in a given societal and historical situation (Keller 2005; Keller 2011a, 76–78). Nevertheless, despite the openness to apply various methodological tools and the flexibility to adjust these tools to the research question that is at stake in a given project, there are a few basic considerations and implications that arise when one decides to engage a discursive research perspective. We can differentiate three important steps in designing and carrying out an historical discourse analysis. I will briefly explain them, using examples from the analysis of RELIGION and SCIENCE.13 This will also serve as an introduction to the main argument of the present book.
Determining the Research Question
In principle, all research is intrinsically bound to a discursive construction of meaning and the organization of knowledge. It is therefore a characteristic of discursive perspectives that they can lead to a better understanding of complex dynamics in the generation of approved knowledge—whether the research in question is itself explicitly discursive or not. But even if everything can be studied from a discursive perspective, not all research is itself a discursive analysis. Discursive studies have concrete research questions that may differ from research questions as they are typically framed in historical, philological, anthropological, or sociological perspectives.
Historical discourse analysis is interested in the processes of communicational generation, legitimization, and negotiation of meaning systems. When we look at the discursive field of RELIGION and SCIENCE, countless research questions are possible. In an arbitrary selection, and limiting our focus to European contexts, we can pose the following general questions that can be framed discursively: (a) “What are the structures that regulate public opinion about astrology as op- posed to astronomy in Europe?” (b) “Why is shamanism so attractive to many people in Europe and North America?” (c) “How can we explain the rise of psychological interpretations and their link to metaphysical and religious ideas in the twenthieth century?” (d) “What is the role of religion in the field of natural sciences today?” To be sure, these are general questions that have to be broken down into more concrete sets of subquestions, but the examples should make clear what kind of questions we typically ask in a discursive analytical framework, and which questions will be addressed in the subsequent chapters of this book.
We can translate these questions into discursive language and design a project that reconstructs the discursive entanglements of RELIGION on the one hand and ASTROLOGY, SHAMANISM, PSYCHOLOGY, and SCIENCE on the other. All these discourses lead to further discourse strands that characterize the respective entangled discourse. After a thorough screening of the available data (see the next passage), it turns out that in the case of SHAMANISM, these further discourse strands include HEALING, SOUL, NATURE, THERAPY, and CONSCIOUSNESS. It depends on our research focus whether we will need to include all of these discourse strands or only a selection of them.
Selecting Data and Building a Corpus
When we have formulated a research question and possible sets of subquestions that concretize what we are interested in, the next step is the selection of the data that is most suitable for finding an answer to our question. It is important to note here that a sound discursive study has to start with a substantial reading of the material under consideration, in the broadest possible sense. That contributions to a discourse must be “repeatedly observable” is a relevant detail in my definition of ‘discourse.’ This also means that the establishment of a discourse for further analysis can be reasonably combined with methods of Grounded Theory. For instance, to find the basis for my construction of the discourses of SCIENCE and RELIGION, I had a close look at a large number of texts in which ‘science’ and ‘religion’ are used; subsequently, I noted down the other concepts that were repeatedly used as entangled discourse strands in these documents; this led to what can be called groupings of entangled discourse strands, such as SOUL, VITALISM, MONISM, POWER, ENERGY, etc. Put differently, a discursive study is a hermeneutical circle that starts with a broad reading of the most various documents, proceeds with heuristic groupings of discourse strands, and refines the construction of these groupings by revisiting the material and including more documents; this circular refinement can be repeated until the understanding of the respective discourse and its historical development will be good enough to be put into a thesis that is clearly based on relevant material.
Since discursive approaches are not limited to textual sources, data can be found in all forms of communication that are operative in the attribution of meaning. With regard to the research questions formulated above, possible data sets would include (but are not limited to): (a) books about astrology and astronomy, media and newspaper coverage, Internet discussions, official statements about astrology in governmental documents or scientific organizations (research organizations, Nobel Prize documents, etc.); (b) books on shamanism (often popularizing academic theories), workshop programs, Internet forums, interviews with shamanic practitioners; (c) academic as well as popular publications in the field of psychology, academic correspondence and ego-documents, governmental documents that regulate the psychological and therapeutic market; (d) books on contemporary science, presentation of research results in the media and on the Internet, museum exhibitions. In order to address the genealogy of the discourses under scrutiny, it is important to add an historical dimension to the research outline as well. My analyses in this book are clearly focused on the genealogy and thus the formation of the discursive entanglements that are derived from the overall research questions.
