Why I’ve Changed My Mind About The Importance Of Ideological Diversity In Academia

Sometimes conservatives (or liberals) are much better-positioned to take down a bad idea

(Unlocked 6/20/2024.)

I bet if you asked me what I thought about the question of “ideological diversity” in academia five years ago, I would have brushed you off. At the very least, I wouldn’t have had strong feelings on the matter, because it struck me as something of a silly concern. When we think ‘diversity,’ in the traditional sense, we think about groups that have been oppressed over the years, which can at least partially account for why they are under-represented in certain settings. This brings with it certain moral imperatives.

“Ideological diversity” deals with a fairly different set of concerns — concerns not really steeped in historical patterns of power — and I think the term is a bit unfortunate in that it implies some sort of equivalency between two very distinct concepts. People seem to see the two words together and imagine that they are communicating the claim that conservatives are somehow oppressed in the same sense that actually oppressed groups are.

If we were in the odd position of having to choose just one problem to address between ‘traditional’ diversity and ‘ideological’ diversity, I’d certainly choose the former. But we aren’t in that position, and I’m increasingly convinced that anyone concerned with how academia functions should take the problem of ideological diversity seriously. 

I was reminded of this when I read a great article on the Heterodox Academy website by Cory Clark, a psychology researcher whose work, coincidentally, I recently wrote up here. The article is headlined “Ideological Discrimination and Self-Censorship Among Philosophers and Beyond,” and it’s pegged to a new paper in Philosophical Psychology by Uwe Peters, Nathan Honeycutt, Andreas De Block, and Lee Jussim.

The authors simply asked a bunch of philosophers whether and under what circumstances they’d be willing to discriminate against ideas and people in their field on the basis of their politics. Here are some of the results:

Summing things, up Clark writes:

Across the political spectrum, participants reported experiencing ideological hostility, occasionally even from those from their own side of the political spectrum (particularly within the left-leaning camp). But more right-leaning philosophers reported experiencing more hostility from their colleagues (for the statistically inclined, r = .47), a pattern confirmed by third-party reports: participants reported seeing more hostility against right-leaning philosophers than toward left-leaning ones. And participants reported that they would be more reluctant to defend their own argument if it led to a right-leaning conclusion than if it led to a left-leaning one.

On average, participants reported greater willingness to discriminate against right-leaning perspectives and individuals than against left-leaning perspectives and individuals. And although both left-leaning and right-leaning individuals reported a willingness to discriminate against their ideological opponents, left-leaning philosophers confessed greater willingness to do so than right-leaning philosophers (in contrast to research suggesting that liberals and conservatives are similarly discriminatory toward ideologically dissimilar others). More specifically, left-leaning philosophers reported greater willingness to discriminate against their right-leaning colleagues in grant applications, symposia invites, paper acceptances, and hiring decisions than right-leaning philosophers reported in regards to their left-leaning colleagues. The figure below depicts the percentages of right-leaning, moderate, and left-leaning philosophers who responded at or above the midpoint in willingness to discriminate against their ideological friends and foes within the four discrimination categories.

Clark goes on to note that “Philosophers are not peculiar here. The present findings are roughly similar to previous studies in which academics reported willingness to discriminate against ideological opponents in social psychology and across disciplines in California State Universities. Given that ingroup favoritism appears to be a fundamental feature of our human psychology, it seems likely that similar findings would emerge in any discipline whose lines of inquiry touch on political issues. This means that left-leaning academics are also not peculiar.” Psychology is the field I’m most familiar with, and it’s very ideologically skewed — the studies Clark linked to found ratios of Democrats to Republicans of 14:1 (among social and political psychologists) and 8:1 (among a big sample of academic psychologists in California), respectively.

I’m working on a book about why half-baked psychological ideas go viral, and that work has really shifted my views on the importance of ideological diversity. The more you learn about how science works, the more you realize just how much distance there is between some people’s idealized notion of the *** The Scientific Method *** and how things play out in real life. Every step of the way, science is a human institution and human errors can cloud the process. Many of those errors have to do with groupthink and faddishness and confirmation bias, and every single one of these tendencies, in turn, is exacerbated when everyone in the room shares the same political values and priors. (This isn’t too far off from one of the pragmatic arguments for ‘traditional’ diversity, which is that having people from a range of backgrounds in an organization makes it more agile and adept at creative problem-solving.)

Now, oftentimes the way this works isn’t as simple as liberals trying to squelch conservative theories or (much less often given their rarity in many corners of academia) conservatives trying to squelch liberal theories. It’s usually more complicated and subtle than that. Let’s take the implicit association test as an example.

Whatever one thinks of my argument that the evidence for this test, which supposedly measures your unconscious bias, was severely overhyped, the fact of the matter is that even the test’s creators acknowledged in 2015 that it is far too noisy to be used to predict individual behavior. And yet the IAT is still widely administered to individuals, and most people don’t understand how little meaning their results hold.

