Universities Are Not on the Level

Academics should think more about what their industry has done to lose the trust of Americans

Dear readers,

Over the last few months, there’s been a lot of talk about how conservatives have grown increasingly alienated from institutions of higher education. Confidence in colleges and universities is now sharply polarized by party; between 2015 and 2023, Gallup found that the share of Republicans expressing “a great deal” or “quite a lot” of confidence in higher education fell by 37 percentage points. As conservatives have come to look negatively at these institutions, Republicans have increasingly engaged in political attacks on the sector, most recently, in the fact-finding and pressure campaign that caused Claudine Gay to resign this week as president of Harvard.

Less discussed is the fact that public confidence in colleges has fallen significantly across all ideological groups since 2015. Though Republicans’ confidence cratered the most, Gallup found that it fell by 16 points among independents and 9 points among Democrats.

Often, when an issue becomes polarized, you’ll see thermostatic effects in public opinion, as when Democrats became more liberal on immigration in response to Donald Trump’s histrionic attacks on immigrants. But while liberal figures on campus like to talk about themselves as a vanguard in a fight against conservative know-nothings who would take down knowledge and expertise, there is no pro-college backlash among liberals that is apparent in the polls. So it would behoove the champions of knowledge and research and expertise and truth at our nation’s elite universities to be a little less entitled and whiny, and a little more introspective about why everyone seems to like them less than they used to.

I personally have also developed a more negative view of colleges and universities over the last decade, and my reason is simple: I increasingly find these institutions to be dishonest. A lot of the research coming out of them does not aim at truth, whether because it is politicized or for more venal reasons. The social justice messaging they wrap themselves in is often insincere. Their public accountings of the reasons for their internal actions are often implausible. They lie about the role that race plays in their admissions and hiring practices. And sometimes, especially at the graduate level, they confer degrees whose value they know will not justify the time and money that students invest to get them.

The most recent debacle at Harvard, in which large swathes of academia seem to have conveniently forgotten what the term “plagiarism” means so they don’t have to admit that Claudine Gay engaged in it, is only the latest example of the lying that is endemic on campus.

For me, the problem starts with the replication crisis. I was a psychology major at Harvard, graduating in 2005. Most of my coursework was in social psychology. And something I keep seeing in the news since I graduated is that a decent amount of what I was taught in Harvard’s social psychology courses was just wrong. At the time I was a student, there was a vogue for this whole body of research about how “priming” and little behavioral nudges can materially affect attitudes and behavior, but the studies I learned about in class keep getting debunked, with replications failing and the studies often having been p-hacked or even based on fraudulent data. The widespread research dishonesty in psychology stemmed from poor incentives — interesting findings get published, publications get you funding and job security and notoriety and sometimes even corporate consulting contracts, and it’s easier to produce an interesting finding if you juke your data — and was made possible by lax practices that prevailed until recently, including researchers not having to pre-register their study endpoints and not being expected to share their raw data for review by others.¹

Universities’ level of interest in addressing widespread research dishonesty in behavioral science has been… mixed. I can’t be the only one wondering how much longer Dan Ariely is possibly going to remain a professor in good standing at Duke. A lot of the work in cleaning up the field has been done as a side project by three professors who write Data Colada, a blog about data integrity. Harvard Business School, to its credit, aggressively investigated data-fraud allegations against Prof. Francesca Gino that Data Colada first raised, ultimately suspending her last year and thus enduring a lawsuit and the de rigueur allegation that taking action against her was sexist. But if the Data Colada team hadn’t done this sleuthing in their spare time, neither HBS nor the journals that published Gino’s research would ever have noticed that they’d been helping to perpetuate a fraud. And in a preview of the defenses we’ve seen in the Claudine Gay case, Stanford Law School professor Lawrence Lessig looked past the ample evidence of data fraud to tell the New Yorker he rejected the allegations against Gino because of who she is: “I’m convinced about her because I know her,” he said. “That’s the strongest reason why I can’t believe this has happened.”²

I’m not under the impression that the replication crisis in the social sciences looms as large for others as it does for me. But research dishonesty in universities goes beyond the social sciences. In the humanities, it has taken a different form — postmodern research that aims at “my truth” instead of truth.

Matt Yglesias wrote a few weeks ago about a paper by Jenny Bulstrode, a historian of science at the University of London, who alleges that a moderately-notable metallurgical technique patented in England in the late 1700s was in fact stolen from the black Jamaican metallurgists who really developed it. The problem with Bulstrode’s paper is that it marshals no real evidence for its allegation — not only failing to show that Englishman Henry Cort was aware of a Jamaican metallurgical technique similar to the one he patented, but failing to show even that such a technique was ever used in Jamaica.

