The Institution of Sociological Theory in Canada.

Guzman, Cinthya ; Silver, Daniel

FROM ITS FOUNDATION TO THE PRESENT, sociology has been keenly concerned with the state of sociological theory and its position within sociology at large (Camic, Joas, and Levine 2004; Levine 1995). The boundaries of theory have always been contested; and while sociologists share a discipline, they seem to rarely agree about what it means to do so (Michalski 2016). In the U.S. context, recent debates have questioned the purpose and relevance of "theory" to a sociological education, and further, what this means for the role and function of the "theorist" within the discipline (Lamont 2004; Lizardo 2014). At the same time, ongoing debates seek to define what a Canadian sociology should stand for in comparison to its American or European neighbors, and the theoretical orientation that ought to guide the practices of Canadian-based sociologists (Gingras and Warren 2006). The debates about intellectual cohesion and fragmentation have been ongoing for decades, and demonstrate no signs of nearing a clear resolution.

Although much scholarship has outlined similar debates, there is little empirical research on the state of contemporary education in sociological theory and its practices. (1) While a complete account would require studying theoretical practice across various substantive fields, as a crucial first step we examine courses primarily devoted to sociological theory. To do so, we analyze theory syllabi, theory instructor characteristics, and departmental data from sociology departments in Canadian universities. In particular, we highlight the position of theory within the sociological curriculum, and how it varies among universities. We show commonalities and differences in how theorists define themselves as researchers, in reference to one another and to the general population of Canadian sociologists. We highlight the key discursive construct at the core of theoretical education in Canadian sociology--the classic--and trace its ramifications and permutations across the curriculum. We also ask whether and where there are distinctively Canadian sources in theory courses, how theory education varies between sociology departments within and outside Quebec, and touch on the factors that lead to these differences.

Our analyses utilize three sources of information about (1) departments, (2) instructors, and (3) syllabi. Taken together, these show that theory remains deeply institutionalized at the core of sociological education and Canadian sociologists' self-understanding; that theorists as a whole show some coherence in how they define themselves, but differ in various ways, especially along lines of region, intellectual background, and gender; and that despite these differences, the classical versus contemporary heuristic largely cuts across these divides, as does the strongly ingrained position of a small group of European authors as classics of the discipline as a whole. Nevertheless, who is a classic remains an unsettled question, alternatives to the "classical versus contemporary" heuristic do exist, and theorists' syllabi reveal diverse "others" as potential candidates. Our findings show that the field of sociology is neither marked by universal agreement nor by absolute division when it comes to its theoretical underpinnings. To the extent that they reveal a unified field, the findings suggest that unity lies more in a distinctive form than in a distinctive content, which defines the space and structure of the field of sociology.

MULTIPLE TRAJECTORIES IN CANADIAN SOCIOLOGY

Although Canadian sociology begins as early as the 1920s (Brym and Fox 1989:17), it was not fully formed until the 1960s (Curtis 2016; McLaughlin 2005:6; Stokes and McLevey 2016). Canadian sociology is often discussed as "foreign import, a product of the European Enlightenment, French Revolution, [...] and industrial capitalist dislocation" (Curtis 2016:205). Sociology in Anglo-Canada was spearheaded primarily by Porter, Dawson, Hughes, Clark, and Innis, and in Franco-Canada by Gervin and Levesque (McLaughlin 2005:6; Stokes and McLevey 2016). Separate paths unfolded (Warren 2014), producing a "wall of silence" or "canyon" between Anglo and Franco Canada (Rocher 1992). Thus, any empirical research interested in understanding how theoretical practices of Canadian Sociology differ must take seriously the boundaries between Quebec and the rest of Canada (ROC).

In Quebec, sociologists have negotiated their position within international disciplinary currents and intra-Canadian cultural politics for nearly 80 years. First founded in Universite Laval in Quebec City in the 1940s, American and French sociology were the formative influences, even as Quebec sociologists were actively engaged in local political and cultural issues (Curtis 2016:205). In the 1950s to the early 1960s, the Chicago school and Parsonian structural functionalism influenced the entirety of sociology, Quebec included (Brym and Saint-Pierre 1997). The 1960s, however, marked a shift: after the Quiet Revolution sociologists turned to the survival of Quebec's Francophone community and the rapid modernization of society (Brym and Saint-Pierre 1997:545). By the 1970s, these differences had grown to the point that Canadian sociology began to split into two distinct communities, with Quebec sociologists more actively participating in their province's political and social institutions.

Within the ROC, sociology has also unfolded at the intersection of local, national, and international currents, with the 1960s forming a crucial turning point. An influx of American-trained professors brought commitments to functionalism, positivism, and the New Left (Brym and Saint-Pierre 1997:545). John Porter's Vertical Mosaic began to strongly shape the discipline's "explanatory stance" (Stokes and McLevey 2016:177), and by the 1970s, Innis and Marx had become central inspirations (Luxton 2006). For some observers, Anglo-Canadian sociology came to focus on uniquely Canadian social developments, and less on the global forces impacting Canadian societies (Brym and Fox 1989). In the 1980s, feminist movements, methodologies, and women-centered issues became more influential, bringing "new schools of thought and the decline of old orthodoxies" (Brym and Saint-Pierre 1997:545; Luxton 2006). Postmodernism, interpretative and qualitative methods, and a new-found skepticism toward quantitative methods emerged more forcefully (Brym and Saint-Pierre 1997:546). The result is a highly pluralistic field in which multiple methods and epistemologies cohabitate.

Despite cultural, regional, and historical differences, the entirety of Canadian sociology is often seen as committed to a form of public sociology (Burawoy 2009). Yet, this common orientation often rests on divergent grounds, making it difficult to construct a national project. McLaughlin (2005) believes that this division within Canadian sociology weakens the sense of what Canadian sociology stands for, its foundational influences, and its ability to confidently participate in international theoretical dialogue. For instance, in Anglo Canada, there is, he claims, a "weak sense of sociology as a craft with distinctive knowledges, [and] skills" (McLaughlin 2005:7). In Quebec, by contrast, some commentators observe that sociology is often concerned with paying tribute to its intellectual traditions in the hopes of preserving them for the future of the discipline (Fournier 2001). In sum, Canadian sociology is multiplex and plural, and studying how it is practiced requires sensitivity to variation and context, commonalities, and differences--in political commitments, intellectual assumptions, and international reference points.

