Capturing the Canon: A Spatial Exploration of Anthropology’s Most Taught Authors
Sara M Conrad
A. Caitlin Lester
Anthony M Ross
This project was born of the idea that the canon is an always-unfolding process. Marcus hints at this when he acknowledges the role of teacher-student interaction as a site of canonical transmission, arguing that “tradition in anthropology thus depends now more than ever on oral transmissions, in courses, corridor talk, teacher-student relations, seminars …” (1991:396). One way to take a “snapshot” of the canon in space (rather than in time) is through current course reading lists. For our purposes, syllabi for graduate level anthropological theory courses are considered to be a reflection of what professors deem to be important for future anthropologists, and representative of the intellectual history of the discipline and the great debates that have reoriented the direction of that history. These required courses can be seen as existing at the nexus of multiple interpersonal processes – departmental debates about what is essential for new members, the influence of previous instructors and students on what is included on course syllabi, the interpretations which texts may elicit, and so on. Regardless of the value the professor attaches to each author – he/she may or may not make explicit judgments in a theory course – the student leaves with the names of Geertz, Bourdieu, Marx, etc. etched upon his/her memory, and in this way is equipped to reproduce the canon for a next generation of scholars. Our aim was to reveal the process of reproduction by uncovering patterns in the selection of certain authors based on their inclusion in course syllabi.
This is a quantitative rather than a qualitative analysis of syllabi that places an emphasis on frequency. In order to determine who the most influential figures in anthropological literature are we first used an online database to identify the number of universities in the U.S. with doctoral programs in socio-cultural anthropology, which gave us a preliminary sample size of 101 programs. Second, after some online investigation we found that not all of these programs had a graduate program in cultural anthropology. We narrowed this list to eighty-eight schools, and we then divided them into four regions according to demarcations used by the United States Census Bureau: Western, Southern, Northeastern, and Midwestern. After each taking one region, we wrote emails to the professors teaching (or who within the past two years had taught) these programs’ introductory theory courses in order to find out which books were on their reading lists. In the event that the contact people in the various departments failed to respond, we used a combination of departmental websites and course descriptions via the registrar’s offices to gain second-hand access to the information we wanted. In collecting data, we also noted the professors’ genders, the schools where they received their Ph.D.s, and their years of graduation. We utilized these various methods to collect data from forty-three of the eighty-eight schools, and we compiled it on Excel spreadsheets in order to be able to extract certain pieces of data and to make easy comparisons between regions. This also served to reduce the chances for coding and data entry errors, which was of utmost importance to us because of our emphasis on empirical data.
We collected 1510 citations from the forty-three syllabi [Fig. 1], and as we had initially predicted some theoreticians’ names seemed to appear over and over. Our analyses revealed that some points of comparison yield more far significant results than others. In regard to determining which authors were most prevalent on course syllabi we found that there were some names that appeared with greater frequency. The fifteen most cited authors in descending order were Michel Foucault (41 appearances), Clifford Geertz (36), Pierre Bourdieu (35), Karl Marx (32), Sherry Ortner (27), Marshall Sahlins (23), Émile Durkheim (22), Franz Boas (21), Eric Wolf (20), James Clifford (19), Claude Lévi-Strauss (18), Max Weber (18), Bronislaw Malinowski (16), A.R. Radcliffe-Brown (15), and Arjun Appadurai (14).
Fig. 1 Distribution of graduate programs assigning top five authors
After this there were a number of authors who appeared frequently on course syllabi, some who appeared on a handful of syllabi, and a striking number who appeared on only one or two syllabi. Two key points struck us while reviewing this finding. First, there is no very clear delimiting point over which the most frequently included authors appear; in other words the distribution here is gradual and emphasizing the fifteen most cited authors was a boundary externally imposed by our group not one internally justified by the data. This suggests no clear-cut “canon” that seems to be more widely agreed upon in instructors in the discipline. Second, it is striking that only one female author, Sherry Ortner, is included among the most frequently included authors on course syllabi for required introductory theory courses, a finding which does lend support to Lutz’s argument (1995) that “theory has acquired a gender insofar as it is more frequently associated with male writing” (qtd. in Mcelhinny et al. 2003: 318).