The paper, because it fit into the fashionable category of Historian Finds Yet Another Thing That Is Racist, garnered a lot of credulous press coverage. And when people pointed out that the paper didn’t have the goods, the editors of the journal who published it came out with a What Is Truth, Anyway-type word salad in defense of the article, including this:

We by no means hold that ‘fiction’ is a meaningless category – dishonesty and fabrication in academic scholarship are ethically unacceptable. But we do believe that what counts as accountability to our historical subjects, our readers and our own communities is not singular or to be dictated prior to engaging in historical study. If we are to confront the anti-Blackness of EuroAmerican intellectual traditions, as those have been explicated over the last century by DuBois, Fanon, and scholars of the subsequent generations we must grasp that what is experienced by dominant actors in EuroAmerican cultures as ‘empiricism’ is deeply conditioned by the predicating logics of colonialism and racial capitalism. To do otherwise is to reinstate older forms of profoundly selective historicism that support white domination.

These ideology-first, activism-oriented, the-truth-depends-on-who’s-looking approaches also extend into the soft social sciences — see, for example, the theme of the 2024 American Anthropological Association annual meeting, which declares an intention to “reimagine” anthropology in a way that breaks down the barrier between theory and practice to make more room for more social activism, so that anthropology better serves as a tool to respond to “systemic oppression.”

People (including me) look at papers and statements like this and conclude that a lot of what’s happening at universities isn’t really research — it’s social activism dressed up as research, which need not be of good quality so long as it has the right ideological goals. Of course, this is not what all (or even necessarily most) professors in the humanities are up to. And I see arguments like this one from Tyler Austin Harper, a literature professor at Bates College, that faculty in the humanities generally aren’t even pleased with these changes — that it’s administrators at the top of these institutions pushing departments in politicized directions. He writes:

The reigning assumption is that scholars of color are disproportionately represented in activism-oriented fields such as “decolonial theory,” which means that deans—always seeking more brown faces to put on university websites—are more likely to approve new tenure lines in ideologically supercharged, diversity-rich disciplines. It is often faculty who are trying to safeguard their fields from the progressive machinations of their bureaucratic overlords. But faced with a choice between watching their departments shrink or agreeing to hire in areas that help realize the personnel-engineering schemes of their bosses, departments tend to choose the latter.

So this is another form of dishonesty.³ Because it’s illegal to use racial quotas in hiring, universities can’t explicitly admit to setting positions aside for candidates from underrepresented minorities. So, instead they use ideological screens and DEI statement reviews as a proxy for race. This approach has many drawbacks — in addition to involving a dishonest concealment of the university’s true objectives, it is of no use to black and Hispanic candidates who are not interested in “ideologically supercharged” areas of study, and sometimes it leads to the hiring of white candidates anyway, if they know best how to include the magic trendy words in a DEI statement. Harper has more to say about non-racial reasons why university administrations push the humanities in these directions — he says selling students the unfounded hope that the humanities will make them more effective social activists is one way to get them to enroll, for example — but none of it speaks well of universities’ commitment to honest research and teaching.

The last form of dishonest use of academic output I want to talk about is one that exploded as a huge problem when the COVID pandemic hit: subject-matter specialists using the guise of expertise in an effort to impose their values and policy preferences on the public. This phenomenon isn't limited to universities — some of the medical and public health professionals doing this were on faculties, some were at hospitals, some worked for the government, and some just posted a lot on Twitter. But I’ll simply say that several years of hearing “Science Says” prior to claims that weren’t science as such, but rather were applications of scientific claims through a very specific value framework I didn’t share — part-communitarian, part-neurotic, part-left wing — made me feel more negatively about experts, including those at universities, and I’m far from alone in that sentiment.

The dishonesty at universities extends beyond their research output. Let’s talk about admissions. Harvard has had a longstanding practice of using race as a factor in college admissions, producing a class that is less Asian and more black and Hispanic than it would be if they did not consider race. And they also have a longstanding practice of lying about it. Throughout the litigation over Harvard’s admissions policies, they didn’t just defend the appropriateness of race-conscious policies to promote diversity; they denied that they were discriminating at all. They played word games — similar to the “what even is plagiarism?” bit deployed by Gay’s defenders — arguing somehow that race could be used as a positive factor for admission without ever being a negative one, a mathematical impossibility when awarding a fixed number of admission slots.

It’s part of a broader dishonesty in how people in higher education insist on talking about affirmative action. Affirmative action policies prefer personnel of certain racial and ethnic backgrounds as part of an effort to alter the institution’s demographic balance — this is the point of affirmative action — but apparently it’s racist to admit this is what’s happening (or sometimes it would constitute an admission of illegal activity) so there’s a bunch of obfuscatory fudging of what the universities are really up to when they look at race. And since the Supreme Court’s ruling last year prohibiting race-conscious admissions practices, institutions across the country have been blatantly obvious about their search for ways to flout the law. It’s dishonest and, in the last few years, it’s been all over the news, which can’t have been good for public trust in universities.

