Race and white dominance in American anthropology

anthro phd production by race

In demographic terms, anthropology in the United States continues to be dominated by white Americans. Consider this graph of the racial distribution of anthropology doctorates over the last twelve years (incidentally, the NSF had no data for 1999, so there should really be a gap year inserted here, but I trust you can all manage without one). The enormous top segment of this graph shows the very large fraction of new U.S. anthropology doctorates that go to white Americans. This decade, on average, 65.7% of new anthro phds were white. And yet we also observe that this dominance is falling, slowly, over the years; you can see that here visually. 75% of new anthropology doctorates went to whites in 1995 but only 63.6% in 2007. And other minority groups have grown, slightly, as demonstrated by the widening of those bands that indicate black Americans, Hispanics, and Asians (which includes Pacific Islanders in the NSF-supplied data I use). But racial equity is far from attained.

To really get a sense of these dynamics, it’s helpful to look at specific groups historically and to look at groups overall in proportion to their share of the U.S. population. Consider the following data for 2007:

Race/Ethnicity BA MA PhD US Population
American Indian 1.43% 2.06% 1.15% 0.8%
Asian/Pacific Islander 6.60% 4.20% 4.43% 4.5%
Black 4.43% 3.38% 3.66% 12.6%
Hispanic 7.87% 5.52% 3.47% 14.7%
White, non-Hispanic 69.89% 66.80% 63.58% 66.3%
Temporary Resident 1.36% 8.65% 16.57%
Other/Unknown 8.41% 9.39% 7.13%

What this table shows, interestingly enough, is that the proportion of white Americans declines as you look higher up the degree structure. That goes against a typical demographic principle of social hierarchy, according to which a more culturally dominant group is better represented at higher levels of the social scale. (We saw an excellent example of this in looking at gender in anthropology, earlier this fall — even though men are a minority of anthropologists, they are increasingly well represented at higher educational levels.) However, it seems in the data that the lower representation of whites at higher levels, like the doctoral level, should not necessarily be understood as a promising sign of racial equality. Rather, the dip in relative white dominance seems related to the huge number of foreign students who appear at the MA and above all at the PhD level, where they constitute the largest demographic bloc (16.57%) after American whites. Very few foreign students come here for anthropology BAs (1.36%), but they do seem to come here for, in essence, upper level professional training.

Distressingly, we can also see in this data that blacks and Hispanics (the two largest American minority groups) are radically underrepresented at all degree levels in proportion to their share of the population. And their presence in anthropology is not constant across degree levels: there is a noticeable drop between the fraction of blacks and Hispanics who get BAs and the fraction who get graduate degrees. Worth noting, on the other hand, that Asian-Americans and American Indians are quite well represented in proportion to their fairly small fraction of the American population. I tend to suspect that class is a hidden variable in the relative success of Asian-Americans, since they are (at least ostensibly) better off in the American class system, but the NSF’s national statistics are beautifully and outrageously silent on the question of students’ class origins. So that has to remain pure hypothesis for the time being.

To get a clearer image of the slow decline of general white dominance, you might consider this:

anthro graph of american whites

The overall pattern — where bachelors being the most white, doctorates relatively the least white, and masters are in the middle — appears to have been fairly constant throughout this data, in spite of the masters’ line fluctuating somewhat between the two. The overall spread seems fairly consistent; whatever sociological processes lead certain sets of white people to go into anthropology appears somewhat consistant. If we look at the situation for black Americans, however, we can see a strikingly different sociological picture:

number black americans in anthro

This one I had to format in terms of absolute numbers instead of percentages, since at very low absolute numbers, small changes translate into large proportional changes and the graph becomes unreadable. But actually, I think it’s helpful to have the absolute numbers here. They serve to remind us that in the United States, a country with some 34 million black inhabitants, the discipline of anthropology, which ostensibly prides itself on its progressive understanding of race, is graduating barely two dozen black anthropologists per year. You can see it here with your own eyes: the PhD line in this graph is just barely halfway to the line that would mark 50. And the number of MAs is above that, but pretty similar. To depress you a little and bring the point home, if in 2004 there were 34,772,381 black Americans (per the Census data linked above), and the same year there were 26 new black anthropology PhDs, that works out to one new anthropology PhD for every 1,337,399 black Americans. By contrast there was in 2004 one new (white) anthropologist for every 551,183 white Americans — that’s more than twice as much, though of course still tiny. (Anthropology, I can’t stress enough, is still a tiny field in a large world.)

