«Sociocultural Anthropology in American Anthropologist (1901-2011) Scholars examining the history of sociocultural anthropology in the United States ...»
Sociocultural Anthropology in American Anthropologist (1901-2011)
Scholars examining the history of sociocultural anthropology in the United States (e.g., di
Leonardo 1998; Moore 2009; Murphy 1976; Paterson 2001; Stocking 1976) often make
generalizations about changes in the field in the past century. They note how certain research
areas, topics, methods, and theoretical approaches have become prominent at certain times and
then fallen out of favor. Despite an abundance of carefully-researched books and articles about the history of American anthropology, authors only occasionally (Charnley and Durham 2010;
Chibnik1999; Chibnik and Moberg 1983; Erasmus and Smith 1967: Lewis 2009; Lutz 1990;
Stern and Bohannon 1970) provide quantitative data about changes over time.
The lack of numerical data can contribute to disagreements about anthropological practice in the past. For example, many anthropologists writing in the 1990s (e.g., Keesing 1994; Turner
1993) argued that the field prior to 1960 or so usually ignored history and treated cultures as isolated units. In a response to such generalizations, Herbert Lewis (1998) showed that some American anthropologists in the first part of the twentieth century wrote about “acculturation,” colonialism, and globalization.1 Lewis asserted that many anthropologists urging anthropology to become more historical were ironically ignorant of the history of their own discipline. In the absence of quantitative data about publications, however, it is difficult to determine if the critics of past anthropological practice were for the most part right.
Even when generalizations are widely accepted, there may not be information about the extent to which they hold. The following are a few such summaries that have been made about trends in
(1) In the first part of the twentieth century, anthropologists from the United States mostly wrote about American Indians. Later on, such research became much less common.
(2) Prior to the large-scale entry of women into the field in the latter part of the twentieth century, studies of gender were uncommon.
(3) Quantitative methods, which were rarely used by American sociocultural anthropologists in the first half of the twentieth century, became common between 1950 and 1990, but now are less popular.
While few would dispute such claims, questions about the magnitude of these changes cannot be answered without relevant quantitative data.
In this article, I provide numerical data about the contents of articles in sociocultural anthropology published in American Anthropologist (AA) between 1901 and 2011. I look at changes over time in research areas, topical coverage, and ancillary material (numerical tables, drawings and diagrams, photographs, maps). I also examine the extent to which the journal’s editor and the gender of authors influence the contents of articles.
There are obvious limitations in the extent to which articles in AA provide a window into changes over time in sociocultural anthropology in the United States. AA is only one of many venues for publications available for U.S.- based anthropologists. Moreover, from its inception AA has published articles written by scholars based outside of the United States. Nonetheless, I would argue that the content of articles published in AA provides a good, if imperfect, indicator of trends over time in sociocultural anthropology in the United States. Until 1974, AA was the only major journal published by the American Anthropological Association (AAA). Articles in AA have always had high citation rates; the journal also scores well on newer measures such as downloads and impact factors.2 Furthermore, the great majority of authors of AA research articles (86% of those examined here) were based in the United States at the time of publication.3 I fully recognize that the broad-brush, often crude, analytic methods in this article are no substitute for careful historical studies of changes in American anthropology. The contents of topical categories such as “religion,” “economics,” and “expressive culture” coded as “present” or “absent” from articles have changed greatly over time. An example of such changes can be seen by looking at two articles from different time periods coded as including content “about religion.” A piece written in the first part of the twentieth century (Fewkes 1906) carefully describes ceremonies associated with Hopi shrines. Little attempt is made to interpret the meaning of these ceremonies. The approach taken in Fewkes’s article contrasts greatly with the one taken in a piece appearing a few years ago in AA, in which the author examines “how Pope John Paul II’s canonization of Juan Diego, the Indian to whom the Virgin of Guadalupe first appeared, was variously represented by sections of Mexican society as an acknowledgment of the indigenous element in Mexican Catholicism and thus a restitution of past wrongs; conversely as a final domestication of the Indian; and as an evangelical move against a resurgent Latin American Protestantism.” (Beatty 2006:324) Counting these two articles as both being “about religion” obviously ignores significant differences in their content. Nonetheless, I would argue that such rough categorizations and counts are useful. Of the 78 articles in database used for this article that were written between 1900 and 1919, 34 (44%) were coded as being about religion. Of the 121 articles in the database written between 2000 and 2011, only 26 (22%) were coded as being about religion. These counts, when combined with data from other time periods, clearly show a decline in emphasis in the AA on religion over the years. Without the numerical data, we would know less about the extent of this decline.
METHODSMy analyses are based on an examination of every research article related to sociocultural anthropology published in AA between 1901 and 2011 that was at least six pages long and appeared in years ending in 1,3, 6, or 8. I decided to focus on sociocultural anthropology because the contents of some articles about archaeology, biological anthropology, and linguistic anthropology were sufficiently different from those on sociocultural anthropology that comparisons did not seem useful. However, the great majority of articles I looked at – including many pieces focusing on archaeology, biological anthropology, and linguistic anthropology – were sufficiently related to sociocultural anthropology that I was able to include them in my analyses. The sample was restricted to pieces at least six pages long because shorter pieces often were informative notes about specific topics rather than full-fledged “research articles.” I restricted the sample to four years out of each decade because I thought this would provide sufficient information about trends over time. The sample ended up including 875 articles, more than enough to allow me to make the comparisons I sought to make.
