Langdell Library at Harvard Law School in March 2020.
Langdell Library at Harvard Law School in March 2020. By Ryan N. Gajarawala

The Academic Policing of Academics on Policing

In 2022, professors Christopher Lewis and Adaner Usmani argued that to reduce violent crime, the U.S. needs to drastically shorten its prison sentences — and increase its police force by half a million officers. Their ideas soon become a flashpoint of online discourse.
By Jem K. Williams

In 2022, the editors of the American Journal of Law and Equality invited Harvard Law School associate professor Christopher Lewis to contribute an article on the subject of inequality in policing. Little did they know, the article Lewis would submit — co-authored with assistant Sociology professor Adaner Usmani ’08 — was to become a flashpoint of online discourse.

In their piece, “The Injustice of Under-Policing in America,” Lewis and Usmani argue that to reduce violent crime, the U.S. needs to drastically reduce its prison population via shortening sentences — and increase its police force by half a million officers. Their work cited evidence that crime is reduced more by increasing the probability that an offender will be caught and detained than it is by enacting stringent repercussions for those who are caught.

Harvard Law School professor Randall L. Kennedy, the editor of the journal who reviewed Lewis and Usmani’s article prior to publication, said he was “grabbed” because “it was such a counterintuitive article.”

He said the article came out around a time — just two years after the death of George Floyd — when the idea of abolishing or decreasing police presence was gaining popularity as a route to racial justice. That Lewis and Usmani were self-professed socialists only made the argument more perplexing.

Perhaps unsurprisingly, then, their prescription for almost doubling the police force provoked outrage.

Civil rights lawyer and Harvard Law School graduate Alec G. Karakatsanis, perhaps the most vocal objector, said the piece was brought to his attention by several students and professors.

“They weren’t really in a position to comment on it,” Karakatsanis explained, “but they thought there were serious problems with it.”

“So I took a closer look at it, and was really shocked and appalled by the lack of intellectual rigor in it,” he said. So he took matters into his own hands.

Karakatsanis wrote out his response via a combination of posts on Substack and X, alleging that the article made flawed arguments, lacked rigor, and amounted to “dangerous, de-politicizing nonsense dressed up as fancy scholarship.”

“The article reveals systemic ethical and intellectual flaws in elite academia,” he wrote.

Lewis and Usmani quickly disputed these claims. And that was only the beginning.

In what followed, not only did the two sides disagree about the specifics of the professors’ argument, they also disagreed about which side was truly pushing against the status quo. At the heart of their exchange lie questions central to academia itself: How can we disagree productively? And, on pressing social problems, how much should theoretical research impact real-life policy?

Counting and Costs

Lewis and Usmani’s article begins with a provocative fact. The United States, by their estimate, has far fewer police per homicide and per incarcerated person than other comparable countries.

This was the first point on which Karakatsanis disagreed. In their count of U.S. police, Lewis and Usmani only include publicly employed officers in the United States. Karakatsanis argued that a full picture of policing in America should include federally employed officers such as the FBI, the DEA, and border patrol, as well as private officers, such as those employed by universities and residential communities. Including these categories would increase the count by hundreds of thousands.

In a reply two weeks later, Lewis and Usmani countered that changing the number of police officers per capita to the higher estimate did not change the crux of their argument. Even using Karakatsanis’ count, they were still able to demonstrate that the United States has a lower police to homicide ratio than other comparable countries.

Even more troubling to Karakatsanis was the lack of weight Lewis and Usmani placed on the costs of policing. In his view, the professors fail to account for the disproportionate harm that police have on disadvantaged demographics, such as people of color and low-income individuals.

The professors addressed Karakatsanis’ criticism in their response by stating that crime can engender some of the same detrimental social effects as policing on disadvantaged people.

But according to Aaron J. Chalfin, a criminal justice researcher who is cited by Lewis and Usmani, meaningfully quantifying social costs can be difficult.

“The benefits are easier to measure because we have good national data on crime. The costs are going to be a little bit harder to measure — which means they’re going to be understudied,” he said. “What that means is that we are going to have a wide range of opinions about how big the costs of policing are.”