SOCIOLOGICAL THEORY AND DISCIPLINARY IDENTITY

If it has been difficult for Canadian sociology to cohere around a collective project and shared history, finding agreement about a common theoretical orientation has been equally difficult. Many observers note clear regional differences. Brym and Saint-Pierre (1997) for instance believe that theory outside of Quebec is oriented more toward "intellectual radicalism without a concrete policy orientation" (p. 543). By contrast, sociological theory in Quebec is more concerned with formulating public policy (Brym and Saint-Pierre 1997) and maintaining intellectual traditions. Various common vantage points have been proposed to bring unity to the discipline, such as "critical and leftist tradition" (McLaughlin 2005:4), critical realism (Carroll 2013), or a renewed form of staples theory (Matthews 2014). None has carried the day, however, and new rounds of criticisms and responses follow (Puddephatt and McLaughlin 2015; Stanbridge 2014; Tindall 2014).

The form of these debates indicates the central place that theory has in the reproduction and self-understanding of the discipline. While there is little agreement about which theoretical orientation should define the discipline, there is general agreement that whatever could do this definitional work must involve some form of theoretical understanding--and perhaps more importantly, that the debate itself should take place on the level of theoretical discourse about the meaning and purpose of sociology. In sociology, theory seems to play a crucial role in drawing the boundaries of the discipline, and in providing a common reference point to its members. Disciplines draw their boundaries in many ways (Bourdieu 1988). Some do so through a shared method, such as archival research in history or formal modeling in economics (Lamont 2009), or a common subject matter, such as plants in botany or brains in neuroscience. But where some draw their boundaries using methodological distinctions and others use subject matter, sociology arguably defines itself through a "canon" of knowledge or texts that they consider critical both to their identity as a discipline and to the curriculum that they teach students (Baehr 2017).

While a disciplinary understanding rooted in a "canon" serves a unifying function, "canons" also may become a place of contention. Specifically, the question of who and which theories are included in the canon can lead to struggles over disciplinary identity. However, tensions that emerge from this type of boundary work are not only evident in sociology. Other notable cases include controversies in English literature over which authors are to be considered "classics," and in History about which events and people should be included in Western Civilization courses (Bastedo, Altbach, and Gumport 2016). Similar controversies unfold in less public view, but are nevertheless quite contentious within academic circles. Such debates turn on questions around what kind of research to reward and discuss, but also come to a head around controversies concerning who and what to teach students.

The classics of sociological theory specifically provide a shared reference point around which methodological and epistemological debates can play out in sociology (Alexander 1987; How 2016; Levine 1995). This too--the discursive centrality of "classical theorists"--is relatively distinctive to sociology. For example, in political science, theory is often divided between ancient (e.g., Plato and Aristotle) and modern (e.g., Hobbes, Locke, Rousseau); in economics, between micro and macro; in physics between classical mechanics and quantum theory. Sociologists may disagree on what their theoretical orientation should be, but whatever it is, there is a good chance they can find an argument for it in Weber, Marx, or Durkheim. The coherence and the stability of the discipline are marked by the coherence and stability of its reference to a core group of classical theorists, rather than any allegiance to particular methodology or topic.

Because the discipline's classical authors have been interpreted as "theorists" (Bargheer 2017), sociological theory becomes perhaps the central venue in which general questions about disciplinary identity unfold that cut across subfields. Advancing a new paradigm often means demonstrating its core ideas were already implicit in the classics, as in the case of efforts to advance the sociology of emotions by reinterpreting the classics as theorists of the emotions (Shilling 2002). Here, the existing classics are the setting for interpretative debates about sociology's legitimate topics.

Given the disciplinary centrality of classical theory, theorists and theory courses have a special place in the discipline. This special status is often understood to emanate from their status as "founders" (Baehr 2017). "Founders" anchor sociology's disciplinary identity in charismatic figures understood as fountainheads of the field's major institutional and intellectual traditions. Despite legitimate concerns about the coherence of the very notion of an intellectual or institutional disciplinary "founder" (Baehr 2017), the discursive centrality of the notion grants the question of who is and is not a "founder" a corresponding urgency. This was evident already when Parsons included in his pantheon of "recent Europe authors" Weber, Durkheim, Marshall, and Pareto--but not Simmel or Marx (Alexander 1989; Levine 1989; Parsons 1949; Sciortino 2001). Parsons evidently excluded Simmel because he considered Simmel's brand of relational and pluralistic sociology to be pernicious to his own synthetic vision (Levine 1989). The subsequent effort to elevate Simmel to "foundational" status was in many ways an effort to elevate relationalism as a theoretical alternative. Likewise, the tumult of the 1960s and 1970s issued in a "new" founding figure: Karl Marx. Current debates about WEB Du Bois and the Chicago School (Morris 2015) reveal similar dynamics and tensions (Dodd 2017). Advocates advance Du Bois as a true but historically excluded founder of American sociological theory, and thereby hope to integrate concerns about race into the very basis of sociological thought (Loughran 2015). While they seek to admit a new founder into the canon, the basic structure of a discipline defined by canonical founders persists.

If the theory course is of particular relevance in examining the structure of the discipline, so is the theorist. In line with the religious overtones of the word "canon" (Baehr 2017), theorists in sociology are in some ways comparable to a priestly class: custodians and guardians of a tradition, tasked with teaching new generations its customs, rituals, and signature styles of thought and speech, perpetually debating the subtle meaning of its sacred texts. If (this understanding of) the role of the theorist is in fact losing its grip--as Lizardo (2014) and Lamont (2004) suggest--then so too would this way of forging disciplinary identity. As Durkheim might well say, a religion without a priest is no religion at all.

RESEARCH QUESTIONS AND DATA

These observations guide the questions that inform this study, and the data we collected to answer them. We pursue five major questions.

