Introduction

More than 15 years after his death, Pierre Bourdieu (1930–2002) is without question one of the leading social scientists of the twentieth century. Not only has his work been cited to an almost unparalleled degree; the principles guiding his research are increasingly employed across many disciplines within and outside the social sciences. However, even today, this comprehensive reception of Bourdieusian concepts, and the research carried out using them, suffers from a relative lack of attention to the methodological implications and foundations of his work. The aim of this book is to explicate the research principles and methodology of Bourdieusian sociology and to contribute to its continuing systematic elaboration.

For a great number of researchers from the social sciences and neighboring disciplines, Bourdieu is especially appreciated for his theoretical work, his critical stance, and his impassioned approach to practicing science. In his early Algerian studies, as well as in the The Weight of the World (Bourdieu 1999a), his intention was to ensure that the living worlds (and world views of the suffering population) were made visible by his research; ultimately, the goal was to contribute to improving their life conditions. Consequently, Bourdieu argued that sociological analyses need to adequately grasp the actors’ subjectivity (in order to escape the pitfalls of ‘social physics’) by way of participant objectivation (Bourdieu 2003).

As a matter of fact, objectivation is a core aspect of Bourdieu’s sociology: In his concept of research, it is essential to break with one’s everyday understanding of the actors (including scientific actors) in order to prevent a reproduction of one’s everyday ideologies (Bourdieu 2006[2001]). This condition can be fulfilled by systematically objectifying the respective phenomena, thus revealing the relational system in which they manifest themselves. At this point, the concepts of social spaces and fields come into play: These key theoretical terms do not merely constitute analytical means of thinking; their usefulness only emerges in immediate interaction with empirical field work. Although he was opposed to rigid methodological fetishism, Bourdieu’s empirical research was based on the relational construction of fields and spaces. This core facet of his methodology motivates a specific type of statistical data analysis: the construction of social spaces using the techniques from L’Analyse des Données (Benzécri et col. 1973; with respect to social spaces: Rouanet et al. 2000; Lebaron 2009), or Geometric Data Analysis (GDA) (Le Roux Rouanet 2004).

In La Distinction (1984[1979]), which is sometimes regarded as Bourdieu’s most important work, he described the correspondence between professional fractions and lifestyle information based on statistical analyses of survey data. In his field-theoretical work Homo Academicus (1988[1984]), he undertook a formal construction of the Parisian academic field using multiple correspondence analysis (MCA), which is a major part of GDA. Finally, in one of his most recent articles – Une Révolution Conservatrice dans L’Édition (1999b), on the decline of French publishing culture – he combined MCA with clustering techniques.

This specific methodological program of quantification and formalization is the consequence of a critical reflection on the limitations of traditional quantitative techniques. Bourdieu’s research practice of relating modern social theory with advanced statistical modeling would have been impossible without the contributions of Jean-Paul Benzécri, and his colleagues and followers, who provided Bourdieu with the tools necessary for the objectivation of social spaces and fields.

In this introductory chapter, we will first outline the principles of Bourdieusian reasoning. We will then discuss the methodological implications of his paradigm with a particular emphasis on the concept of relation. Finally, we give a short overview of the contributions to this edited volume.

Fundaments of Bourdieusian Sociology

Bourdieu’s research is characterized by a consistently relational view: With his dictum “the real is relational” (Bourdieu 1998: 15), he put relations as constituents of (social) reality – in the foreground. Thus, (human) entities or (interpersonal) structures, which are traditionally at the forefront of sociological thinking, have a secondary analytical value compared to the relational constitution of the (social) reality:

I must, at every stage, make sure that the object I have given myself is not enmeshed in a network of relations that assign its most distinctive properties (Bourdieu and Wacquant 1992: 228).

From an epistemological point of view, Bourdieu emphasized a relational mode of object construction; that is, a process of identifying and modeling similarities and differences between entities and their characteristics. This fundamental epistemological concept of difference and similarity is addressed by the concepts of “social spaces” and “social fields”, which are defined as the totality of mutually external positions.

The notion of space contains, in itself, the principle of a relational understanding of the social world. It affirms that every “reality” it designates resides in the mutual exteriority of its composite elements (Bourdieu 1998: 31).

