The Diversity that Dare Not Speak Its Name
This is a guest post by Dave Frame. Many thanks to him for contributing!
Over the last few years, researchers have pointed out a dimension along which there is an extraordinary lack of diversity in the academic social sciences and humanities. And the response from social scientists has been striking. Usually, statistics like these trigger strident calls to reflect diversity and address systematic bias; in this case – political bias – everyone just smiles and winks. But on what basis should political diversity not matter, given how highly academics prize diversity in regards to gender, ethnicity, religion dis/ability and so on?
One obvious response is that politics is chosen and ethnicity and gender are not. But ethnic, religion and cultural affiliations are at least partly chosen, and we insist on sensitivity towards these important forms of human diversity. The fact an aspect of identity is chosen is not, in itself, an argument against sensitivity to the resulting diversity, except in the unusual case that a person’s choice is somehow illegitimate.
Another answer draws on the theoretical underpinning of social science disciplines. Since much of sociology (for instance) draws on theory from the political left, it’s unsurprising that conservatives choose not to pursue careers in sociology.
But when feminists have pointed out that the theoretical underpinnings of academic work are gendered, it’s been considered unacceptable to shrug and say “that’s why there aren’t many women here, then.” Just the opposite: it’s seen as a reason for change.
A further response is that the bias may be the result of more general psychological relationships between cognition and politics: conservatives don’t become academics because they’re hostile to new ideas/don’t care as much about other people as leftists.
These are sweeping claims predicated on fairly limited evidence regarding relationships between cognition and politics. Regardless of whether or not there is a nugget of truth in these speculations, it seems relevant that hostility to new ideas doesn’t stop conservatives from becoming academic engineers, lawyers or economists, or government policy analysts or members of think tanks (all of whom traffic in ideas relevant to other people’s welfare). Neither does caring less” preclude the significant levels of private philanthropy we see among conservatives. So why should either stop them becoming sociologists? Furthermore, most on the academic left completely reject similar arguments in the case of, for instance, women occupying senior positions in private sector companies, black people entering academia, or students from disadvantaged backgrounds (including racial minorities) entering tertiary education. Ask Larry Summers.
In any case, when we encounter the levels of under-representation Cardiff and Klein present or which Jonathan Haidt discusses we usually start looking for amplification mechanisms of some sort – social processes that act to exacerbate bias. One possible mechanism could be a sort of Schelling tipping argument which would interpret the observed macro-scale bias as an unintended consequence of fairly weak individual preferences for fellow travellers.
What’s interesting about Schelling’s model is that it explains a socially undesirable outcome (racial segregation of American cities) via individual preferences through private transactions (buying and selling houses), without recourse to allegations of nefarious motives or swivel-eyed paranoia: sometimes bad outcomes at a macro level arise from perfectly defensible private behaviour at the individual level.
But even this sort of fairly benign causal account is not a normative justification. It would only amount to a justification if we buy the libertarian argument that distributional outcomes are uninteresting as long as long as individuals are behaving justly. This would seem a strange move for leftist academics. In the absence of such a move, though, the observed distributional bias in the social sciences and humanities would seem still to require justification.
A final answer, which tends to come out after a pint or two, is occasionally offered: if you educate intelligent people about politics they don’t become conservatives. (Just as you don’t find innumerate people in maths departments or young earth creationists in geology departments, so you don’t find conservatives in the social sciences. QED.)
But this is palpably false (as well as insulting): there are loads of well-educated, intelligent conservatives working as policy experts and analysts in the public sector, in think tanks, in business circles, in the arts, the sciences and other facets of public life. The academic social science sector is the entity with the massive diversity problem. My point is just that this statistical bias lacks any obvious normative justification: why we should treat it differently from all those other diversity issues that universities have grappled with over the last few decades?
I suspect the reason for the smile-and-wink response is that academics feel threatened. It’s hard to be inclusive of diversity when that diversity threatens your dominance. Just as embracing racial diversity was stressful and difficult for many white people brought up in the Jim Crow era American South, I don’t expect modern leftist academics would find it easy to embrace political diversity in their own institutions since it amounts to yielding (some) power to people with whom they feel little in common. It’s easier to smile and wink and explain it away.
But this easy, lazy political bigotry stores up bad institutional karma. People tolerate elites because elites bring something useful to society. Where significant minorities correctly perceive elites/universities to be failing, utterly, to represent their perspectives they develop a legitimate grievance with the elites in question. And in such a case they can and should (and, I hope, will) deploy an approach much, and rightly, loved by leftist academics: ask you to justify your bias or remove it.
Dave Frame is Professor of Climate Change at Victoria University of Wellington and Senior Visiting Fellow at the Smith School of Enterprise and the Environment. He has a PhD in physics, an undergraduate degree in philosophy and has worked as a government policy analyst and academic. He regards himself as a political moderate, and voted for Labour in the last UK election.
 Christopher F. Cardiff and Daniel B. Klein, “Faculty Partisan Affiliations: A Voter Registration Study”, Critical Review 17 (2005), nos. 3–4. ISSN 0891-3811, 237-255. Daniel B. Klein, Charlotta Stern, and Andrew Western. Klein, Daniel B. and Stern, Charlotta, “How Politically Diverse Are the Social Sciences and Humanities? Survey Evidence from Six Fields”, Academic Questions Winter 2004-05, Vol. 18, no. 1, pp. 40-52. See also John Tierney’s article on Jonathan Haidt’s work, “Social Scientist Sees Bias Within”, New York Times, February 7, 2011, available online at http://www.nytimes.com/2011/02/08/science/08tier.html?_r=3&nl=todaysheadlines&emc=tha210.
 These responses, and my responses to them, are culled from various conversations over the last few years. Thanks especially to Matt Sharp for some fruitful sparring here on practicalethics which has helped clarify my thoughts, and to Nadira Faulmueller for suggesting this post. Readers should note that I’m compressing arguments for brevity, and that this is lossy compression. Apologies for the bits I have thrown away.
 Thomas C. Schelling (1969). “Models of segregation”, American Economic Review, 1969, 59(2), 488-493.
 Though an aggravating factor in the academic situation would be that academics’ behaviour isn’t entirely private, since they choose (i.e. hire and promote (or not)) their “neighbours”.