The first concerns the position of theory within the sociology curriculum in Canada. How deeply and broadly is theory institutionalized in sociological education, and what typical forms does it take? To address this question, we compiled information about course offerings in the 64 Canadian sociology degree-granting departments that offered a minimum of a four-year Bachelor of Arts degree. For each university, we documented its location, the types of degrees it offers (B.A., M.A., Ph.D.), and its university type according to Maclean's Magazine (primarily undergraduate, comprehensive, medical-doctoral). We classified each course as required or elective, and classical, contemporary, general, or specialty. "General" courses are typically a department's theory courses that combine classical and contemporary; "specialty" courses generally cover specific topics, such as a single author, or cultural, feminist, queer, and postcolonial theory. Table 1 summarizes basic information about the departments included in our study.

Our second set of questions is about those who teach theory courses in sociology. Do they understand themselves as theorists? How shared is their self-understanding, and in what ways do their intellectual orientations diverge? To answer these questions, we gathered information about instructors who taught theory courses from 2012 to 2015 (N = 259). Using publically available information on university websites, LinkedIn, and CVs, we documented each instructor's research interests (following Lamont 2004), rank, gender, Ph.D. granting institution, and current institution. To provide a comparative baseline, we gathered similar information about the general population of Canadian sociologists as well.

Figure 1 summarizes basic information about sociology and theory instructors in contemporary Canadian sociology. Compared to the discipline at large, the population of theory instructors has a somewhat higher proportion of men, Canadian Ph.D.s, and French Ph.D.s, and a somewhat lower proportion of U.S. Ph.D.s. Additionally, about 60 percent of theory courses are taught by associate and full professors, with assistant professors and sessional instructors covering the rest.

Figure 2 maps Ph.D. locations separately for instructors inside and outside of Quebec. Theory instructors in Quebec were far more likely to receive training in Quebec and France, whereas instructors in ROC were more oriented toward the United States and the United Kingdom.

Third we ask questions concerning the content of theory courses. Is there a generally shared canon, and in what ways does its composition vary? What are the major intellectual traditions from which Canadian theory courses draw? To answer these questions, we compiled a database of sociological theory syllabi, for 2012 to 2015 (N = 285). (2) We extracted from these a variety of information: authors taught, period covered, course description and objectives, and the number of authors assigned per syllabus. Table 2 and Figure 3 show some basic descriptive statistics. The typical syllabus lists around 10 authors, and about half of the theory courses offered in Canada include Marx, Weber, and Durkheim.

Our fourth and fifth questions probe our database of syllabi further. We ask specifically about the division of theory between "classical" and "contemporary": How does the classical-contemporary heuristic structure the institutionalization of theory (1)? We ask questions about the extent to which there is a distinctively Canadian form of sociological theory by examining the nationality of authors listed in theory syllabi. How often, and in what contexts, are Canadian authors assigned in sociological theory courses? To answer this question, we assigned a nationality to all authors in our database. Figures 4 and 5 map the number of authors from each country.

To analyze these diverse data sources, we use a variety of methods, such as basic descriptive statistics, computational text analysis, network analysis, and correspondence analysis. Specific methodological details emerge in the course of our analysis.

ANALYSIS

Question 1: How Deeply and Widely Is Theory Institutionalized?

Given the divergent historical trajectories and self-understandings that characterize Canadian sociology, one might well wonder if the discipline possesses a broadly shared and reproducible identity. A simple indicator of how broadly a practice is integrated into a given institution is how regularly it occurs across time and space. A more deeply integrated practice is one that members of an institution must observe. Regularity and necessity together are signs of a practice with a high degree of normative authority.

On these indicators, theory is strongly and widely integrated into the institutional framework of the Canadian undergraduate sociology curriculum. All Canadian sociology departments offer some type of theory course. And all but two departments require their students to take a theory course (and those two require theory for their honors B.A.). Despite the deep and enduring divisions across Canadian sociology discussed above, the discipline does possess a widely shared practice with relatively unquestioned normative authority: theory (Table 3). (3)

If theory is central to the common institutional practice of sociology, theory also strongly defines the discipline's collective content: the substantive texts and ideas that any sociologist would be expected to know. Most students of sociology must encounter the classics. Of the 64 undergraduate programs, 36 require a course explicitly identified as classical theory. When departments do not require classical theory, they nevertheless usually require a general theory course instead (14 programs). Still, these courses typically cover the same core authors (--75 percent include Marx, Weber, and Durkheim). Thus, not only does theory in general define a core feature of the discipline's identity, it also structures that identity according to a particular heuristic: the classics.

Quebec does not differ substantially from the ROC in terms of the position of the classical theory course. Half (four) of the departments in Quebec require a course in classical theory. Where Quebec departments do stand out to some degree is in offering and requiring more contemporary and specialty courses, often on individual authors, traditions, or problems, such as Marx and Marxism, Weber, Durkheim, functionalism, democratic theory, identity, systems theory, or globalization. Three of eight Quebec departments require specialty courses of this type, whereas only 13 percent of departments in the ROC do. The difference becomes more evident if we look to our syllabi database, and compare types of courses offered across regions.

Between 2012 and 2015, Quebec departments were considerably more likely to offer specialty courses in theory. While the numbers are not large--there are only eight departments in Quebec, after all--they do at least suggest some evidence of a hybrid model there. If in the ROC, theory courses primarily provide the classical grounding of a general sociological education, in Quebec theory is more strongly institutionalized as a distinct specialty area. Students do not only take introductory survey courses, but are more often expected to undertake close studies of particular theorists or theoretical orientations, receiving specialized training on the way toward adopting the institutionally recognized role of "theorist." In the ROC, as we will see, while many theory instructors present themselves as theorists, there is little formal training into this role for their students. This model of theorist-as-specialist is more in line with French practice, and may account for some of the regional differences we find below.

Even so, the classics remain the key discursive pivot around which theoretical education in Canadian sociology turns. No doubt few Canadian sociologists would be surprised to discover that classical theory is a core disciplinary requirement. Yet, it is worth reflecting on the implications of this taken-for-granted arrangement, since theory in general--and classical theory in particular--is by no means the most common or obvious organizing disciplinary principle.

