The other day I got an email that contained a great implicit question. My interlocutor wanted to assign my recently published Presidential Address to Peace Science (ungated proof here), but expressed reluctance to do so given the paucity of work written by women cited within it. The subtext seemed to be: given your activity supporting greater awareness of sexism in the discipline, what’s up with that reference list?
A bit o’ hyperbole: 90% male is close enough.
I had not eyeballed the gender composition of the References in the piece, and thus did so. Eighteen of the twenty publications (90%) were authored by one or more males, one by a female, and one by a mixed pair (see table below the notes). D’oh! That is an arresting percentage. So, indeed, what’s up with that?
Confronting the Implicit Biases Systems Impart
Like any systemic bias, sexism is a complex web and I am not going to try to briefly sketch it: we all have grown up in a sexist society, and it impacts all institutions and human relations, including, of course, political science and its many subfields. That said, I can think of several specific domains in the profession where I have sought to address implicit biases.[1]
- PhD Student & Faculty Recruitment
- Mentoring (students & colleagues)
- Letters of Recommendation
- Syllabus Construction
But it turns out that I have always been ambivalent about vetting References in my writing and I have never pushed myself to explore my ambivalence. The outcome, this recent email forced me to realize, is that I have not made it a habit to scrutinize my References for gender (or other) composition. Which raises the questions: Why not? and Should I change?
The Gender Citation Gap
It would be fabulously weird, given sexism in society, if there was not a gender citation gap (for documentation of said gap, see here, here, here, and here), and it would be equally surprising if that gap did not decline over the decades. The question is: what should we do about it?
I think syllabus construction is an easy case to assess. First, when we create a syllabus we are producing a public good. When we publish we are making a case for our ideas in a competitive marketplace of ideas. When we produce syllabi we are training and socializing. In this latter role we have an obligation to make our students aware of a variety of ideas, and this is especially true when it is the Core course / Proseminar.
As long as one does not reject the exist of sexism, it is not difficult to observe that (1) women were systematically excluded from the profession through the 1960s, and have only gained equal access to PhD programs during the 1990s,[2] and (2) this fact will dramatically impact women’s representation on syllabi. The syllabi we create are strongly influenced by the syllabi of the courses we took in graduate school, and as a consequence, active intervention is needed to prevent change from moving at a snail’s pace. In brief, this is why I vet my syllabi for under-representation[1] and urge others to do so as well.
Let us now turn to the work we engage in our research, and lands in our References. Writing at the Duck, Dan Nexon conducted a self experiment and argues in favor of actively reviewing one’s reference and then making adjustments to include more work authored by women. What is the case for ambivalence toward such arguments?
The truth is, I don’t want to talk about this publicly. Why? Because I harbor something close to contempt for the modal citation practice in political science. I think the community cites poorly. And don’t get me started on the nonsense policies of misguided journal editors to place limits on the number of citations (at the precise historical moment when Adobe’s PDF language had made paper irrelevant–you have to be fucking kidding me!). But I digress.
If you have an opportunity to buttonhole one of my PhD students, ask her/him whether I gave them static about their crappy citation practices during graduate school. But, as you can now see, I have no incentive to share this: randomly ragging on large groups of people tends to be counter productive. Yet, I can see no way out of the quandary, so damn the torpedoes.
Citation, in my book, tells a carefully crafted story and should never be used strategically. When we cite we are acknowledging an intellectual debt and we are producing a narrative that reveals the development of knowledge in a given area of inquiry. That is precisely what makes the fuktup citation practice in an article like Collier & Hoeffler (2004) [5,384 cites on Google Scholar] so mind-bendingly frustrating. The research reported there is interesting and valuable. The fealty to community is bankrupt.[4] And while that paper is an outlier, the modal paper I read falls well short of what I believe we should strive for (and those I review usually fall even farther from the mark).
So, in thinking about the source of my ambivalence over the past few days, I have narrowed it down to a clash of values. Ceteris paribus I like the idea of vetting References for exclusion of under-represented classes of voices, but my concern about a different value—the careful construction of the story of the germination of my ideas—has taken precedence, leading me to not even conduct the review.
Show Us Some Data
OK, so what does the gender citation pattern look like for someone who has paid zero attention to it? I decided to look, and restricted the sample to the six solo-authored articles I have published (1990 Political Communication & Persuasion, 1995 PRQ, 1995 JCR, 1998 AJPS, 2000 JCR and 2010 PS).[3] Four were published in the 90s, one in the 00s, and one in the current half decade.
Note: the red comment boxes indicate 11 self citations.
What do you think? I’m not sure what to make of the specific values, but the pattern seems to be precisely what one would expect, especially for a white male pursuing a rational institutionalist research agenda studying dissent-repression using statistical models to conduct hypothesis testing.
Hello, Can we Consider Marginalized Groups who aren’t Privileged White Women?
As Jeff Colgan observed in a footnote to his recent post on the gender composition of work assigned in IR graduate syllabi:
Diversity in IR, beyond gender diversity, merits attention and study. The data are a lot harder to collect. I know, that’s a crappy excuse.
But I am less concerned about the paucity of data than the paucity of discussion. So since I am navel gazing, let me state that the authors of the 313 works I cited in my . Put another way, the ideas of people of color and people who live and think in the Global South are poorly represented in my research. That fact makes my work quite typical for the literatures to which I contribute–even absent data, I am confident that the data will reveal this to be so–but that doesn’t make it any less of a problem.
Whether that problem is better understood as a problem of access to training and tenure-track positions that are biased against people of color and those from the Global South or the intellectual hegemony of white people and the Global North I leave to the reader to consider.
Outro
So, what will my practice be going forward? My navel gazing leads me to conclude that the my ambivalence was overdrawn and the post draft review proposed by Nexon makes sense. I should have incorporated it earlier. I can’t say how much of an impact it will make on my citations, but a practice that pushes one to confront implicit bias will do not harm. It is not obvious that I cannot “have my cake and it eat it too” (i.e., the trade-off I have implicitly imagined may not exist).
In closing, I want to state that I do not wander this cold orb under an illusion that I know what’s best. The best thing about a scientific community is that none of us can dictate: the community decides. This is, of course, true of all democratic systems. Any given argument I make or position I take may be ill considered, stupid, or otherwise flawed. And I reserve the right to change my mind as I learn from others.
Finally, I am in my early 50s, and folks in their 20s and 30s have been socialized in a more equal, less biased world than I. As such, their age cohort has higher expectations than mine. A non-trivial portion of socio-political change involves younger people pointing at their elders and jeering at their flaws. This is a good thing. So if my position here (and elsewhere) warrants, point and jeer away.
@WilHMoo
[1] Not only with respect to sexism, as I briefly note further down in the post.
[2] This is a blog post, so I am killing nuance: it is part of the genre. Yes, that claim uses a dichotomy that obscures, fails to address faculty, and a variety of other issues. I’m living with that.
[3] I cannot vouch for the extent to which my co-authors did (not) pay attention to the gender composition of References, and thus exclude co-authored articles from the sample.
[4] The specific problems are a garish use of straw-“men” characterizations and absence of interest in an enormous amount of relevant published work from a generation of scholars who published during the 60s-80s, and some in the 90s.
Citations in Moore (2015, CMPS)
Will Opines 