Chalfin describes how various types of research are vulnerable to error depending on the numbers chosen for calculations, but he says those discrepancies are a common problem in empirical research.

“Statistics is about estimation,” he said. “There’s not perfect accounting of everything.”

‘Our Intellectual Culture’s Worst Features’

Lewis and Usmani expected that some scholars might disagree with points of their argument, but not in the manner Karakatsanis did.

When they responded, they were “not interested in engaging in a Twitter-style debate,” so instead they opted for a letter uploaded to DropBox and linked on Lewis’ website, the pair wrote in a statement to The Crimson. Though it was titled “Reply to Alec Karakatsanis,” Krakatsanis said they did not send him the letter directly.

The professors began by calling many of his objections “unserious” and said “we would have preferred to ignore them.”

“But because his accusations of scholarly malpractice have garnered some attention, it has seemed necessary to set the record straight.”

Sociology professor Adaner Usmani '08 co-authored 'The Injustice of Under-Policing in America'
Sociology professor Adaner Usmani '08 co-authored 'The Injustice of Under-Policing in America' By Jonathan G. Yuan

Besides addressing Karakatsanis’ methodological critiques, the professors also took issue with the format of the lawyer’s reproach.

Karakatsanis believed the argument was dangerous because it could help advocates for a police state in “a time of rising fascism.”

Yet by suggesting that their research was “dangerous or could cause harm,” the professors claimed, Karakatsanis had “pandered to our intellectual culture’s worst features,” including an “anti-intellectual tendency” among students “unable to engage people who disagree with them in good faith.” (Usmani is a member of the Council on Academic Freedom at Harvard.)

These students, they claimed, “look to institutions to censor views which make them uncomfortable.”

The professors also noted they “have no attachment to our proposal if someone can convince us that we are wrong.” They were merely unconvinced by Karakatsanis.

“We hope for better debates in the future,” they concluded.

Karakatsanis, a self-proclaimed outsider to academia, disagreed with the professors’ chosen discursive setting. According to him, responding outside the confines of academic cordiality was more conducive to productive disagreement.

So if both the guidelines of academia and the wide ranges of social media weren’t the right setting for this debate, where might that setting lie?

While Lewis and Usmani do not view social media as the optimal space for productive discourse, they’re more optimistic about other spheres. The professors have debated their argument on the New York Magazine podcast and dissected it in a class called Sociology 1185: “Crime, Punishment, and Policing in an Unequal America.”

The class was in fact built entirely around constructive engagement with opposing perspectives.

They accomplish this type of engagement with the deployment of two principles: “First, we require students to argue positions that they may not agree with themselves. Second, we require students to develop these arguments in number-premise form (i.e., reasoning carefully from premises to conclusion),” they wrote. They claim these principles enable healthy debate by “detaching people’s identities from the positions they’re arguing.”

Madina A. Jenks, a third year Law School student who took the class in fall 2022, explained that one of the main goals of the course was to emerge from debates “with a fuller understanding of the topic than when you came in — without any real hurt feelings.”

Students read arguments from a variety of perspectives, including drafts from Lewis and Usmani’s upcoming book — which is based on their 2022 article.

“They wanted to hear some input from us, and they want to hear honest pushback on their ideas,” Jenks said. “Which there was plenty of.”

“They don't necessarily want to push you in any kind of direction,” she added.

Across published anonymous course reviews, students consistently encouraged others to “Take this course,” with some even calling it the “best class I’ve taken at Harvard.”

However, Lewis and Usmani’s teaching style is not without its objectors.

Jordan A. Rogers, a practicing attorney who took Lewis and Usmani’s class in his third year at the Law School found the course’s focus on a “philosophical understanding” of policing frustrating, because class discussions “didn’t necessarily take into account the reality of our system.”

In a second chain of posts on X, Karakatsanis also took umbrage with Lewis and Usmani’s pedagogy. He obtained an image of a mandatory first year exam question from one of Lewis’ law school classes, which asked students to “write an essay laying out and evaluating what you see as the most powerful objection(s)” to Lewis and Usmani’s proposal for increasing the police force — while assuming that shifting funding from prisons to police “is the most efficient revenue-neutral way to reduce homicides and other serious crimes that is presently feasible” and “the rate of police killings is unlikely to rise in such a scenario.”