This stance characterizes Bourdieu’s conception of the social: His theoretical concept of social spaces (that is, his conception of society itself) and social fields (as an expression of differentiated social spheres) are the basis of his research program. Social spaces and fields are characterized by their endogenous structuration – structures of congruity and oppositions that are conceptualized as specific capital dimensions and class (or group) relations. Social spaces and fields also have specific meanings, orientations, and rules, which are expressed by concepts such as illusio, nomos, and doxa.

Bourdieu’s concept of action is that of practice, thus emphasizing the practical meaning of everyday action. This model of practice challenges both structuralist notions of automatic compliance with rules and the idea of autonomous action (cp. Lamaison 1986). An essential feature of Bourdieu’s paradigm is to analytically and practically transcend both the traditional structure-practice dichotomy and the distinction between the macro and the micro level.

The concept of habitus illustrates this relational and integrative position: Habitus means a bundle of patterns of perception, evaluation, and action (dispositions) which can be analytically located within a space or field (or position). In the context of the social space, Bourdieu constructed class habitus relationally; for instance, the habitus of the middle classes is constructed in contrast to those of both the upper and the lower classes. He further focused on the interrelations between male and female habitus and on the intersections of classes and gender (as well as, among other things, ethnicity). Likewise, he described social fields in the context of ‘ideal-type’ forms of habitus and their interrelations: For example, in the context of the academic field, he compares the habitus of (more autonomous) scientists to that of (more heteronomous) research managers (Bourdieu 1988[1984]).

Based on their acquired habitus, actors exhibit particular practices in their responses to current environmental conditions. Their habitus are oriented to certain social fields and result in practices that fulfil the requirements of these fields. However, it is possible for incompatibilities between habitus and social fields or social spaces to become a new basis for social practice: Within social fields and spaces, actors and groups of actors (classes) engage in conflicts (and co-operation) over goods and objects of interest, as well as for definitional sovereignty over rules and values. At the same time, social fields and social spaces are also characterized by the fact that actors install themselves in material and meaningful niches, acknowledging (and, under certain circumstances, misunderstanding) their own position in the social space. This (sociological) view of social conflicts is based to a considerable degree on mechanisms of cognitive theory: The objective structural conditions of social spaces and social fields are incorporated into the perceptual structures of the actors, so that the actors perceive the social world through them and reproduce them in their practice. Bourdieu summarized these phenomena as relations of symbolic power between actors.

These basic concepts and their relational architecture guide the Bourdieusian approach to conducting research.

Methodology

The fundamental principle of Bourdieu’s empirical work follows the same ‘logic of relation’ that was previously noted as being central to Bourdieu’s stances on social theory. First, relationality manifests itself in terms of an epistemology in the tradition of Karl Mannheim and Gaston Bachelard; the social position of the researchers, as well as their position in the disciplinary field, is understood to also be a genuine factor in their own position-takings, and thus in the production of scientific goods itself.

Second, this relational logic also manifests itself in the form of a particular closeness between theoretical and empirical work, that is, their reciprocal combination. In fact, the common differentiation between theory and empirical work is replaced by an integrated understanding of scientific research.

Third, the process of object construction is based on a relational logic: Social fields and social spaces are empirically constructed, beginning with the collection of the data and continuing until their analysis (Le Roux and Rouanet 2004; Blasius and Schmitz 2014; Lebaron 2018; Lebaron and Le Roux 2015). With respect to the practice of research, the logic of relation – and the relational concepts (such as fields, habitus, or practice) derived from it – served Bourdieu as (implicit) hypotheses for the control of observation and questioning. In this process, the theoretical concepts introduced above play a crucial role in selecting the most important aspects from the infinite variety of social reality. An empirical investigation requires ‘feature carriers’ (statistical units such as humans or institutions) in order to describe with regard to their objective and subjective features – features that are not simply observed or measured, but rather identified in relation to each other. Thereby, the collection of suitable data is characterized by a process of recursive, mutual categorization of actors (or institutions). Even in his early field work in Algeria, Bourdieu’s approach can be characterized by its inclusion of the relational collection and coding of various types of data, in order to identify relations and representations of social structures (Blasius and Schmitz 2014).

Fourth, in doing so, data types and data collection comprise both ‘quantitative’ and ‘qualitative’ connotations. In fact, Bourdieu was an early proponent of mixed methods and triangulation.