Defining theory in terms of classics has at least two immediate consequences. First, it automatically generates a residual category, theory that is not classical. In Canada, nonclassical theory primarily is taught under the heading "contemporary," the second most commonly required type of theory course. Contemporary theory courses, as we will see in more detail below, are much less coherently patterned than classical are. They also exhibit widespread confusion about what it means to be "contemporary"--a sign of their residual status. Second, as elaborated above, the classical theory heuristic makes the question of who is taught in a classical theory course a major source of symbolic struggle in the discipline. As authors migrate from mere contemporaries to classics, they become consecrated; placing a relatively obscure author next to an established classic allows the former to bask in the halo of the latter. Adopting "the classic" as a basic structuring principle builds these dynamics into the very heart of the discipline. We follow the ramifications of these tensions in more detail below.

Question 2: How Shared or Divergent Are Theorists' Self-Understandings?

The institutionalization of a practice exists not only in its regular and mandatory occurrence. It also depends on participants publically defining themselves by its logic. A widely shared, publically avowed self-conception is a mark of a collective practice with the normative authority to define an identity.

To examine sociological theorists' public self-definitions, we look to their publically declared research interests. While these statements are by no means rich evocations of subjective experience, they do have certain methodological advantages. They are public. They are widely available: we were able to collect data for 87 percent of all theory instructors and for 84 percent of all sociology instructors, a much higher response rate than a survey would yield. They are produced in the course of sociologists' normal professional activities, rather than in response to a research inquiry. Hence, they mitigate response bias. And because they are generated both by theorists and the general population of sociologists, we may compare the self-definition of the former to the latter to discern what is and is not distinctive to theory instructors.

We use some basic techniques of computational text analysis to investigate the role of theory in the self-definition of Canadian sociology. Cosine similarity is a common metric used to compare corpuses in text mining. It shows the degree to which words in two sets of documents tend to occur at similar rates, and ranges from 0 to 1. The cosine similarity between sociologists as a whole, and theory instructors is .91. This high value suggests that theorists do not markedly differ from the rest of sociology in how they define their research orientations publically. This shared style of speech is another indicator of a generally shared disciplinary culture, despite differences in what the content of that culture ought to be.

The specific role of theory becomes more apparent if we look at the most common words sociologists use to describe their research interests. Figure 6 is a slope graph comparing the 20 most frequent words used by all sociologists with those most commonly used by theory instructors.

Both sociologists in general and theorists in particular frequently use "theory" in their self-description, though the latter do so more often (13th and 3rd, respectively). This finding contrast sharply with Lamont (2004), who based her claim on the declining disciplinary commitment to theory on the fact that relatively few professors who teach theory make reference to theory among their primary research interests. In Canada, at least, theorists and sociologists in general make such a reference quite routinely.

The slope graph also indicates how the two groups differ. Theorists stand out as the keepers of the classics, listing an interest in "classics" far more frequently than does the average sociologist (over 100th). In line with Lamont's (2004) findings, theorists often pair their theoretical interests with interests in history, politics, and culture, which she calls "theory satellite areas." Other areas tend to move in Canadian theorists' orbits, such as gender, media studies, and social movements. Compared to sociologists as a whole, theorists showed less interest in the sociology of labor, education, race/ethnicity, and policy. Overall, "theory" remains a core part of Canadian sociologists' self-definition in general. Theorists in particular tend to define themselves as custodians of the canon and via interests in areas with a strong symbolic and interpretative component, such as culture, media, and politics.

Examining how theorists in particular describe themselves helps to reveal the commonalities and cleavages in their theoretical orientations (Figure 7). The left graph compares theorists from Quebec versus non-Quebec universities; the right compares women and men. Both groups show high degrees of similarities to one another. The cosine similarity for Quebec versus ROC is .87, for men versus women .89. Whatever their other differences, Canadian theorists describe their interests in similar terms.

The graphs make some cleavages within the theory community evident, somewhat in line with the debates reviewed above about divergent regional trajectories and the impact of feminist thought (the major differences we flag are statistically significant). Quebec theorists, for example, show a markedly more prominent interest in "Quebec," and in "current" topics. Perhaps by the same token, Quebec sociologists more rarely avow an express interest in "classical" sociology. Women's interests more frequently include terms such as "feminist" and "women's", and women are more likely to describe their work as an "exploration" (not shown, p < .05). Men by contrast are more likely to use the term "science."

These results reveal some signs of ongoing cleavages in the research orientations of Canadian theory professors. Theorists differ (by region and gender) in their avowed interests in the Quebec national project, feminism, and "science" as an appropriate way to describe the sociological endeavor, and perhaps in their commitment to the contemporary relevance of theory. But these cleavages subsist within a broadly shared set of interests, in which theory, culture, history, and critical thought are common modes of self-definition.

Question 3: What Canonical Authors Define the Discipline, and How do These Vary?

Our first two questions addressed the organizational arrangements and personal self-definitions that sustain the institutionalization of theory in the discipline. We now turn to the shared content of theory courses, and examine the authors most frequently assigned in theory courses. This too highlights a key aspect of institutionalization: a practice is more deeply institutionalized to the extent that it provides members something they can all talk about, and expect one another to know about. Theory courses are crucial vehicles for creating these shared expectations among sociologists.

We use network analytic techniques to examine patterns of author assignment across syllabi. We treat authors (e.g., Marx, Foucault) as nodes, and consider two authors connected when they appear on the same syllabus. Strongly connected authors indicate regularly recurring patterns of coassignment. First consider a relatively schematic representation of a figure showing the entire field of sociological theory in Canada.

Figure 8 indicates the overall pattern of assignments in theory courses. There is an extremely dense core, in which the same authors are repeatedly assigned with one another. Around this core various satellites orbit: authors who appear together in a small number of courses, which are bound to the overall field through sharing a few more commonly assigned figures. Viewed from this high altitude, sociological theory courses exhibit a remarkably coherent common core.

The specific authors at the center of this network are not surprising: Weber, Marx, and Durkheim. Network measures allow us to codify the scale of this centrality, and identify other central authors as well. To do so, we use Kleinberg's authority centrality score. Higher "authority" scores indicate nodes that are linked with many others (Figure 9).

The graph makes evident the special role the "Big 3" "founding figures" of Weber, Marx, and Durkheim plays in the discipline. There is a clear gap between their authority and that of "everybody else"--as clear evidence for an undisputed canon as anybody could wish for. This centrality is even more decisive in classical courses, with roughly 85 percent of them including the Big 3. If anything, this number underestimates their centrality. When an individual classical course does not include the full triumvirate, it usually includes one or two of three, often as part of a two-semester sequence in which for instance Marx might be assigned in one and Weber and Durkheim in the other.

