‘OK, I get it. This makes sense.’

The proposal to rein in inflated grading at Harvard College has dominated campus discussion and ricocheted around the Ivy League and across higher education since its Feb. 6 release.

The plan is the result of five years of discussion and study, spanning multiple reports, considerable faculty input, and careful review of alternatives, culminating in more than a year of concerted work by the Undergraduate Education Policy Committee’s Subcommittee on Grading.

The faculty of FAS will vote on the plan in April and, if approved, the initiative would take effect for the coming academic year, with the before-and-after line clearly noted on transcripts.

The policy would limit flat-A grades to 20 percent plus four of the enrolled students in a course, meaning up to six A’s in a 10-person seminar or 34 in a 150-student lecture, with no cap on A-minuses.

At the same time, the Faculty of Arts and Sciences would adopt an internal measure for evaluating students for Harvard honors and prizes, shifting from GPA to an average percentile rank (APR) that would not be included on transcripts but would provide more meaningful data about relative performance.

That calibrated combination seeks to address the challenges that Amanda Claybaugh, dean of undergraduate education, laid out in a 25-page report last fall.

“Our current grading practices are not only undermining the functions of grading; they are also damaging the academic culture of the College,” Claybaugh wrote in “Re-Centering Academics at Harvard College.”

She emphasized that exhortations alone “won’t be enough, nor is there a single policy fix. But coordinated action — individually, collectively, and institutionally — can restore the integrity of our grading and return the academic culture of the College to what it was in the recent past.”

In real terms, the plan would pull grading back to 2010 levels, when A’s accounted for one-third of transcript marks. By 2025, more than 60 percent were flat A’s while the median cumulative GPA at graduation climbed from 3.56 to 3.83 in 15 years. Where it was once newsworthy for two students to tie for the Sophia Freund Prize for the highest GPA, dozens now share the honor. And summa cum laude, reserved for the top 5 percent, has ratcheted to a hair’s breadth from 4.0.

Beyond diluting the classroom experience, the proliferation of A’s has shifted stress to extracurriculars and penalized students from less-resourced high schools by magnifying the cost of a first-semester B, the subcommittee found.

We sat down with four subcommittee members to discuss the proposal: chair Stuart Shieber, the James O. Welch Jr. and Virginia B. Welch Professor of Computer Science; Paulina Alberto, professor of African and African American Studies and of history; Joshua Greene, Alfred and Rebecca Lin Professor of Civil Discourse; and Alisha Holland, Gates Professor of Developing Societies. This interview was edited for length and clarity.

---

Tell me about the feedback you’ve received since the proposal went public.  

Greene: In psychology, there’s a framework associated with the late Daniel Kahneman, contrasting thinking “fast and slow,” meaning intuitive reactions versus more deliberative judgments. People’s first responses to this proposal have been all over the map.

Some said, “Great, I’ve been worried about this for a long time, and I’m glad someone’s doing something.” Others said, “Ach, this is terrible, I hate the idea of a cap!” or “This is an assault on academic freedom!” or “This will lead to ruthless competition among students!” 

We’ve found that the more time people spend with the proposal and ask themselves how they would balance the competing considerations, the more likely they are to come around — not necessarily grinning with delight, but saying, “OK, I get it. This makes sense.”

Alberto: At the second town hall, Amanda asked the students to imagine a world in which it was normal to have an honors transcript with a mix of A’s and A-minuses and a few B-pluses and B’s, and to consider how much freer that would make them to follow their curiosity instead of chasing the 4.0.

I watched the students’ eyes widen. They were nodding in slow agreement, beginning to see what that would be like. That was a moment that made me realize the full potential of this plan.

Holland: A recurring theme from students and faculty has been concern about student mental health and whether the proposal will increase competition at a time when students are already pulled in many different directions.

Our current system causes high levels of competition and anxiety already. Many students feel immense anxiety around losing their perfect 4.0 GPA. A big thrust of this reform is to normalize getting an A-minus or B-plus.

Even very talented students are not always going to excel at every course, and that’s OK. Part of what college is about is exploring and finding your personal strengths.

Greene: Stress is largely a product of competition for opportunities beyond college. No grading policy can change that. But it can bring stress down a notch by getting people out of damage-control mode.

It’s like you start college with this shiny new car, and your goal for four years is to make sure it doesn’t get a single scratch. If that’s your attitude, you’re never going to go off-roading.

Many students see their academics as hoops to jump through as efficiently as possible: Get your A, then move on to the thing that will really distinguish you, like impressing someone who can get you into the consulting club.