And so that takes me to where this current news kerfuffle started: the hearing last month, where the presidents of Harvard, MIT and Penn got the shit beat out of them by Elise Stefanik,⁴ of all people. The university presidents were right on one of the points that was officially at issue in their exchange: if a university’s policies on expression are designed to follow the First Amendment, then even a call for genocide will only be prohibited in certain specific circumstances. The problem, in my view, is that the university presidents were full of shit about the if part of that statement: they do not, in practice, take a hands-off approach to all kinds of speech. Universities find ways to extend forbearance to speakers who break rules in the process of expressing favored ideas, and they impose sanctions on the speakers of disfavored ideas, often through process-as-punishment.

Gay herself was personally responsible for one such breach of neutrality discussed at the hearing: the firing of Prof. Ron Sullivan as a faculty dean at Harvard College over his legal representation of Harvey Weinstein. Officially, that’s not why Sullivan was fired — the university laundered students’ complaints about his legal activities through the notion that his behavior was disruptive to the “climate” inside Winthrop House, and Gay continued that laundering in her testimony. The case of Carole Hooven⁵ also shows that a claim as banal as saying that there are only two human sexes (note sexes, not genders) is unprotected in practice at Harvard.

Gay’s detractors have correctly identified Harvard as hostile to certain kinds of speech, but they had a variety of other complaints too — they want restrictions of certain kinds of speech about Israel, and they have broader objections to DEI and the adoption of more explicitly left-wing institutional positions at universities well beyond Harvard.⁶ But what they found, when searching for dirt on her, was unrelated to all of this: she plagiarized, repeatedly, in her body of academic work. She copied paragraphs of text nearly wholesale, without quoting the scholars whose text she used, and sometimes without even citing them. Anyone who went to college knows you’re not allowed to do that. It’s not just a rule — it’s a rule that universities are obsessed with and borderline sanctimonious about. They beat into students’ heads that they must correctly attribute their sources. If I copied like Gay did when I was a Harvard student, and if I got caught, I would have expected the university to require me to withdraw.

Dall-E 3 image generation for “academics undergoing a restorative justice process in which they atone for their plagiarism and redress the wrongs they committed against others.” Dall-E rejected the prompt “academics plagiarizing each other” on the grounds that “creating an image of academics plagiarizing each other would be inappropriate and unethical, as it promotes a negative and harmful activity.” It also rejected the prompt “academics being justly punished for plagiarism” because “I can't create an image that depicts punishment or negative consequences for individuals, even in the context of addressing unethical behavior.” It rejected a third prompt, “academics receiving a stern talking to for their plagiarism,” because this prompt “still centers around a negative and potentially harmful scenario.”

And that’s why it’s been so jarring over the past month to watch some academics and journalists announce a new, laxer standard on plagiarism that was unknown to us when we were students. What seems to be happening here is they are suffering from Chris Rufo Derangement Syndrome. That is, they know conservative activist Chris Rufo is a bad guy, and therefore the only way they can analyze a question on which he has opined is by assuming that the opposite of whatever he said was true. If Rufo says Gay plagiarized, then she must not have plagiarized, regardless of whatever near-duplicate paragraphs we can see with our own eyes. In addition to being a terrible approach to learning the truth, this mental model endows Rufo with tremendous power: If you have Rufo Derangement Syndrome, all Chris Rufo has to do to make you look like a total idiot is be right about something, once.

And so we got a lot of idiotic statements. Gay was merely guilty of “duplicative language,” the Harvard Corporation said, back when it was still defending her appointment. We were told that everybody does it: “Claudine Gay has resigned on the basis of a plagiarism charge that could have been leveled at anyone we know via the power of text mining applied without sound standards of how to assess the results,” wrote Jo Gludi, a history professor at Emory. (Really? Anyone we know?) Charles Blow even wrote in the New York Times that the expectation that the president of Harvard should not plagiarize (or should not be the subject of “questions about missing citations and quotation marks,” as he more verbosely described plagiarism) constitutes a “Wonder Woman problem” in which black women in positions of power “are trapped in prisons of others’ demands for perfection.”⁷

The demand that we should define academic honesty down in order to address the fact that Harvard’s first black female president is a plagiarist is insulting to academics of all races who don’t copy other people’s works, even the “banal” parts of them. And the insistence that this is how it always was, that actually this kind of copying is a standard industry practice, is just gaslighting. I went to college. I know that’s not true.

