How do you fix academia?

Part I: Why Viewpoint Diversity Isn't Going to Save Us

There are many challenges facing academia currently, but there is one I’m particularly interested in: the internal politicization of academic research. We could also talk about the role of external politicization (and its long history) or the uni’s deteriorating political economy/bureaucratization/adjunctification issues, for example. I care about this one because it’s coming from inside the house, it affects me directly more than the other issues, and it goes to the very foundation of what we do. Further, it exists across presidential administrations; this problem probably won’t be solved, or put on a path for being solved, if a pro-education candidate wins in 2028.

I’ve previously written about solutions to this problem of internal politicization, like how to fix a field, with some additional field-level ideas at the end of a recent Theory and Society article. But in this series, I want to write about how to fix academia on this dimension.

For starters, I’m going to make a bold claim that might upset some friends and folks I really respect: I don’t think we should hitch our star to viewpoint diversity. In this essay, I use organizational sociology and sociolegal studies to explain my skepticism.

The GPT summary of what I say in this post: “Viewpoint diversity is a good academic norm but a poor institutional reform strategy because organizationally it is too ambiguous, too process-oriented, too politically coded, and too burdened with multiple goals to survive implementation.” Or: Viewpoint diversity is not wrong; it’s just the wrong vehicle for reform.

No, I’m not Anti-Viewpoint Diversity

I’m picking on viewpoint diversity because it’s getting a lot of attention—and funding. And I’m sometimes asked about what should we do to improve viewpoint diversity. It’s very tempting to dive right in to answer that question because viewpoint diversity is so important.

I mean, really, who would make the argument that viewpoint diversity is a bad thing? Lots of academics, actually. Usually, this takes the skeptical form of a question like why we should admit flat-earth theory and creationism to our pantheon, as though that’s what viewpoint diversity proponents are demanding. Or that viewpoint diversity is a smokescreen for conservative ideology. Rarely does someone who actually understands what viewpoint diversity means oppose it on its own grounds.

I’m also a member of an organization (the Heterodox Academy) devoted to promoting viewpoint diversity on college campuses. So just to be clear, I’m not actually anti-viewpoint diversity and I do support it, strongly. I’ve even taken the HxA oath!

So let me be really clear in what I’m arguing here by repeating my above thesis statement in different words: viewpoint diversity isn’t what saves us from internal politicization. Obviously, this is speculation and only time will tell. I do think it’s entirely possible that I’m wrong and I do think there’s really good evidence that suggests viewpoint diversity, even the narrow version of political rebalancing, could fix a lot of our problems. Even so, this isn’t where I’d place my money if I was a betting man. You can evaluate the logic of my argument and (eventually) my alternative suggestions and then tell me why I’m wrong if you disagree.

Why Viewpoint Diversity Won’t Save Us

Viewpoint Diversity Means Too Many Things to Different People

For a variety of complicated and sometimes silly reasons, viewpoint diversity sits at a disadvantage precisely because it’s such a compelling and important idea. Viewpoint diversity is catchy and has become faddish, so now it means far too many things, everything from DEI for conservatives (i.e., hiring initiatives) (apologies, Jesse—please read his excellent article on why I’m wrong even though this is how other people see read these initiatives) to creating structures to nurture and protect open inquiry to efforts to encourage high-quality heterodoxy. While organizations like Heterodox Academy very carefully define their term, others like the AAUP only look to politicized efforts to encourage viewpoint diversity and read them in a negative light and then oppose the entire concept of viewpoint diversity.

One consequence of this proliferation of viewpoint diversity’s meanings is, at this point, viewpoint diversity is super right-coded, which instantly brings more resistance than necessary. Note: I’m not saying viewpoint diversity is a conservative concept—I’m not a political scientist and current politics are so weird I have no idea where anything sits on the political diamond anymore. I’m saying enough academics and academic leaders think viewpoint diversity is a conservative idea, is only supported by conservatives, or—my favorite—is a conservative plot to hijack higher ed. The really good and necessary idea is now tainted.

Viewpoint Diversity: Process or Outcome?