While the authority of the Big 3 is undisputed within classical courses (regardless of region) and in the overall network, it does vary by region, as do the authors in the next tiers of authority (Figure 10). The authority scores outside Quebec largely correspond to those for the country as a whole. Quebec differs, however, in form and content. In content, Quebec theory courses clearly feature French authors more predominantly. The ROC by contrast features more American and Anglophone Canadian authors, such as Dorothy Smith, WEB Du Bois, C. Wright Mills, and Robert Merton (though Parsons' influence cuts across the two regions). Thus, despite a stock of shared reference points from the Big 3 to Foucault, Mead, Goffman, and Bourdieu, the historical and current differences in the regions' international intellectual connections to France, the United States, and the United Kingdom means they place that common stock in somewhat different dialogical contexts.

In terms of form, the graphs have different shapes, indicating a different structure. Although in the ROC the Big 3 occupies its own tier, Quebec authority follows a more continuous gradient. This largely reflects the fact that, as we saw above, in Quebec a majority of theory courses are on specialty topics, a relatively smaller proportion are in classical theory as such, and a relatively larger proportion cover contemporary theory. The result is that core contemporary authors such as Bourdieu and Foucault suffuse the Quebec curriculum: they sometimes appear in classical courses, almost always in contemporary, and quite often in various specialty courses. By contrast, in the ROC, they are primarily segregated into contemporary courses, and do not even appear in those uniformly.

The ROC pattern largely corresponds to the structure of theoretical education in the United States, where the classical versus contemporary heuristic predominates, and a classical course without the Big 3 is nearly unthinkable. The Quebec pattern largely reflects its more hybrid character. Aspects of the American pattern mix with a more French one, in which the theorist is a more clearly delineated social role, and is more attuned to contemporary continental intellectual currents. These curricular differences are likely grounded to some degree in the different intellectual background of theory instructors noted above.

Question 4: How Does the Classical-Contemporary Heuristic Structure the Institutionalization of Theory?

While regional differences account to some degree for variation in how deeply theory is institutionalized according to the classical-contemporary heuristic, the heuristic itself imposes its own normative order onto the practice of theory in sociology. We can get some of this order into view by mapping the network of authors assigned in classical and contemporary courses, separately.

Figures 11 and 12 show the classical and contemporary networks. They weight node sizes by the degree centrality of each author, and edge thickness by their edge weight. For legibility, only edges with at least a weight of five are shown.

The figure again shows how strongly the Big 3 dominate the definition of classical sociological theory. They are the triune sun around which the entire system orbits. But the graph also shows multiple potentially competing logics defining how the classics are understood, depending on their "others." (4) Many courses seem to treat classical sociology as itself the outcome of a presociological ferment, assigning Marx, Weber, and Durkheim along with various philosophical "protosociologists," such as Hobbes, Comte, and Rousseau, or Smith, Kant, and Locke. These adopt what we might call a foundationalist approach, highlighting the transformation of general philosophical insight about social life into a discipline especially devoted to that topic. Other courses adopt what we might call a "perspectives" approach. They feature alternative classics that offer points of view that cross boundaries of gender and race, such as Du Bois, Gilman, or Martineau. Others highlight interactionism, bringing Mead, Simmel, and Cooley in as contrasts to the largely individualistic, structuralist, and historical-materialist orientations of the Big 3.

While further and alternative analyses might reveal other patterns, (5) the major principle governing classical theory is clear. The Big 3 provide an unquestioned foundation to the disciplinary notion of what a classic is. Alternatives exist around the margins, and a theory professor has some discretion when it comes to defining what they believe it means to be "marginal," typically by gender, race, or theoretical orientation. Depending on the proto-sociologists included, a professor may place the foundation of the discipline in various narratives: as an outcome of philosophical reflection about the nature of categories, an activist project of social and political critique, an extension of the liberal tradition of political thought, and more.

Contemporary theory exhibits a somewhat different overall pattern. Overall, contemporary theory exhibits less coherence around a common core than classical does. (6) Foucault and Bourdieu are at the center of the network, but their authority as the definition of "contemporary theory" is less uncontested: whereas roughly 85 percent of classical syllabi include the Big 3, only roughly 50 and 65 percent list Bourdieu and Foucault (respectively). Similarly, the authority scores of Bourdieu and Foucault are not as distinct from the rest of the field: Goffman, Habermas, Giddens, and Dorothy Smith have scores around .7 or higher (whereas in classical other than the Big 3 only Simmel's is above .5, around .6).

The specific authors assigned also reveal different conceptions about what is important in contemporary theory. Some courses feature what we might call the question of modernity, spanning authors from traditional modernization theory (Parsons, Merton), the Frankfurt School critics of modernity (Habermas, Adorno, Horkheimer), and postmodernists (Lyotard). Others highlight interactionism and phenomenology especially: Mead, Garfinkel, Goffman, D. Smith, Blumer, and Collins. Others adopt a more avowedly radical stance, including critics of colonialism, empire, and heteronormativity, such as Said, Gramsci, and Butler.

What is most striking about contemporary theory, however, is the structural confusion it exhibits regarding what it means to be "contemporary." Of the 20 authors with the highest authority scores, only 15 "contemporary" theorists are still alive. Some have been dead for over half a century, such as Mead, Fanon, or Mills. Several authors are sometimes listed as classical, sometimes as contemporary. For example, Foucault appears on roughly 10 percent of classical courses, Mead on 15 percent of contemporary. Marx, dead and buried for over 130 years, appears on almost 20 percent of contemporary theory syllabi. This is a perplexing notion of "contemporary" to say the least.

This situation becomes somewhat clearer when we treat "contemporary" not in terms of its manifest content, but via its structural meaning, as the opposite of "classical" in a semiotic binary (Alexander 1987). The meaning of "contemporary" depends on what it means to be "not classical," which in turn depends on what it means to be classical (Levine 1995). For example, if classical means "founders," then "not classical" can mean quite generically: anything not formative of the discipline. If classical has a humanistic meaning in which classics are great texts to be read closely for the profound insights contained therein, then "not classical" encompasses more transitory texts that cannot sustain such intense engagement. If classical means generically "of the past," contemporary means "relevant to understanding the present situation."