Eventually (belatedly) the Harvard Corporation realized nobody was buying these defenses, and that Gay’s position was untenable — that it would not do to have a university president stay in place when she’d done things the university kicks ordinary students out for doing.⁸ But that Harvard’s first instinct was to lie and obfuscate — to say there was nothing to see here — is reflective of the university’s overall posture of dishonesty and non-transparency. In fairness to the members of the Corporation, they usually do get away with it.

And all of this colors the way I feel about the conservative “war on higher ed.” Liberals in academia, including Claudine Gay herself, are very agitated about it. But it’s not clear to me exactly what one is supposed to be defending and why. I’d much rather see this industry do some introspection about why it’s losing public trust — again, as I discussed at the top, not just among conservatives — and what it could do to be more deserving of it. After all, if the strategy is simply to polarize views about universities and turn them into a liberal cause célèbre, that’s not only going to be a disaster for university budgets in red states; it doesn’t even look like it’s working as a strategy to build support among liberals.

Very seriously,

Josh

1. The lies endemic in psychology have mattered in the real world, but not necessarily in a left-wing direction. Arguably, the lies were often neoliberal: as Gideon Lewis-Kraus put it in the New Yorker, this flawed body of research promised that “kitschy modifications of individual behavior will repair the world” — e.g., you can get insurance policyholders to be more honest simply by moving the attestations they must sign to the top of forms instead of the bottom — and the alleged viability of these One Weird Trick approaches concealed situations where improving individual behavior actually requires a heavier regulatory hand, while distracting from more structural changes that might have been more effective in bettering society. And probably the most influential bit of BS psych research I learned about in college is one that came to form a cornerstone of DEI initiatives. This is the Implicit Association Test, a widely misused tool that purports to find implicit biases in human subjects who don’t even consciously know they hold them. The wildly over-broad lesson so many people and organizations took from IAT — that racism lies hidden within all of us and the way to cure it is to climb up our own assholes and look for it — has led to immense amounts of time wasted in woo-woo employee trainings while distracting from the more pernicious and still widespread problem of explicit racism. In other words, you can lay a lot of the blame for the rise of Robin DiAngelo at the feet of social psychologists peddling crappy research.

2. This excessive level of professional courtesy is one reason it took so long for Gino’s fraud to be discovered. After Harvard suspended Gino, Zoé Ziani wrote about how she had tried several years ago to raise concerns about data integrity in one of Gino’s papers, as part of her own Organizational Behavior PhD dissertation. But two of the members of her dissertation committee strongly pressured her to delete the section raising the concerns, essentially on the grounds that doing so would not be collegial, or would be disrespectful to more senior researchers. This is another sign of a culture that does not make research honesty a priority.

3. I would note that, while Harper makes a good argument that universities’ commitment to radical politics is largely insincere, the pressure he describes to create faculty lines in radical areas of study represents a very real financial commitment from universities toward the ideas they espouse insincerely. This is the worst of both worlds, in terms of how we should feel about what universities are doing with our money: their “decolonial” turn is simultaneously fake as a matter of sentiment and real as a matter of resource allocation. And if this trend in hiring goes on long enough, those attitudes will come to be quite sincere among a growing proportion of university faculty.

4. ::sigh:: who was a year behind me at Harvard

5. Speaking of the fact that I went to Harvard: Carole Hooven actually taught my discussion section about 20 years ago when I took The Evolution of Human Nature, a large lecture course taught by Profs. Richard Wrangham and Marc Hauser. Hooven was a dynamic and engaging teacher. Hauser, unfortunately, turned out to have been fabricating and manipulating data for some of the monkey studies we learned about in class. Oh well!

6. Those critics may be having some success at swinging the pendulum on pro-Palestinian speech from institutionally favored to institutionally disfavored — an outcome they probably saw as achievable because of these universities’ long track record of failing to uphold their purported commitments to neutrality.

7. A more reasonable response to Gay’s departure is one I saw from Keith Boykin, who wants to see analyses of other college presidents’ writings to see if they plagiarized too. Great, let’s do it; I’ll get the popcorn. My guess is this sort of copying is not that common, precisely because academics know it’s so easy to catch if you go looking for it. But I’m not sure, and I’ll be interested to know. What I do think there’s a lot of out there, undetected, is data fraud of the sort we’ve seen in the Hauser and Gino cases. Data fraud is a lot easier to conceal than plagiarism, and a lot more time-consuming to identify and prove, and as such I’d assume it’s more tempting to academics under pressure to publish. I’d love to see more journalistic resources devoted to uncovering it, too.

8. Plagiarism of a commencement speech — not even an academic paper — led to the resignation of the white, male president of the University of South Carolina in 2021. I guess he didn’t meet the Wonder Woman standard either.