Another problem is I’d say—and I think many viewpoint diversity supporters would say—viewpoint diversity is a means to an end, not the ultimate goal. I love this quote from HxA: “The most vibrant universities are not echo chambers dominated by groupthink, but arenas where ideas are tested, refined, and replaced.” Yes! Viewpoint diversity is a necessary (but not sufficient) condition for getting there—“Viewpoint diversity keeps the frontier of scholarly inquiry open.” Also, yes! This is super important. But notice: the goal is vibrant universities with stimulating teaching, solid research, and the expansion of knowledge to new frontiers. The goal is not that we have intellectual diversity on campus; that’s a means to an end. (Some folks argue intellectual diversity on campus is the goal and is a good thing. I agree it’s a good thing, but for me, it’s a means to an end.) Viewpoint diversity efforts, though, are often about increasing viewpoint diversity, protecting viewpoint diversity, or setting up structures that allow for (encourage and protect) viewpoint diversity. It’s a procedural focus—if we adopt these procedures, we’ll get to a thriving university.

In all of these conversations, we’re talking about fixing a problem—we’re looking for reforms. The problem with reforms that emphasize procedure and not outcome is: it’s really easy to get hijacked by symbolic compliance.

Symbolic compliance is a concept used in organizational theory and law and society (and pioneered by the late Laurie Edelman) to understand why legal reforms fail to achieve their stated objectives. The concept describes situations where organizations make a big deal of adopting formal structures (official language and policies, build new offices or hire new officers, etc.) to show how they are complying with a legal mandate. Symbolic compliance is paired with a concept called loose coupling: those formal structures are loosely coupled with actual practice, or said in regular language—the new policies and practices don’t change how the organization does things. The organization is safe on compliance because look at those policy statements, look at that new office devoted to this issue, look at all the things they are doing to show how much they are complying with the new law. In practice, nothing meaningful changes. The generalizable lesson: You can require that procedures are fair, but if you don’t require the preferred outcome, you’re not going to achieve your goal. This goes double for ambiguous phrases like “affirmative action” or “viewpoint diversity” (in fact, we first learned these lessons about symbolic compliance from corporate compliance with the Civil Rights Act). So process-oriented + ambiguity = recipe for symbolic compliance.

So you take a concept that already has multiple meanings (see above) and is kind of fuzzy (how many viewpoints do we need? I hear critics asking) and you focus on procedures rather than outcome (viewpoint diversity rather than, say, solid research), and you’re going to get symbolic compliance. And universities, like other large organizations, are great at symbolic compliance: open up a new office of viewpoint diversity, have some seminars, issue some formal statements, and go back to business as usual.

A third factor for predicting symbolic compliance rather than substantive compliance is laws with no enforcement mechanisms. In some states, we are seeing enforcement mechanisms, but the ambiguity is causing problems there; moreover, I’m really thinking about universities in states where they don’t have these laws and it’s more a response to either a zeitgeist or funding pressure. In those circumstances, I don’t expect enforcement mechanisms because we don’t really do enforcement mechanisms, except for things that actually keep the lights on or where we actually might get sued. And what does enforcement look like: folks will go for the easiest metrics (like how many liberals do you have and how many conservatives do you have) because proper evaluation of research is really hard, time consuming, and insanely controversial.

Too Many Goals, One Tool

But there’s another problem with viewpoint diversity: among those who support it, the reasons for their support differ because it’s a powerful tool that is being asked to solve many problems at once. Two of the most central problems it solves: 1) asking for viewpoint diversity is basically a baby step toward ameliorating discrimination against and hostile work environments for “conservatives” (that is, conservatives, moderates, libertarians, and those faculty or students simply believe are conservative) 2) viewpoint diversity is thought to pave the way for good basic research and for stronger educational environments for students (especially by filling in blindspots, biases, and taboos and doing the bare minimum of our job to present and evaluate all sides of an issue, not just offer our preferred side in the best possible light). There are others, like civil revitalization, or generally helping students (citizens and next-gen leaders) to learn to get along in a diverse society and respect people’s rights to have different beliefs instead of refusing to talk with one another because that person’s belief makes them sub-human and disqualifies them from basic respect and decency.