When you look at this test’s evolution, it’s clear that it was carried along a fair distance by political currents. If you are the sort of person who believes diversity and diversity trainings and discrimination are major societal concerns, this test spoke to those concerns. It offered a solution (revealing and potentially ‘treating’ implicit bias) to a problem (discrimination in the workplace and elsewhere) that most liberals believe to be pressing. It’s basically impossible to separate, in the human mind, the sentiment This idea is true from the sentiment This idea stokes my own values. (If you disagree with that, look into Dan Kahan’s research.) “The problem is that implicit measures, and the IAT in particular, became a critical part of a political narrative about why disparities between social groups exist in the United States,” Patrick Forscher, a psychologist at the University of Arkansas who studies intergroup disparities and who has conducted research on the IAT, told me when I was first reporting on the test. “Thus, claims about implicit measures became, to a certain extent, political claims, not just scientific claims.”

Of course, not everyone believes diversity is a major issue, or that diversity trainings are important, or that it’s likely the case that most white people, as the IAT’s proponents long claimed, are unconsciously biased against nonwhite people in a manner that is relevant to the real world. The people who are skeptical of these claims tend to be called ‘conservatives.’ And to many liberals, their stances are obviously, if not offensively, wrong. But you know what? From a scientific point of view, where they’re coming from, and whether they have the ‘right’ values, doesn’t matter. In fact, it can’t matter, because either their critiques of the IAT are correct, or they aren’t. And based on what we know about social and political psychology, if a liberally oriented theory has some holes in it, a conservative is going to be much more likely to find those holes and publicly point them out, because they can do so less hindered by various cognitive and sociological roadblocks.

Just off the top of my head:

1) In the case of a ‘liberal’ theory, a conservative will be less likely to be forced to grapple with the uncomfortable cognitive dissonance of finding out a theory that seems to line up with their values is false, rendering them freer to understand its shortcomings.

2) If, as is often the case, the conservative in question has friends who are mostly conservative rather than mostly liberal, the ‘cost’ of publicly speaking out against the popular, ‘liberal’ theory will be lower, shifting their personal incentives in a positive, pro-rigorous-science direction.

3) Conservative values, whatever we liberals might think of them, might in some cases offer useful lenses or angles for exposing the weaknesses of a ‘liberal’ theory — lenses and angles liberals don’t have such easy access to.

Suffice it to say if we weren’t talking about the IAT, but rather some theory which “points in” a conservative direction, we could flip all the above labels and the arguments would still hold. And we should also keep in mind that, in this example, a conservative is more likely to identify problems that don’t actually exist in the IAT (or any other ‘liberal’ idea), because they will be cognitively motivated to disbelieve the idea in much the same way a liberal will be cognitively motivated to believe it. (That said, at the moment psychology has a much bigger problem with false positives than false negatives, so I’m less worried about that.)

This same general argument applies to any claim in any field. Yesterday I recorded a video chat with a philosopher about the wars raging there over the concept of gender identity. Because philosophy is overwhelmingly liberal, and because this is a particularly morally charged subject at the moment, the philosophical discourse on gender identity only reflects a really tiny sliver of (far-left) public opinion, which is rendering philosophy rather irrelevant to helping the average member of the public understand how to think through some novel practical issues pertaining to (among other things) identity, gender, and the question of what we morally owe other people. In the view of some philosophers, certain rather unsupported, overhyped claims about gender identity have gained purchase simply because they tickle progressive sensibilities on this issue, and they would not have survived peer review were the field a bit more ideologically diverse. (To be clear, I don’t think philosophers have a duty to have their work be relevant to the public in every context, but in this context they are writing about a subject of great societal interest that would benefit from their insights.)

Or, substituting in a different hot-button issue but sticking to the same discipline, take this comment from one of the respondents quoted by Peters et al quote in their study of ideological diversity in philosophy: “I suspect that men and women are predisposed to have different interests, and that this accounts for the disparities in gender ratios across disciplines/professions. Yet this view is not one I am able to voice openly […]. I don’t know what reaction people would have if I were to make this view public, but I suspect it [sic] ‘hostile’ would be an understatement.”

In my view, this is an overstatement — I’d be more comfortable with a claim like “partially accounts for” — but that’s beside the point: It strikes me as eminently reasonable to believe that this professor really does feel some fear about expressing this viewpoint in a liberal-dominated discipline. If so, that’s a shame, because in fact we do not know the precise reasons for gender ratios in different professions. People like to pretend we do, and to issue sweeping monocausal statements on the matter — which, to be fair, this professor is a bit guilty of — but we don’t: It’s an almost completely unresolved issue, a giant cliffside researchers have barely begun to scale. And will they be able to get to the top if a sizable percentage of researchers and theorists remain stranded at the bottom, unable to help because of fear they will endure professional or reputational harm?

If I’m right and ideological diversity should be taken seriously as an issue within academia, that brings with it certain questions, like what to do about this problem and what causes these ideological disparities in the first place. I’m going to punt on those questions — they’re complicated. My main goal, for now, is to try to convince skeptics that a lack of ideological diversity within academia constitutes a real problem that shouldn’t be laughed off, because it may have profoundly distorting, unscientific (or unphilosophical) effects on what gets published and what gets debunked.

---

Questions? Comments? Concerns about this newsletter’s lack of ideological diversity? I’m at singalminded@gmail.com or on Twitter at @jessesingal.