Fifth, one of Bourdieu’s goals in the objectification of any particular phenomenon was to apply an ‘epistemic rupture’, i.e., a break with the perceptions and dispositions of the investigated actors to whom the relational analysis is being applied, but also a break with the researcher’s own views and presuppositions. Despite the relevance Bourdieu ascribed to qualitative data, such objectification requires “statistical analysis”, which “is the only means of manifesting the structure of the social space” (Bourdieu 1985: 725). Even in his early Algerian studies, he was already analyzing complex data; in this case he used punch cards, which he marked in order to map simultaneously occurring or mutually exclusive characteristics, thus enabling the visualization of contrasting and equivalent relations (Bourdieu 1990: 8). His main problem was the evaluation of more than a small number of oppositions within each analysis. Although Bourdieu applied several graphical visualization techniques and different statistical approaches, he could not find a satisfactory representation of dimensionally structured relations, as embedded in his theoretical concept. Being deeply convinced that the world is multi-dimensional, he rejected all forms of linear approach, such as ordinary regression models: These methods usually have the goal of explaining a dependent variable over a series of independent variables, and were thus methodical approaches that did not match the methodological principles of his theory. Consequently, his research was quite unlike the usual practice of statistical modelling: Instead of constraining the data with strong assumptions such as linearity between variables (and between categories), he focused on the relational structure of the data by using geometric modeling (construction and interpretation of geometric maps).

In the same period of the 1960s, Benzécri started his work on L’Analyse des Données, focusing especially on simple and multiple correspondence analysis (CA, MCA) (Benzécri et col. 1973); his underlying idea was that a “model should follow the data and not vice versa” (cp. Blasius and Greenacre 2006: 6). CA and MCA differ from traditional statistics, which involve significance tests, normality assumptions, and attempts to fit the data to a previously specified model. CA and MCA identify the structure of data without fitting the data to a statistical model; this was the non-linear technique Bourdieu was searching for:

If I make extensive use of correspondence analysis, in preference to multivariate regression for instance, it is because correspondence analysis is a relational technique of data analysis whose philosophy corresponds exactly to what, in my view, the reality of the social world is. It is a technique which “thinks” in terms of relation, as I try to do precisely with the notion of field (Bourdieu and Wacquant 1992: 96).

From the geometric point of view, as is typical for the French tradition in the GDA framework, any data set is conceptualized as clouds of points, and thus a social space that relates these points and these clouds to each other. Although CA and MCA interpretations are usually based on maps (social spaces), and (geometric) distances within these maps, the underlying geometric model provides a numerical solution that can be interpreted much like the well-known principal component analysis (PCA). In fact, MCA can be understood as PCA applied to qualitative data. From the mid-1970s to the present day, Bourdieu and his adherents have used CA and MCA to empirically objectify the relational conception of the social – both methods transform data into (two-dimensional) maps (for a differentiation between these two methods in the Bourdieusian framework, see Blasius and Schmitz 2014). The formally constructed social space allows social scientists to interpret non-linear relations between indicators (for example lifestyle indicators), between individuals (for example academics Bourdieu 1988[1984] or economists (Lebaron 2008) or between groups of individuals (or example groups of class fractions (Bourdieu 1984[1979]).

A core difference between relational methodology and individualist methodology is that the former does not assume theoretical and statistical independence of the analyzed entities. Even advanced ‘causal’ approaches, such as fixed and random-effect regression models, assume the independence of entities, reducing causality to the ‘averaged internal mechanisms’ of artificially separated individuals. As a consequence, individualistic perspectives fail to reveal causality between agents; that is, relational causality (cp. Schmitz 2016). CA and MCA, in contrast, ‘rehabilitate’ the individual and its interdependencies within the world of statistical modeling as entities, and their relational positions become genuine objects of analyses.

Bourdieu used CA in Anatomie du Goût (Bourdieu and De Saint-Martin 1976) as well as in La Distinction (Bourdieu 1984[1979]). The aim of these publications was to develop an integrative vision of the French social space with a particular emphasis on the dominant classes. With increasing computational power and new possibilities for analyzing complex data (Benzécri 2006), Bourdieu later primarily applied MCA, e.g. in La Patronat (Bourdieu and De Saint-Martin 1978) and in Homo Academicus (Bourdieu 1988[1984]), but he was also interested in extending the set of methodical tools within the framework of GDA (cp. Rouanet et al. 2002).

In sum, Bourdieu’s work resulted in a powerful sociological research program that provides social sciences with comprehensive insights, while preserving the impulse to develop it further. All attempts to do so should keep in mind its paradigmatic core: methodological relationality.