Of course, one of the most powerful arguments in favor of viewpoint diversity understood as a political rebalancing of universities is that even well-meaning, pro-viewpoint diversity liberal faculty won’t do a good job emulating conservatives or otherwise presenting conservative viewpoints, and certainly not in an energetic/enthusiastic way. This is huge for teaching. The research version is we’re not very good at vetting different ideas equally—we have blindspots, so even if we try to be attuned to multiple perspectives, we will give the edge to those ideas that are more similar to our own beliefs. Instead of organized skepticism, we’ll engage in systematic discrimination of those ideas that we find foreign or distasteful. And you can see this in the echo chamber of our syllabi, the topics of our leading journals, and at conference presentations.

Honestly, checkmate. That’s the hardest argument I have to overcome on the political rebalancing version of viewpoint diversity (at least one version of which I generally oppose—any preferential hiring based on one’s political affiliation/beliefs or that commits a person to a particular viewpoint in their research, as opposed to, say, their willingness to evaluate particular hypotheses, certain variables, or certain theories). I simply can’t refute it—I’m mostly convinced. I don’t want to be (because I’m worried that the only way to solve it really will be to hire for political diversity, which I oppose because I think all extracurricular characteristics should be off limits for hiring and promotion), but I am. Mostly.*

*Why am I only mostly convinced, despite the really strong evidence? 1) I think you can have different politics from other folks and still be so immersed in an echo chamber that you will be just as flawed at evaluating things as someone with different politics. So politics here wouldn’t be sufficient to fix the problem. I also think a lot of really cogent critiques of academia are coming from liberals, even hard-core liberals. So politics here wouldn’t be a necessary issue. 2) I think the dominant issue isn’t actually your core politics but the normalization of not doing basic steps—and not training folks to do them, much of which is correlated with the further leftward tilt. 3) I suspect that if you selected academic jobs based on a love for research and not for social change, you’d course correct in a hurry. 4) I’d say the argument for political diversity also misses the fact that it’s not even just liberals v. conservatives but a very narrow range of far-left views that have come to dominate the overall culture of academia, and much of that is coming from folks who believe they have a political mission. 5) I think a lot of this happened because people were given license to do so when we broke our own rules—at some point, it became okay to say some arguments are not acceptable or desirable, so we can’t even have that conversation. And around the same time and for similar reasons we said our one and only goal isn’t about knowledge but social justice, narrowly defined. That’s not a viewpoint diversity issue, that’s a violation of the basic conditions for knowledge exchange. In Tim Urban’s usage, we went from idea lab to echo chamber. And then it snowballed. And it spread beyond academia to the elite sectors of society. I don’t think adding in conservatives and stirring will fix things if we don’t fix that first or also—it will just be window dressing while the train keeps rolling. Instead, we need to prioritize hiring people who are willing to ask questions and provide answers regardless of their politics. I suspect that will actually fix a lot of the habits we’ve normalized. I’m going to add 6) even though it’s less about the empirical research of political bias in research evaluation I’m talking about that gives me the most pause and gets more to one of the arguments in favor of political rebalancing (it will help us get to under-explored topics). I don’t think we actually need conservative hiring initiatives for that, we just need to incentivize people to do that work. My decadal survey suggestion or a more recent suggestion to incorporate the American people and ask them to weigh in on what we’re missing (as well as policymakers, business people, NGO staffers, practitioners, and scientists in for-profit labs) would also help us figure out those blindspots.

But this bit of doubt doesn’t change my argument about why viewpoint diversity won’t save us: it is still trying to solve too many problems at the same time. (You could entirely pivot and say, look, we need political rebalancing for this one reason, and we’re going to adopt it in order to solve this one problem, but stop there. Don’t add other things that it’s trying to fix—that way ruin lies.)

To be fair, this is a common problem in coalition building: you get people who sign on to a project for multiple reasons. This is how legislation gets watered down so that you can bring more people on board. I sign on for this reason, you sign on for that reason, and now the thing we signed on to has a lot to accomplish. The more diverse the coalition, the more things in needs to do in either different directions or in more watered down versions.