In addition to concrete sets of data, the research should also include an analysis of dispositives that serve as the ‘infrastructure’ of the discourse under scrutiny: for the present study I looked at new technologies and media (electricity, television, Internet); the history of associations and scholarly organizations (such as the Theosophical Society, the German Monist League, and new book series for a lay audience); governmental rules (for instance with regard to the juridical and tax status of shamanic, astrological, or pagan practitioners); the funding of scientific research; programs in teaching and research at universities; etc.
The lists of possible—or even necessary—sets of data make it clear that a full-blown discursive analysis requires a lot of time and resources. The selection of data and the building of a research corpus will thus be dependent on possible constraints and practicalities. Individual projects can be part of a larger research program or even consider themselves simply as a contribution to an overall analysis of a discourse. It is methodologically acceptable to design an exemplary study that highlights one aspect of a discourse.14 What is more, within a larger research program it may be necessary to set up small projects first that make relevant data available for further discursive analysis (such as the critical edition of a text or a social-scientific survey).
Choosing the Most Suitable Method to Analyze the Data Sets
As noted above, the methods that are applied in a discursive analysis can vary, depending on the most suitable way to interpret the research corpus. Typically, a discursive study of religion makes a selection of the following research methods: content analysis, conversation analysis, media analysis, participant observation, textual interpretation, surveys and interviews, and historical methods (see Engler and Stausberg 2011). For the research I needed to prepare the present book, I applied a combination of historical methods, content analysis, and textual interpretation.
The generation and interpretation of data is not a goal in itself. It serves the overall question that is formulated within a discursive referential framework. Consequently, the data is used and interpreted with reference to the organization of knowledge in a given setting as well as to the attribution of meaning to things and events. It is fully acknowledged that the research results are themselves elements of the discourse under scrutiny; hence, they do not represent the ‘truth’ about the issue at stake but provide insight into the mechanisms, historical dimensions, and implications of the construction of meaning in a discourse community.
Although the main argument of this book is based on an extensive reading of documents, which gives me enough confidence about the relevance of the material under consideration, it would be presumptuous to claim that my (re)construction of the processes of changing entanglements of science and religion during the past two-hundred years is the concluding result of an analysis that would include many more sources than one author can possibly manage. If readers regard this book as a ‘pilot study’ of the scientification of religion and test my hypothesis against a rereading of the same material or against other contributions to the discursive knots under consideration, the goal of this book would be accomplished.
Beyond Binaries: An Outline of the Argument
Often, analyses of the processes that I engage in this book operate within a binary pattern, juxtaposing religion and science, professional and amateur, or modern and pre-modern. Discursive approaches are skeptical of these binary constructions. In a way, discursive approaches are themselves discursive materializations of questions that have been raised with reference to binary models of interpretation, such as true and false, insider and outsider, or culture and nature. The twentieth century has seen a fundamental break with these binaries, and the new cultural studies have taken up the challenges that came with the break. Self-consciously presented as “a new paradigm in cultural studies,” the notion of the third has recently gained influence as a new way to think beyond the binaries and to include the ‘in-between’ as significant characteristic of contemporary culture. Albrecht Koschorke, one of the leading thinkers in this field, explains that the third (das Dritte) as a way of thinking was developed in the twentieth century and cannot be ignored anymore.
Here, the exceptional state is made permanent as it were. When in the encounter of two parties none of the two sides can assert a hegemonic claim—a claim that brings back the Other into the Own, viewing the opposite as derivate of a higher-level order that is identical with the own—then a new grammar of cultural and epistemological negotiation is needed, which traditional means cannot achieve (2010, 13, emphasis original).