Now, it seems to me that a very interesting demographic phenomenon here, and a striking departure from the parallel lines on the graph of whites above, is the fact that black Americans are in fact growing quite substantially as a fraction of the anthropology bachelors’ population — the population has almost doubled since 1995 – and yet things seem to be changing very slowly at the doctoral level. Of course, there can be a time lag between these two lines — if more of group X suddenly get more BAs, it would still take that same group of new BAs most of a decade to get PhDs, if any of them want to — and yet I’m still struck by the extremely low numbers at the graduate level. Still talking about black anthropologists, then, one does notice a climb at the PhD level from 5-6 (total for the year) in 1995-6 to a couple of dozen on average this decade, but at the master’s level, things are about constant. As we know, affirmative action is not doing so well these days; and although we can guess that diversity fellowships play a role in the small increases we do see, it’s hard not to think that it’s radically insufficient.

(I guess this is the time to acknowledge that it’s possible that things are not as bad as they seem; the “unknown” category in this data is quite large (7-9% of all anthro degrees), far larger than the black or Hispanic category, and so I suppose it’s at least likely that there are greater numbers than get reported. Unfortunately, I don’t know enough yet to do a proper error analysis of this data.)

In lieu of a conclusion, let me just note that it would be good to look at racial statistics among American anthropologists at large, on the AAA membership, on graduate students (are there major race-related selection effects that function within graduate education? it seems quite possible)… I can’t do that now, but I would love to have thoughts on what else might be worth examining. Other disciplines? Other countries? But for the time being, I confess I find this set of data somewhat disheartening. There’s progress towards racial equity, but it’s slow, and not enough.

8 thoughts on “Race and white dominance in American anthropology

  1. Thanks for sharing this analysis Eli,

    I definitely think looking at other disciplines would help put things in perspective. If one more black anthropologist means one less black sociologist, then its unclear which would be better. The strong increase in the number of BAs to blacks is interesting. I’d also like to see the trend in degree productions for whites. If you think the NSFs data are “outrageously silent” on students class background, then I take it you wish universities were required to report this data? 😉

  2. Thanks, Mike. Frankly, I’m still somewhat stunned by these numbers. I knew that phds are fairly unusual in the world, but I wouldn’t have guessed that any major ethnic group would be in the double digits, yearly, across the nation. Yeah, it would be interesting to know whether there’s a zero-sum situation across social science disciplines; my instinct is not necessarily, but I guess you’d have to check. A good follow-up post might examine a particular racial group across different social science disciplines… not that I have time to write that just now.

    But yes, I definitely support increased collection of data about class backgrounds 😉 I don’t know if I am inherently in favor of nation-wide reporting, though; can’t you find this stuff out through good surveys also? What are the costs/benefits of mandatory reporting vs survey research?

    (ps I finally added an option to subscribe to the comments on a post — can you try it and let me know if it works?)

  3. ok, I checked the box to be notified of follow up comments via email. My guess is that if you send voluntary surveys to departments, many departments will neglect or downright refuse to fill them out. They never asked me my class background, so they would be unable to answer that question. If the surveys go to individual students the non-response problem will be even worse.

    I don’t think 100% of an attempt to increase diversity in anthropology comes at the expense of other social sciences, but some portion probably does. Another large chunk comes at the expense of diversity in business, politics, etc.

  4. Well, there does appear to be successful survey research on these topics, based on sampling individuals (not departments). The big survey of social science phds that I know of, by CIRGE, had a 45% response rate (sample size of 3025). They believed that the responding group was skewed towards those who had stayed in academia and not gone into the private sector…


    I don’t know anything about sample research; is a 45% response rate considered decent?

  5. Yes, some survey research is certainly possible using only volunteers. 45% is a fairly high response rate, given the nature of the study, but still low in an absolute sense. I haven’t given much thought to this but we might think about how the writers of the survey are highly constrained in what they might ask, time consuming or controversial questions might kill their response rate.

  6. Hi Eli,

    Thanks for sharing this with us! I agree with Michael, I would be interested to see how your figures compare across disciplines as well in relation to PhDs afforded to different groups in general.

    But actually what I was even more curious about is whether you had an idea on how to bring in “class”? I’d suspect that if you were to ask, most people would respond they were middle class (if you could get them to respond of course), since to my understanding the majority of Americans consider themselves to be middle class. One of my professors once joked about how at an University Chicago alumni event he’d asked the alumni what class they considered themselves to be. Almost all of them self-identified as middle class, even though many of them made well over $100,000/year, and had substantial assets.

    So, how would you unpack the category of the middle class then? Would you ask people about their parents’ income level? Or education level? Or what stores they shop at for clothes, food?

    Here in Brazil with race- and “class”-base affirmative action finally setting in at most universities, the “quotas” are filled by students who have gone through public middle and high schools, and whose family income is below a certain level, and who self-identify as black (unlike the US, this is often a much less self-evident category than class in Brazil).

Comments are closed.