We see this with punishment all the time (a major theme of my entirely written but on-hold book on the history of American prisons). Prisons are frequently supposed to solve several crime-reduction-related goals (deterrence, incapacitation, and rehabilitation), some justice-related or moral goals (retribution/“punishment”), and it’s also supposed to be cost-effective (or sometimes even generate revenue, historically at least) while being humane and not killing or torturing its prisoners and also making sure everyone (prisoners, staff, and visitors) are safe and secure. But it turns out humane treatment is expensive, as is a truly rehabilitative prison, and the most secure prison is not necessarily going to be the most humane or cheap; meanwhile, a truly rehabilitative prison might not be the best deterrent or retributive prison, while retribution, incapacitation, and deterrence can also sometimes be at odds as they are with other justice related issues of equal treatment based on one’s commission crime. It’s very messy when you actually try to balance these competing goals. More importantly, when you get these tensions across goals, there’s always a hierarchy: some goal is going to take precedence over the others. Also very important: different people subscribe to different goals or they have different hierarchies—so someone is going to be disappointed when invariably one goal is prioritized over the others. Under these conditions, it’s very easy for prisons to “fail” according to their stated goals—we simply created an impossible set of conditions for them. (This isn’t a new development, but something we’ve faced since the very beginning of prisons. And we don’t just do it for prisons, we do it for courts, too, as Malcolm Feeley argued in the 1970s and 1980s.)

Now, prisons are an extreme case, but this challenge arises in all sorts of organizations from universities to hospitals. Just think of all the competing considerations universities face of fundraising, academic excellence, research excellence, student life, avoiding litigation, etc. Hospitals increasingly face a balancing act with legal demands, medical considerations, and the wants and needs of families and patients. You have a lot of different goals or criteria of evaluation; they are in tension with one another; something gets prioritized; and someone is disappointed.

Now come back to viewpoint diversity: if some people think viewpoint diversity is about undoing anti-conservative discrimination, and some people think it’s about ensuring norms of open inquiry, and some people think it’s about restoring under-explored theories, and some people think it’s about restoring the university to its basic mission, someone is going to be disappointed when one of those goal is prioritized over the others. Or each one is going to get watered down until it doesn’t actually achieve any of those goals and we go right back to symbolic compliance because we couldn’t agree on what the actual goal is, or how to get there, so we asked for all of the above.

Pick Something With More Staying Power

Since viewpoint diversity is basically a fad right now, and because of the particular political moment, it has some power to get passed into law, adopted for legitimacy concerns, or advocated for and incentivized by funders. But I suspect that as soon as Democrats are back in power at the national level, viewpoint diversity will be asked to move off stage. This goes double because the term is right-coded and already getting resistance from the very groups we need on board. So even if we get symbolic compliance in the short-term, I expect any initiatives that get passed will go unfunded and disappear in another political moment.

Instead, choose something with more staying power because it’s even more fundamental, based on a very clear goal, and is less controversial—something like good teaching and good research. Yes, I said that’s hard to evaluate, but I think there will be more enthusiasm and creativity invested in that than in figuring out something as controversial and nebulous as viewpoint diversity.

Closing

For these reasons, I don’t think viewpoint diversity is going to be the thing that saves us from internal politicization. We should definitely think about viewpoint diversity as a necessary academic norm, but not necessarily a viable strategy for reforming academia.

If not viewpoint diversity, what will work? I’m going to be a broken record on this front: We should organize around excellence in knowledge production/dissemination rather than around political or discursive balance. The bonus is I do think the former will achieve the latter, if done carefully. For example, one of the things one will look for is paying attention to counterarguments and alternative perspectives, which is arguably viewpoint diversity in teaching and research (but of course it’s not the only thing that makes for good teaching or research). But we can’t set out to achieve that goal, or else we risk piling on too many goals.

In another post, I’ll offer some concrete and fuzzy suggestions along those lines since I’ve been thinking about this a lot.