Other meanings may be at play, but these examples are enough to illustrate how "classic" and "contemporary" mutually define one another, with the former generally functioning as the more active partner in the binary. Because "classical" possess such (positively) charged significance in this system, its definition correspondingly defines symbolic struggles to acquire its halo. If classical means "foundational," then the struggle is about who is the true founder. Similarly, when classical is defined humanistically, authors become classics when their works are read exegetically and interpretatively. Foucault and Bourdieu show signs of moving in this direction, especially in Quebec. If classical however means "old" and contemporary means "relevant," moving an author into the latter category can acquire a significant meaning, as a statement for instance of the ongoing value of Marxian thought for analyzing current social problems.

Question 5: Is There a Distinctively Canadian Sociological Theory?

Debates about the distinctiveness of Canadian sociology often turn on whether there is a distinctively Canadian approach to sociology. Considered in the empirical light of this paper, this question could be taken up in at least two ways: in terms of form or content. In terms of content, we can ask if and when Canadian theory courses include Canadian authors. In terms of form, we can ask if the origins of authors assigned in Canadian theory courses exhibit any distinctive patterns.

Figure 13 shows countries from which at least 3 percent of authors assigned hail. While authors from the United States predominate in Canadian syllabi, Canadian authors comprise a sizable proportion, around 10 percent. It is unlikely that Canadians are assigned with similar frequency in other countries, and this does suggest that overall Canadian sociology does give special weight to Canadians' ideas (Figure 14).

The distribution of authors across syllabi, however, is strongly patterned by their national origin, along two major axes. The first and strongest differentiates German and Canadian authors by type of course. German authors, while comprising only around 3 percent of all authors, are strongly clustered in classical courses and to some degree in general theory courses, but appear outside of these relatively rarely. Canadian authors, though they make up around 10 percent of all authors, appear in classical courses exceedingly rarely. Though they do appear to some degree in contemporary courses (e.g., D. Smith, Goffman), they are most common in specialty courses--suggesting that to the extent that the debate (reviewed above) about the need for a distinctively Canadian sociology has issued in the study of Canadian theorists, it has done so largely in specific specialty areas. To a large degree, the classical core stands outside this debate. The second axis involves authors from the United States and France. Professors from the United Kingdom and United States are most likely to assign U.S. authors, while those teaching in Quebec and with French Ph.D.s tend to assign French authors. Canadian Ph.D.s are in the middle.

Thus, while Canadian authors do hold a fairly prominent position in the curriculum, they do not define the classics in Canadian theory education--these continue to be grounded in the European origins of the discipline. Yet, this figure may also reveal a more distinctively Canadian form of sociological theory: Canadian theory as (structurally) defined by its position between France and the United States. U.S.-trained sociologists tend to pull the field in one way, toward American authors; French-trained sociologists pull it the other, toward French authors and away from the notion that courses in (mostly German) classics define the discipline; and Canadian-trained sociologists exist in the space between. While it may be difficult to isolate a distinctively Canadian set of sociological topics or a canon of Canadian theorists, perhaps what is truly distinctive is this formal position in the to and fro between continental traditions of thought and disciplinary structures.

CONCLUSION

It is certainly possible to push this research further. One direction runs deeper into sociologists' subjective understanding of what theory is and why they do it. Surveys and interviews can usefully supplement the more objective data analyzed here. Another direction runs wider. We can for instance compare theorizing conventions inside and outside of explicit theory courses, and how such conventions are informed by a professor's own theoretical training and who they studied under during their doctoral studies. We can also compare the Canadian situation to other countries. In this way, we can more precisely determine the extent to which sociological theory education has a broadly shared international form, and identify sources of national patterns of variation. We are pursuing these directions in our ongoing collaborative research.

This paper, however, took a narrower view and sought to bring some empirical evidence to bear on questions about the sources of unity and division in Canadian sociology, highlighting the role of theoretical education. Findings are decidedly mixed. Broadly shared assumptions provide the background for local variation, and common reference points for ongoing debate and controversy. Across Canada, theory is deeply institutionalized in the curriculum, a small group of authors is taken for granted as classics, and theorists describe their research orientations in similar terms, often accentuating the centrality of "theory." In Quebec, theory education has a more contemporary orientation geared toward training theory specialists, and is more in touch with currents of continental thought. In the ROC, theory education primarily preserves and transmits the classics. Canadian authors are rarely anointed as classics, and are discussed more in specialty courses often dealing with substantive themes. Confusion reigns as to what it means to be "contemporary," as the strong discursive force of "the classic" defines its other as a residual category. Canadian-trained Ph.D.s find themselves in a tug of war between their colleagues trained in Europe and the United States.

The overall impression given by these results is of a field marked neither by universal agreement nor by absolute division. To the extent that they reveal a unified field, they suggest that unity lies more in a distinctive form than in a distinctive content. That form involves several key structural features: theory as a core requirement; the classics as the core of the core; Canada between the United States and Europe. In somewhat Bourdieusian fashion, we might conclude that the shape of theory helps to define the space and structure of the field. The nature of this structure becomes clear when we contemplate what it would mean to change it. A true revolution would involve not the introduction of a new classical theorist, but the abandonment of the classics as a structuring principle; not a new theoretical orientation but the abandonment of theory itself as a core requirement; not the adoption of more Canadian authors or different European and American authors, but a radical reorientation of Canadian intellectual life outside of its tense position between the two. The current form of the field continues to reproduce itself, however, by continuing to define not only who and what we discuss, but also the unquestioned way in which we do so.

Cinthya Guzman and Daniel Silver

University of Toronto

Cinthya Guzman, Department of Sociology, University of Toronto, 725 Spadina Avenue, Toronto, ON, Canada M5S 2J4. E-mail: c.guzman@mail.utoronto.ca

References

Albers, C. 2003. "Using the Syllabus to Document the Scholarship of Teaching." Teaching Sociology 31(l):60-72.

Alexander, J. 1989. "Against Historicism/for Theory: A Reply to Levine." Sociological Theory 7(l):118-20.

Alexander, J.C. 1987. "The Centrality of the Classics." Pp. 11-57 in Social Theory Today, edited by A. Giddens and J.H. Turner. Stanford, CA: Stanford University Press.