Following Lebaron (2011: 87ff.), a modern relational research program based on a Bourdieusian perspective aims to: (1) Reveal the structure of a social field and/or of a specific social space. Here, descriptive procedures should always come first, while an inference procedure should follow as a ‘natural’ extension of descriptive conclusions (Rouanet and Le Roux 2010: Chapter 5). (2) Show possible structural homologies between different social fields and/or social spaces, such as the (varying) positional homology between producers and consumers of cultural goods. (3) Determine the relative autonomy of social fields and/or social spaces. Bourdieu put a particular emphasis on the relations between social fields, where the first step of the analysis is to consider the embedding of the social fields within the field of power (Bourdieu and Wacquant 1992; Schmitz et al. 2017). (4) Study sub-spaces within a global social space. For example, a social class can diverge from an overall capital structure, and is structured by the composition of different forms of capital, for example, cultural and economic capital (Bourdieu 1984), as well as by the weighted sum of capitals, i.e., the capital volume (for the relation between the forms of capital, capital volume, and the composition of capitals within a (two-dimensional) social space, see Blasius and Friedrichs 2008). (5) Explain social practices such as position-takings. A good example is the necessity for objectifying the scientific subject within Bourdieusian sociology. Even seemingly scientific statements can be (partially) traced back to objective positions within both the social space and the scientific field. (6) Assess the importance of various effects, especially field effects. This includes the effects social fields may have on the habitus of their individuals, as well as effects on individuals of different social fields. (7) Study the dynamics of social fields: Bourdieu’s concepts genuinely involve historicity and change. Issues of dynamics can be found in many of his empirical investigations: for example, the transformation of Algerian society, reforms in the French educational sector, trajectories of classes, and changes in the dominant capital forms.

These aims are realizable with the application of the techniques and concepts of GDA, which have been available since sociologists first began to construct social spaces. New theoretical considerations motivate the further developments of statistical techniques – for example, specific MCA (Le Roux and Rouanet 2004, 2010) and class-specific MCA (Le Roux and Rouanet 2004, 2010).

Contributions of the Authors

The book is organized in three parts. Part one deals with the social space and its construction, part two is devoted to social fields as sub-segments of the social space, and part three addresses methodological and methodical questions.

Social Spaces

The first section of the book is devoted to the construction and analysis of social spaces; that is, entire societies, thus treating societal core issues such as class relations and capital distributions.

The first section starts with Loïc Wacquant’s contribution, which argues that the triad of ‘habitus, capital, and field’ can be replaced by the dyad of ‘social space and symbolic power’. This shift from a triad to a dyad of analytical core concepts illuminates the inner logic of Bourdieu’s concepts, and also serves to re-orient observers towards the ways the symbolic is woven into comprehensive societal power relations. Following Bourdieu’s work on the grandes écoles, Ida Lidegran, Mikael Börjesson, Donald Broady & Ylva Bergström analyze the field of elite education in Sweden by revealing the structural interplay of study orientations, educational capital, and gender of upper secondary school pupils in Uppsala. The third paper, written by Johs. Hjellbrekke & Olav Korsnes, applies MCA and class-specific MCA to survey data in order to analyze structural homologies within the Norwegian field of power. Jörg Blasius & Jürgen Friedrichs show how lifestyle items, as originally formulated by Bourdieu in La Distinction, and reformulated to apply to contemporary German culture, can be used to differentiate old and new inhabitants of a neighborhood of Cologne. Using data from a dwelling panel, the authors show how lifestyle patterns can change over time in the process of gentrification. Nora Waitkus & Olaf Groh-Samberg analyze capital portfolios and accumulation strategies in Germany using latent class analysis. Based on disaggregated measures, they construct a ‘space of capital’ and analyze class mobility over time. The paper by Stine Thidemann Faber, Annick Prieur, Lennart Rosenlund & Jakob Skjøtt-Larsen discusses, using as an example class structure in contemporary Denmark, five ways to apprehend classes: as structures of distribution, as forms of the habitus, as symbolic boundaries, as symbolic structures of domination, and as consciousness or identity. Nicolas Robette & Olivier Roueff discuss cultural domains and class structure by assessing homologies and cultural legitimacy. The authors test several hypotheses related to the social differentiation of lifestyles, and examine whether a structural homology exists between the entirety of social and cultural spaces. Starting from the observation that, today, reproduction strategies are no longer limited to national boundaries, but rather operate across borders, Martin Munk focuses on the intergenerational reproduction of transnational and national cultural capital, and shows how educational reproduction and mobility occur within the global space of university institutions. The concluding paper of section A, by Magne Flemmen, Vegard Jarness & Lennart Rosenlund, investigates class, lifestyles and politics. Referring to Bourdieu’s homology thesis, in their empirical analyses the authors reveal structural similarities in both the space of lifestyles and the space of political stances.