Karakatsanis objected to how the mandatory nature of the exam forces students to engage with an argument that he views as flawed.

For him, it seemed, classroom discussion may be limited by expectations placed on students and the power imbalance that accompanies the professor-student dynamic.

In their written statement, Lewis and Usmani emphasized that they received positive feedback in student evaluations, with students specifically citing how the course allowed them to “engage with us as equals.”

They also noted that most professors learn from their students — “If they don’t, they’re probably not listening to them.”

“In our book we plan to attribute (with permission) certain positions to the students who most powerfully defended them in our class,” they added.

‘The Game of Elite Academia’

In February of 2023, Lewis and Usmani’s article was picked up by the San Francisco Deputy Sheriffs’ Association, who tweeted it out with the caption, “I recommend you read this in its entirety.”

It seemed like something Karakatsanis might have predicted.

In an interview, Karakatsanis said the combination of the Harvard name, the academic nature of the publication, and the progressive beliefs of Lewis and Usmani all lend credibility to their argument.

But is Karakatsanis’ fear of Lewis and Usmani’s paper falling into certain hands something they should be held accountable for mitigating? And should their research have the direct power to impact implementation of policy?

As a fellow researcher, Chalfin is wary of making policy recommendations.

“I’ve written papers that consider the impact of police hiring on crime. I wouldn’t tell someone that they should hire more police. That’s really a decision that people with skin in the game should be making,” he said.

Chalfin points to the limited ability of social science research to reach definitive conclusions, noting that “research can’t predict the future. It can only study the past.” To maintain responsibility for his research, Chalfin said, he frequently talks to journalists, notes the limitations of his findings, and never makes policy recommendations.

Though “lobbying groups will always pick up papers that suit their needs,” he tries to teach his students to engage with a wide range of scholarly material.

“I think our role as academics should just be to be scientists and to try to be intellectually honest about this stuff because other people are going to play the role of being advocates,” Chalfin said.

Lewis and Usmani agreed that “academics have a responsibility to contribute to the public good.”

“This is why we write about morally important issues about which people disagree,” they wrote in their statement. They added that they try to “guard against misinterpretation” by being transparent about their reasoning.

In Lewis and Usmani’s “Reply to Karakatsanis,” guarding against misinterpretation also meant being transparent about their political philosophies.

“We are both socialists: our central concern has always been the plight of the most disadvantaged,” they wrote, after emphasizing their support for “social democratic redistribution that would address the root causes of most crime.”

Yet because, according to them, this redistribution is politically infeasible and highly expensive, they opted instead for a “revenue-neutral” proposal.

Lewis and Usmani are no strangers to this kind of disassociation between political alignment and research. In the 2022 class, Jenks recalled a reading arguing in favor of reparations from a libertarian perspective.

Karakatsanis takes a more pessimistic approach to this kind of academic disjuncture — calling it “the game of elite academia.”

He describes this “game” as “presenting yourself as progressive so you can sort of fit into the charade of the liberal academic elite, but not be too threatening to the people who really control how these institutions function.”

“It’s part of the pretense of these as institutions that actually care about knowledge, and learning, and teaching, and rigor,” he explains, “when in fact what’s going on is they are institutions for the reproduction of social hierarchy and for preserving the status quo.” (A Harvard spokesperson did not respond to a request for comment.) But he clarifies that this is not unique to academia.

“In any profession, if you serve the interests of people with power and money, you’re going to do well. That’s just the way our society works,” he said.


After everything, including an onslaught of responses ranging from “extreme enthusiasm to death threats,” the professors never considered retracting their article — which will inform the basis for their upcoming book. Their editor at the American Journal of Law and Equality, Kennedy, still stands by the article’s publication as well.

Although Kennedy does not agree with every perspective published in the journal, he tries to keep an open mind.

“I think a perfectly fine thing is to publish something that you disagree with very strongly, but that shines a light on something or causes people to rethink conventional beliefs,” he said.

— Associate Magazine Editor Jem K. Williams can be reached at Follow her on Twitter @jemkwilliams.

CommencementYear in ReviewThe ScoopEditors' Choice