Baehr, P. 2017. Founders, Classics, Canons: Modern Disputes over the Origins and Appraisal of Sociology's Heritage. London: Routledge.

Bargheer, S. 2017. "Anthropology at War: Robert H. Lowie and the Transformation of the Culture Concept, 1904 to 1954." Journal of the History of the Behavioral Sciences 53(2): 133-54.

Bastedo, M., P. Altbach and P. Gumport. 2016. American Higher Education in the Twenty-First Century: Social, Political, and Economic Challenges. Baltimore, MD: JHU Press.

Bourdieu, P. 1988. Homo Academicus. Stanford, CA: Stanford University Press.

Brym, R. and C. Saint-Pierre. 1997. "Canadian Sociology." Contemporary Sociology 26(5):54346.

Brym, R.J. and B. Fox. 1989. From Culture to Power: The Sociology of English Canada. Toronto; New York: Oxford University Press.

Burawoy, M. 2009. "Disciplinary Mosaic: The Case of Canadian Sociology." Canadian Journal of Sociology 34(3):869-86.

Camic, C., H. Joas and D. Levine. 2004. The Dialogical Turn: New Roles for Sociology in the Postdisciplinary Age, edited by C. Camic, H. Joas and D. Levine. Lanham, MD: Rowman & Littlefield.

Carroll, W.K. 2013. "Discipline, Field, Nexus: Re-Visioning Sociology." Canadian Review of Sociology tRevue Canadienne de sociologie 50(1): 1-26.

Curtis, B. 2016. "The Missing Memory of Canadian Sociology: Reflexive Government and 'the Social Science.'" Canadian Review of Sociology / Revue Canadienne de sociologie 53(2):203-25.

Dodd, N., ed. 2017. "2016 BJS Annual Public Lecture (Special Issue)." British Journal of Sociology 68(l):573-770.

Fournier, M. 2001. "Quebec Sociology and Quebec Society: The Construction of a Collective Identity." Canadian Journal of Sociology / Cahiers canadiens de sociologie 26(3):333-47.

Gingras, Y. and J.-P. Warren. 2006. "A British Connection? A Quantitative Analysis of the Changing Relations between American, British and Canadian Sociologists." Canadian Journal of Sociology / Cahiers Canadiens de Sociologie 31(4):509-22.

Goyder, J. 2009. "A Behavioural Approach to Ranking Academic Departments by Prestige: The Case of Sociology in English-Speaking Canada." Bulletin of Sociological Methodology 101(1):28-41.

Grauerholz, L. and G. Gibson. 2006. "Articulation of Goals and Means in Sociology Courses: What We Can Learn from Syllabi." Teaching Sociology 34(l):5-22.

How, A. 2016. Restoring the Classic in Sociology. Traditions, Texts and the Canon. London: Palgrave Macmillan.

Lamont, M. 2004. "The Theory Section and Theory Satellites." Perspectives: Newsletter of the ASA Theory Section 27(1):5.

Lamont, M. 2009. How Professors Think. Cambridge, MA: Harvard University Press.

Levine, D. 1995. Visions of the Sociological Tradition. Chicago, IL: University of Chicago Press.

Levine, D.N. 1989. "Parsons' Structure (and Simmel) Revisited." Sociological Theory 7(1):11017.

Lizardo, O. 2014. "The End of Theorists: The Relevance, Opportunities, and Pitfalls of Theorizing in Sociology Today." Pamphlet Based on the Lewis Coser Memorial Lecture. Presented at the 2014 annual meeting of the American Sociological Association, August 17, San Francisco, CA.

Loughran, K. 2015. "The Philadelphia Negro and the Canon of Classical Urban Theory." Du Bois Review: Social Science Research on Race 12(2):249-67.

Luxton, M. 2006. "Feminist Political Economy in Canada and the Politics of Social Reproduction." Pp. 11-44 in Social Reproduction. Feminist Political Economy Neo-Liberalism, edited by K. Bezanson and M. Luxton. Montreal: McGill-Queen's University Press.

Matthews, R. 2014. "Committing Canadian Sociology: Developing a Canadian Sociology and a Sociology of Canada." Canadian Review of Sociology /Revue Canadienne de sociologie 51(2):107-27.

McLaughlin, N. 2005. "Canada's Impossible Science: Historical and Institutional Origins of the Coming Crisis in Anglo-Canadian Sociology." Canadian Journal of Sociology 30(1): 1-40.

Michalski, J. 2016. "The Epistemological Diversity of Canadian Sociology." Canadian Journal of Sociology 41(4):525-56.

Morris, A. 2015. The Scholar Denied: Web Du Bois and the Birth of Modern Sociology. Berkeley, CA: University of California Press.

Parson, L. 2016. "Are Stem Syllabi Gendered? A Feminist Critical Discourse Analysis." Qualitative Report 21(1): 102-16.

Parsons, T. 1949. The Structure of Social Action. A Study in Social Theory with Special Reference to a Group of Recent European Writers. New York: Free Press.

Puddephatt, A. J. and N. McLaughlin. 2015. "Critical Nexus or Pluralist Discipline? Institutional Ambivalence and the Future of Canadian Sociology." Canadian Review of Sociology / Revue Canadienne de sociologie 52(3):310-32.

Rocher, G. 1992. "The Two Solitudes among Canadian Sociologists." Pp. 65-70 in In Fragile Truths: Twenty-Five Years of Sociology and Anthropology in Canada, edited by W.K. Carroll. Ottawa: Carleton University Press.

Sciortino, G. 2001. "The Structure of Social Action's 'Missing' Chapter on Simmel: An Introduction. Appendix: Talcott Parsons, Letter to J.C. Alexander, January 19th, 1979." Pp. 45-67 in Parsons' the Structure of Social Action and Contemporary Debates, edited by G. Pollini and G. Sciortino. Milan: FrancoAngeli.

Shilling, C. 2002. "The Two Traditions in the Sociology of Emotions." The Sociological Review 50(2): 10-32.

Stanbridge, K. 2014. "How to Commit Canadian Sociology, or 'What Would Innis Do?"' Canadian Review of Sociology I Revue Canadienne de sociologie 51(4):389-94.