One of the more surprising findings was female and male professors assigned Sherry Ortner’s work with relatively equal frequency [Fig. 2]. Since the work of Mcelhinny et al. suggests that women are more likely to cite other women we hypothesized that female instructors might be more likely to also assign women authors more often, which is not supported by our data. However, the marked lack of women among the most included authors limits us to a single point of comparison which, while perhaps rather surprising, is certainly not enough to serve as a sufficient challenge to their argument. The only striking difference between the genders was that female professors were far more likely to include Bourdieu and Foucault on their course syllabi.
Fig. 2 Distribution of five most-cited authors by gender of assigning professor.
Next we looked at how the decade in which a professor received his or her Ph.D. affected choice of one of the top five authors [Fig. 3]. Foucault and Bourdieu were not included on the syllabi of those professors whose doctoral degrees were conferred in the 1970s. A majority of professors who finished their doctoral degrees in the last three years assigned Geertz, Foucault, Bourdieu, and Ortner, while only 33.3 percent included Marx on their syllabi. More than 60 percent of professors graduating in the aughts assigned Foucault, Bourdieu, and Marx, while only 40 percent or less assigned Ortner or Geertz. There do not seem to be any particular patterns that emerge from this initial analysis of our data.
Fig. 3 Distribution of five most-cited authors by decade in which doctoral degree was conferred to assigning professor.
Finally, we analyzed the data in terms of the region in which each professor received his/her academic training [Fig. 4]. The most variance in distribution is between Foucault and Bourdieu. Eighty-two percent of professors trained in the West assigned Foucault, whereas only 40 percent of those trained in the Midwest did so. Less than 20 percent of professors hailing from schools in the Northeast assigned Bourdieu, while over 70 percent of those from Western institutions did so.
Fig.4 Percentage of professors by region of degree granting institution assigning each of the five most-assigned authors.
We received syllabi from thirty-nine of the eighty-eight doctoral programs in sociocultural anthropology we identified, so our sample included only less than half (forty-four percent) of the programs in the United States. While our sample size does exceed the recommended size of thirty for statistical significance, we do think a more representative sample would allow us to make a stronger case. Additionally, we did not receive strictly comparable data across the regions, because far fewer syllabi were received from universities in the Northeast, which required us to rely on the often partial data that could be gathered through accessing material available online. More syllabi were received from professors affiliated with programs in the West and South, however, due to the greater number of total programs in each region, the percentage of universities represented in each region was not as high as that of the Midwest. While we attempted to account for the fact that our data was skewed during our analyses, we do think a more fully representative sample from each region would have created a stronger basis for regional comparison.
Finally, when attempting to access course information through universities’ web portals, we repeatedly confronted serious limitations of access and were compelled to focus our data collection on courses offered between 2011 and 2013, since archived course catalogs were only rarely available. This limitation compelled us to focus only on variations across space, while eliminating the possibility of exploring the duration of certain individuals on course syllabi. We feel this is a significant shortcoming because as products of textual communities, syllabi are also rendered deeply and fundamentally ephemeral. Indeed, many of the professors who sent us syllabi did so with addendums in their accompanying emails informing us of how a colleague’s approach to the course differed from their own or about the changes they already planned to make the next time they taught the classes. This suggests to us that, at best, our findings represent what may constitute a canon at this particular moment in time, but nothing more. If syllabi are snapshots in time, then our findings show which individuals are relatively popular now, but a more complete examination of the canon would also likely reveal how they wax and wane in popularity. This approach might allow us to envision a far more dynamic “canon” intersected by trends in thought and shaped by political context.
We would like to thank the many professors from universities across the United States who not only so generously shared their time and syllabi, but also often offered further reflections on the idea of canonicity, the role of introductory theory courses, and the role of the instructor. Many of their comments became central to our interpretations of the collected data.
Marcus, George E. (1991). A Broad(er)side to the Canon: Being a Partial Account of a Year of Travel among Textual Communities in the Realm of Humanities Centers and Including a Collection of Artificial Curiosities. Cultural Anthropology 6(3): 385-405.
Mcelhinny, Bonnie, Marijke Hols, Jeff Holtzkener, Susanne Unger, and Claire Hicks. (2003). Gender, Publication and Citation in Sociolinguistics and Linguistic Anthropology: The Construction of a Scholarly Canon. Language in Society 32(3): 299-328.