Social Fields

Section B comprises the conceptualization of societal spheres within the social space, the social fields with relative autonomy. Social fields such as the academic field, the cultural field, the economic field, and the field of the labor market will be discussed in detail.

The section starts with a contribution from David Swartz on the use of Bourdieu’s field concept in English-speaking literature. The contribution illustrates the extensive and widespread use of the concept of field and its appurtenant analytical concepts in current English-speaking social sciences. Richard Münch gives a comprehensive summary of his research on academic capitalism, utilizing Bourdieu’s conceptualization of the scientific field and the transformations it has undergone in the last decades. Christian Schmidt-Wellenburg shows how internationalization affects the position-takings of German-speaking economists on crisis issues by reconstructing the field of economists, its historical changes, and the economists’ involvement in academic, political, and professional practices. Applying and differentiating Bourdieu’s concept of cultural capital, Myrtille Picaud, Jérôme Pacouret & Gisèle Sapiro construct a social space of the public of a literary festival using MCA, thus gaining insights into the structured interplay of the literary market and the corresponding audiences. Håkan Forsberg, Mikael Palme & Mikael Börjesson discuss education as both field and market, using as an example upper secondary schools in Stockholm. According to their findings, the educational market created in upper secondary education through state intervention resulted in a large expansion of the supply of education in terms of schools and study programs. Michael Gemperle analyzes work orientations using the example of nursing, applying the concept and methodology of the social space paradigm. Using specific MCA, he analyzes how the work orientations of nurses correspond to their position within the occupational group’s social space. Tobias Dalberg analyses the dynamics of inequality in the field of Swedish human scientists by comparing two cross-sections: human scientists who held a position at Swedish universities and university colleges in 1945 and in 1965. Christian Baier & Andreas Schmitz conclude the section by proposing a Bourdieusian approach to institutional fields, using the example of German universities. Applying Multiple Factor Analysis, they show how this institutional field changed from 1995 to 2012 in reaction to heteronomous intrusions.

Methodology and Methods

Section C focuses on methods within the Bourdieusian framework, and includes both quantitative and qualitative contributions. A particular focus is placed on further developments of GDA, such as analyzing temporal and process data, and analyzing (sub-) classes within the social space.

The section starts with a paper written by Brigitte Le Roux & Frédérik Cassor studying changes over time, using GDA methods (MCA and clustering) applied to the French Barometer of Political Trust. Using different techniques within the GDA framework, they demonstrate the advantages of visualizing dynamic structures. Building on a Bourdieusian approach to the social, Fionn Murtagh focuses on the geometry and topology of data and information in the analysis of processes and behaviors using three examples: the challenges and opportunities presented by big data analysis, the narrative analysis of sentiments, and the sphere of mental health and depression. Using the European judicial field as an example, Frédéric Lebaron & Philippe Bonnet discuss class-specific MCA, a further development within the framework of GDA that allows scientists to analyze the spatial structure of sub-groups within an overall social space. Rainer Diaz-Bone & Katharina Manderscheid reflect on how to establish correspondence analysis in sociology in German-speaking universities. They argue that for German university researchers who adopt a Bourdieusian sociology, GDA should be a standard procedure; however, this is not yet the case. Comparing the social and cultural spaces of France, Norway, and Switzerland, Dominique Joye, Gunn Elisabeth Birkelund & Yannick Lemel compare approaches to nominal, ordinal, and metric data within the GDA framework, and declare themselves in favor of canonical correlation as a general analytical approach. The final paper, written by Heinrich Schäfer, Leif-Hagen Seibert & Adrian Tovar Simoncic, proposes a qualitative methodology derived from Bourdieu, consisting of a conceptualization of the habitus as an analytical third layer of ‘socioanalysis’, and as a promising complement for the statistical construction of social spaces and fields using GDA.