Stokes, A. and J. McLevey. 2016. "From Porter to Bourdieu: The Evolving Specialty Structure of English Canadian Sociology, 1966 to 2014." Canadian Review of Sociology /Revue Canadienne de sociologie 53(2): 176-202.

Thomas, J.E. and A. Kukulan. 2004. "Why Don't I Know About These Women? The Intergration of Early Women Sociologists in Classical Theory Courses." Teaching Sociology 32(2): 252-63.

Tindall, D.B. 2014. "A Distinctive Canadian Sociology?" Canadian Review of Sociology /Revue Canadienne de sociologie 51(4):395-401.

Wang, D. 2013. "Is There a Canon in Economic Sociology?" Journal of Economic Sociology 14(2): 111-22.

Warren, J.-P. 2014. "The End of National Sociological Traditions? The Fates of Sociology in English Canada and French Quebec in a Globalized Field of Science." International Journal of Canadian Studies 50:87-108.

Wilkinson, L., J. Bramadat, R. Dolynchuk and Z.T. St. Aubin. 2013. "Are Canadian-Trained PhDs Disadvantaged in the Academic Labor Market?" Canadian Review of Sociology 50(3):357-70.

(1) But see Goyder (2009), Stokes and McLevey (2016), Wilkinson et al. (2013), and Warren (2014); while these articles empirically investigate sociological research and Ph.D. training, we feature theory specifically.

(2) See Albers (2003), Grauerholz and Gibson (2006), Parson (2016), Thomas and Kukulan (2004), and Wang (2013) for studies that use syllabi data.

(3.) Theory is of course not the only such practice; method courses are also almost universally required. This combination--theory and methods--exemplifies Lamont's (2009) findings, whereby sociologists' standards of evaluation value the interplay of theory and methods, whereas for example humanists value interestingness and physicists value truth.

(4.) Textbooks clearly have some influence on author selection, but it is far from absolute. Less than half (45%) of the courses in our database use a textbook (primarily undergraduate departments use readers more often). Readers by Ritzer (17), Calhoun (12), Appelrouth and Edles (11), and Mann (9) are most common and thus most strongly set the agenda for theory courses; but there are about 70 other textbooks or readers in use. Moreover, instructors often select which specific readings they assign. Overall, professors exercise a considerable amount of individual judgment in forming their syllabi-whether by not using a textbook or reader at all; choosing which one to use; or selecting individual readings.

(5.) We have for instance examined course descriptions and objectives to uncover implicit narratives and visions of the sociological tradition. Levine (1995) suggests a typology of such narratives: positivist, pluralist, synthetic, humanist, contextual, and dialogical. For example, positivist narratives envision sociology as a progressive movement toward more secure and reliable objective knowledge about society; contextualist narratives envision sociological ideas as responses to major historical trends such as wars or economic restructuring; humanist narratives trace a narrative around sensitive thinkers and fecund texts in need of perpetual reinterpretation; pluralist narratives envision sociology as a multiplicity of valuable perspectives, with ever more flowers blooming. Our preliminary research indicates that contextualist and pluralist narratives are the most common way that the classical material is shaped into a meaningful story.

(6.) The network also includes a small isolated subgraph comprising the Big 3, not shown here.

Caption: Figure 2 Where Theory Professors Inside and Outside of Quebec Earned Their Ph.D.s

Caption: Figure 4 Nationalities of Authors Assigned in Canadian Sociological Theory Courses

Caption: Figure 5 Different Types of Theory Courses Offered in Quebec and the ROC

Caption: Figure 6 Research Interests of Sociology Professors and Sociological Theory Professors

Caption: Figure 7 Research Interests of QC Theory Professors and Non-QC Theory Professors, and of Female Theory Professors and Male Theory Professors

Caption: Figure 8 Author Coassignment Network for all Canadian Sociological Theory Courses

Caption: Figure 9 Authority Centrality for all Canadian Sociological Theory Syllabi

Caption: Figure 10 Authority Centrality for Sociological Theory Syllabi in the ROC and Quebec

Caption: Figure 11 Classical Author Network

Caption: Figure 12 Contemporary Author Network

Caption: Figure 13 Countries with Highest Percentage of Authors Listed in Canadian Sociological Theory Syllabi

Caption: Figure 14 Correspondence Analysis of Author Nationality in Canadian Sociological Theory Syllabi Table 1 Sociology Departments across Canada Department information Quebec Rest of Total Canada Highest degree offering B.A. 3 28 31 M.A. 1 6 7 Ph.D. 4 22 26 Total 8 56 64 Maclean's ranking Primarily undergraduate 1 16 17 Comprehensive 2 13 15 Medical doctoral 3 14 17 Unranked 2 13 15 Total 8 56 64 Notes: Table 1 describes the sociology departments in Canada. Only departments that offer a minimum of a four-year B.A. are included in the analysis. Table 2 Number of Authors per Syllabus Minimum 2 Average 9.5 Max 30 Table 3 Theory Requirements for Undergraduate Degrees Classical Contemporary theory Percentage theory Not required 18 28 21 Required 36 56 33 Required elective 9 14 9 Missing 1 2 1 Total 64 64 General Percentage theory Percentage Not required 33 43 67 Required 52 14 22 Required elective 14 6 9 Missing 2 1 2 Total 64 Specialty Percentage Not required 44 69 Required 9 14 Required elective 10 16 Missing 1 2 Total 64 Figure 1 Characteristics of Canadian Sociology Professors Male Female Sociological Theory 44 98 Sociology 480 515 Sociology Sociological Theory Cdn PhD 489 134 US PhD 183 27 UK PhD 46 19 FR PhD 25 12 DEU PhD 244 45 Other PhD Unknown Note: Table made from bar graph. Figure 3 Most Frequently Listed Authors in Canadian Sociological Theory Syllabi Marx 46% Weber 45% Durkheim 44% Foucault 30% Bourdieu 24% Simmel 22% Goffman 20% Mead 19% D. Smith 14% Giddens 13% Notes: Figure 3 shows the most frequently listed authors in Canadian sociological theory syllabi. The y-axis shows the total number of syllabi in which a given author is listed; points are labeled with the percentage of syllabi in which an author appears. Note: Table made from line graph.