Each article was coded for the following:
(1) Characteristics of the author(s) – number of authors, gender of first author, gender of second author (if applicable), place of residence of the author, institutional type of author (university, museum, government, other) (2) Content of the article – research area, whether or not the articles contained information about 21 different topics (e.g., social change, kinship and social organization, economics, legal systems) (3) Ancillary material (presence or absence of numerical tables, photos, maps, drawings and diagrams) The most difficult coding decisions were about the presence or absence of information about particular topics in an article. The most important problem in this respect was determining how much information was necessary to conclude that an article contained material about a particular topic. I took an inclusive approach, coding “present” for any article including two sentences or more about a topic. Another significant problem was determining whether or not certain discussions in particular articles could be fit into my topical categories. For example, many articles about kinship and social organization that appeared in AA between 1930 and 1960 discussed “economics” only with respect to property rights given to members of particular lineages. Although I decided to code such articles as including material about economics, I could just as easily have chosen otherwise.
The codebook used in the analyses can be found in Appendix 1, along with detailed comments about coding procedures and problems. The codebook includes several categories (e.g. “socieconomic type” – foragers, tribal agriculturalists, etc, “primary data source” – ethnographic, historical) that I eventually chose not to analyze because they proved not to be useful.
I restricted the sample to single-authored articles in my analysis of relationships between the gender of authors and the contents of their articles. Not too much information was lost by doing this. Of the 875 articles in the database, 754 (86%) are single-authored, 91 (10 percent) have two authors, and 30 (4%) have three or more authors.
Most of my analyses of changes over time are based on grouping of twenty year periods (1900-1919, 1920-1939, etc).4 The last period analyzed in these comparisons is shorter (2000These groupings cover fairly discrete periods in the history of American anthropology and allow for easily comprehensible presentations of trends in bar graphs. After constructing the codebook and entering the data, I created a new category in which articles were grouped according to the journal’s editor. My principal motivation for this recoding was to more closely examine the content of articles published in AA in the 1980s and 1990s, where some editors were reputed to be more favorable to quantitative and “positivist” approaches than others.
CHANGES IN RESEARCH AREAS OVER TIMEMost articles in the database about sociocultural anthropology (74%) are based primarily on research carried out in one broadly defined geographical area. The remaining articles are theoretical discussions (17%), cross-cultural analyses (5%) and comparisons of two or more sites in different geographical areas (4%). The analyses that follow are based on the 645 articles in the database that are clearly about one geographical area.
Although it is well-known that sociocultural anthropologists based in the United States conducted most of their research among North American Indians in the first part of the twentieth century, the extent to which this areal focus persisted is not always emphasized in histories of the discipline. Figure 1 shows the percentage of articles in the database focusing on North American Indians in AA during different time periods. While the high figures for the proportion of articles about Native Americans in 1900-1919 (79%) and 1920-1939 (55%) may not be surprising, this figure remained high (38%) even during the period 1940-1959. The percentage of articles about Indians of the United States and Canada dropped significantly in the period 1960-1979 (16%) and has been considerably lower since then (7% in 1980-1999 and 5% in 2000-2011). Despite this decrease, Native Americans remain a group of considerable interest to anthropologists.
The focus in AA until recently on American Indians can be seen clearly via comparisons with the number of articles published about all other groups in the region. Of articles in the database published between 1900 and 1959, 154 (52%) were about American Indians compared with only
9 (3%) about other groups in North America. Even in the period since 1960, 36% of articles about North America in the database are about Native Americans. Articles about non-indigenous groups in North American have appeared only recently in significant numbers. Such articles comprised 19% of those focusing on one geographical area between 1980 and 1999. The rise of this figure to 33 percent during 2000-2011 clearly reflects a significant change in anthropological practice.
Another way of analyzing the figures on research areas is to examine the extent to which AA has focused on the Americas during different time periods. (See Figure 2). In the earliest years covered by the database, 86% of area-specific articles focusing on places in the Western Hemisphere. Starting around 1920, the journal began publishing more articles about other parts of the world. Even so, two-thirds of area-specific articles in the data based published between 1920 and 1939 were based on research sites in the Americans. In the middle of the century the journal became a bit wider in areal scope, with 58 percent of articles published between 1940
and 1959 focusing on the Americas (69% during 1940-1949, 50 percent 1950-1959). This trend became more pronounced during the 1960s and 1970s, with only 41 percent of area-specific articles covering the Americas. In recent years, AA has become more Americas-focused, with 51 percent of area-specific articles published between 1980 and 2011 covering the Americas (49 percent between 1980 and 1999 and 53 percent between 2000 and 2011). This increase can largely be accounted for by the striking increase in articles in recent years in articles about nonnative groups in North America. (See Table 1.)
As can be seen in Table 2, articles in AA about the Eastern Hemisphere have been relatively evenly split among Asia, Africa, Oceania, and the rest of the world (Europe and the Middle East). The attention to these areas, however, is wildly disproportionate to their relative populations. From this perspective, Asia is greatly underrepresented and Oceania greatly overrepresented.
Table 2: Percentage of Articles about Different Parts of Eastern Hemisphere 1900-1919 1920-1939 1940-1959 1960-1979 1980-1999 2000-2011