By: Emma McDonell
Escalating contact between Europe and myriad “Others” produced by colonialism enabled the emergence of the discipline of anthropology in the 19th century. The discipline was founded upon the idea that a rational (Western European) observer could objectively document the cultures of the multitudinous so-called primitive (non-Western) Others, making the enterprise a distinctly Western European project. Anthropology was originally conceived as the science of “less-developed” peoples, as opposed to sociology which laid claim on knowledge production about “advanced” society (Radcliffe-Brown 1952; Ferguson 1997). The spatial patterns of these dynamics are clear in the history of who publishes anthropology, from where, and about whom. The vast majority of the anthropology published in the US or Europe has been about other peoples and places.
Anthropologists problematized the drastically uneven subject/object power relations back in the 1970s and many called for a “de-colonization” of the discipline. While this decolonization would take on multiple meanings and was never conceived as a singular, cohesive movement, a central concern was the colonial legacy of where anthropological knowledge is made, and who are the objects of inquiry (Nader 1972). Yet decolonization cannot be seen as a simple de-linkage from a colonial past, as many have noted that colonialism was replaced by a 20th century project of “development”, something anthropologists both inside and outside of academia are entangled with in complex ways (Ferguson 1997). The discipline’s very infrastructure is constructed around Western academic knowledge institutions, so “decolonization” in a Fanonian sense would necessitate extricating the anthropological enterprise from the architecture of academia.
This project examines the relative changes in the “topography” of anthropological publication between 1975 and 2013, and in doing so aims to produce a basic measure of the success of “decolonization” efforts on publishing. I am interested in two questions: Are more anthropology publications published outside of the US and Europe today (relatively) than in 1970s? If so, which regions have seen the highest growth rates in anthropological publishing?
I used the Web of Knowledge search tools to select anthropology publications in categories: social science, anthropology, English language. I then sorted this data by year (every 5 years from 1975 – 2010 and then 2013) and aggregated countries into region groups based off the United Nations geographic regions. Then I used the UN geographical region groupings to categorize countries into regional blocs. While there is clearly much variation between nations within each block, as I am interested in relative shifts in the poles of anthropology, revealing major changes in publishing locality.
The Web of Knowledge is an interface that allows searching 23,000 academic articles from Web of Science (1955-present), BIOSIS Previews (1969-present), Medline (1950-present), Zoological Record (1864-present), and Journal Citation Reports (1998-2007). For this project, most of the entire came from the Web of Science. Web of Knowledge selects works through an evaluation process based on “impact, influence, timeliness, peer-review, and geographic representation” (Thomson Reuters 2014). Since Web of Knowledge is one of the premier sources for high-impact research, this database serves as an index for high-impact anthropology publishing in general (Smith, personal communication, February 5, 2014). However, the Web of Science does not include local journals which may be popular publication sources for academics outside the US and Europe.
The Web of Science lists the country where the primary researcher’s institutional affiliation is located. While the researcher may be from a different country and is only temporarily located in this country, and also the co-authors from other universities were not included, these issues could not be overcome with the available dataset. I compiled this data in an Excel spreadsheet, analyzing the changes in ratios of published anthropology by region.
As de-colonization is not simply the spatial spreading of publishing, this study uses only one of many possible measures of de-colonization. Also, the “location” category blurs much of the complexity in relationships between researchers and places.
The numbers reveal interesting shifts in the proportions of academic anthropology published in different world regions and in the ascent of English as the dominant academic language. The relative proportion of publications from the US/Canada declined between 1970 and 2013 from 63.6% of total English language anthropology publications to 42.8%, respectively [see Figure 1]. This downward trend is steady since 2000, when the slope of the line becomes straight..
The trajectory of anthropology publishing in Western Europe since 1975 has been more irregular, perhaps due to the many languages spoken in Europe (and thus the growth of academic publications in other languages). While Western Europe contributed only 11.2% of total English publications in 1970, this proportion surged in the 1970s, and by 1975, Western Europe published 42% of total English language anthropology. Since this peak in 1975, the relative proportion published in Western Europe has decreased, with 1985 being the lowest (20.8%). It’s not possible to discern why these changes occurred from the raw numeric data, though it seems likely that language politics played a role.
Publishing from Africa saw dramatic growth. In 1970, 0.6% of English language anthropology publications came from Africa, but this proportion rapidly increased in the 1980s and 1990s. By 2013, 17% of publications came from the continent of Africa. It’s important to note that these publications have not come from all African nations and 80% of publications from Africa in 2013 came from South Africa. Most of the other Africa countries represented contributed 1-2 articles.
I compared the English/Non-English publication ratio for the same time period and found that non-English publishing has declined, from 21% in 1970 to 5% in 2013 [see Figure 3]. In 1975 and 1980, anthropology articles were published in 15 and 14 (respectively) non-English languages, in including Afrikaans and Serbian. In 2013, just 6 non-English languages were represented.
The aggregation into regional blocs proved problematic. I was interested in tracking changes due to the relationship between colonizers and the colonized, and regional groupings glossed over complex relationships. For instance, Australia was in the Oceania category, though Australia’s inclusion skewed the results to make Oceania into a major publishing hub. I removed Australia and created a separate category for Australia and New Zealand, and yet the issue of particular nation-level historical relationships aggregated into single categories operates at varying degrees through the data. In a more extensive study, it might be useful to track individual countries, and aggregate them into groups using different criteria in order to examine different relationships. For instance, it would be interesting to track whether French anthropologists still tend to write about French colonies, due to colonial relationships.
This data could be the foundation for a long-term project that assesses changes by individual country and attaches the spreadsheets to GIS layers to compile a global analysis of anthropology publishing.
Ferguson, James. 1997. Anthropology and Its Evil Twin. In Frederick Cooper and Randall Packard (Eds.) International Development and the Social Sciences Berkeley, UC Press: 150-175.
Nader, Laura. 1972. Up the Anthropologist: Perspectives Gained from Studying Up. In Dell H. Hymes (Ed.) Reinventing Anthropology. New York, Pantheon Books: p.284-311.
Radcliffe-Brown, A.R. 1952. Structure and Function in Primitive Society. New York, The Free Press.
Smith, Moira. 5 February 2014. Personal Interview. Wells Library. Bloomington, IN
Thomsom Reuters. 2014. Web of Knowledge Overview. Accessed at: http://thomsonreuters.com/business-unit/science/pdf/Web_of_Knowledge_factsheet.pdf