Indeed, the intellectual discussions in various disciplines throughout the twentieth century have made unmistakably clear that notions of ‘hybridity’ or the ‘third space’ (Bhabha 2004, 53–56; see Bhabha 1990) not only represent a critique of binary constructions but also a new vocabulary that is needed to understand processes of cultural transformation (see also de Certeau 2010 as discursive considerations on the ‘Other’). Thinking in triads, in which the third is not the synthesis of one of the opposites, is a red thread that runs through many disciplinary contexts, from philosophy to anthropology to sociology to the study of religion, and even to economy and law (Eßlinger et al. 2010, 35–149, see 316–322 for a list of relevant literature; see also Breger and Döring 1998). Gender studies and (post)colonial studies are heavily influenced by these theories. In a recent research program, scholars have identified new ‘figures’ (or, rather, figurations) of the third that have left their pariah position and have become central analytical instruments, among them the messenger, the cyborg, the parasite, the laughing third, the trickster, and the rival (Eßlinger et al. 2010, 153–315). In the context of the present book, the ‘amateur’ would also qualify as a third that is not adequately described as the overcoming of either professional or ‘pseudo’ knowledge. We may even reposition the ‘intellectual’ in new triadic constellations of twentieth-century culture.
Discursive approaches can provide the vocabulary we need for the new grammar of cultural and epistemological negotiation that Koschorke calls for.15 Legitimization, de-legitimization, construction, and reconstruction of knowledge in changing entanglements of discourse strands form the basic structure of this epistemological negotiation.
The following chapters will develop and test such a vocabulary with regard to the question of whether we can come to a better understanding of the complex relationship between religious and secular discourses in Europe. I will reconstruct the genealogy of distinctions between religion and the secular, but I will do so by focusing on subfields in which these distinctions become clearly visible. Legitimization and de-legitimization of knowledge are particularly relevant for the histories of astrology, alchemy, and magic. Often lumped together under the rubric of ‘occult sciences,’ these systems of knowledge have played a special role in European imagination at least since the fifteenth century (see Zika 2003; Hanegraaff 2012; on the concept of ‘occult sciences’ see my introduction to Chapter 3 below). One reason for their ambivalent status as knowledge systems may be the fact that they subscribe to rational philosophies of nature and that they formulate theories about nature; thus, they are perceived as competitors to what has become ‘scientific’ knowledge after 1800.
In the previous section I already described the relevant steps that structure the discursive analysis in this book. Based on this approach, Part One addresses the discursive constellations in which the ‘occult’ sciences, but also nature-based philosophies in general, have gained their meaning during the past centuries. Chapter 2 reconstructs the genealogy of discourses on astrology as a controversial system of knowledge that lost ‘scientific’ legitimacy and at the same time was re-entangled in scientific discourses after the nineteenth century. Chapter 3 does the same with alchemy, arguing that vitalism and related philosophies of nature are relevant discursive entanglements that lent new legitimacy to alchemy in the twentieth century. The same is true for a pantheistic monism (Chapter 4) that was modeled by natural scientists at the beginning of the twentieth century, thus propagating a veneration of nature that was fully compatible with a scientific worldview. Chapter 5 gives credit to the Theosophical Society, which served as an important link between academic and non-academic interpretations of science and religion, with a special interest in ‘occult’ or esoteric forms of knowledge.
Part Two looks at concrete examples of academic experts whose writings had a considerable impact on the attribution of meaning to religion. I address them as “religious pioneers” of the twentieth century, not because they all intended to found new religions, but because they lent scientific authority to new religious interpretations and religious practices. In other words, these scholars were catalysts of religious change and the emergence of new religious communities in the twentieth century. Chapter 6 contextualizes the work of Martin Buber, Gershom Scholem, Rudolf Otto, and Gerardus van der Leeuw in their intellectual environments and describes the concrete impact these scholars had on subsequent understandings of ‘mysticism’ and ‘religion.’ Chapter 7 analyzes the academic construction of the ‘Great Goddess’ and her veneration in European history, which had a clear impact on religious practice in twentieth-century witchcraft and nature-based spirituality. Chapter 8 is devoted to scholarly constructions of shamanism and how they were turned into religious practice in the second half of the twentieth century.
The concluding chapter aggregates the historical material and links the analysis to recent discussions about secularism, secularization, the secular, and modernization. I will argue that if we talk of ‘scientification of religion’ as a discursive constellation we will be able to overcome the binaries that distort many interpretations of the place of religion in contemporary Europe.