We want to encourage students to invest in their classes instead of just treating them as a checkbox.

How have you experienced grade inflation at Harvard? Stuart, you’ve mentioned that a “revelatory” B as an undergrad told you not to pursue physics. You still managed to graduate with high honors in 1981 and join the faculty in 1989.

Shieber: Exactly. There was valuable information in B’s then. But today I give many more A’s than I used to, because I don’t want to burden my students with a distribution that differs from what’s expected.

I’m a culprit of grade inflation but not because I’m acting in bad faith. We have reduced freedom to grade. If I were to give grades now that I gave even 10 years ago, my enrollments would drop, and it would have even less of an effect on fixing grade inflation. We’re at the whim of a classic collective-action problem. 

Alberto: I experienced a real culture shock three years ago coming from the University of Michigan. That’s a top public university with incredible students. I had a much broader grade distribution there. Harvard students are amazing, but there’s room for adjustment.

The problem is the system. This is not about faculty who are not doing their job or students pressuring us against our will. We didn’t set out to impose a grade cap, but we concluded after considering many factors that this was the best way — perhaps the only way — to make this work.

Informed by the experiences of other schools, we decided to restrict just the A’s, to have the lightest touch possible.

Holland: When I came from Princeton in 2019, I was used to giving a wider range of grades, with A-minuses being common, and even maybe the median being a B-plus.

I asked my colleagues and teaching fellows here: What are students expecting? I was shocked to hear they were going to be heartbroken by an A-minus. It felt like the entire curve had shifted up. And that was before the pandemic, which created even greater pressure and expectation to grade in a lenient way.

I had a teaching fellow from Peru with very high standards who gave out much harsher grades than students were used to. She faced terrible blowback, with really rough teaching evaluations and a lot of student complaints. It was hard to witness. Even though I supported her interpretation, it came at a real cost to her.

It sounds like she was trying to follow the student handbook, which defines A-minus and above as work reflecting “excellent quality” and “full mastery,” with A reserved for “extraordinary distinction.”

Holland: That’s a good way to put it. She was trying to textually interpret the rubric, which is not how grades were given out in practice.

With this proposal reasserting that rubric, many students worry that “extraordinary distinction” is amorphous. How do you define it?

Shieber: We don’t define it, and that’s a conscious choice. The instructor needs to decide, and we want to preserve that autonomy.

Greene: Some say that an A should be about mastery in a narrow sense — if everyone learns the material, then shouldn’t everyone get an A? That’s an appropriate model in some contexts, but it provides no incentive for discovery.

At Harvard, we’re not just training people to reproduce what’s known. We are — we hope — training the next generation’s Nobel Prize winners, the people who are going to imagine new possibilities for humanity. And you don’t do that by merely reproducing existing knowledge and skills. We want to encourage students from day one to dig deeper, to be creative, to integrate knowledge from the course in new ways.

Top students should be able to answer questions on an exam that don’t look like the ones they’ve seen before but that draw on principles they’ve learned. The best students should be surprising their professors.

Alberto: Grading and pedagogical change go together. If we continue to have incentives for professors to teach classes that are less rigorous, we won’t create the space to rethink our assignments and coursework in ways that encourage students to reach for that extra degree of creativity, depth of understanding, or originality.

This requires students to up their game, but it also requires professors to ask more of themselves, and we’re going to have to think about how to communicate what extraordinary distinction looks like in our classrooms.

I like to share student writing and have students provide feedback on what’s working and what can be improved. Sometimes when students are presented with the work of someone who really took it to the next level, that’s very powerful. It doesn’t just set an example; it also begins to give them the tools to figure out how to get there themselves.

Holland: Most professors have a good sense of what extraordinary work looks like, even though we might not be applying it with our undergraduate grades. I do it when I review academic work for journals, or when I write letters of recommendation and need to explain why a student is in the top 1 percent or 10 percent of the class. We’re just asking professors to use that judgment and apply it to the courses they teach.

There’s also an interesting gap around senior theses. At least in my department, when we grade senior theses, we’ve continued to use the FAS rubric. A summa thesis represents extraordinary distinction and with revision could go on to be a published academic work.

Professors take that very seriously, and when we hand out grades, many students end up in tears, because it’s the first time they’ve been graded applying the rubric as written.

The Bok Center will be critical in helping instructors, so they aren’t drawing arbitrary lines between A and A-minus but are rethinking their courses to allow for even deeper learning and ways for students to distinguish themselves.

Princeton tried to address grade inflation in 2004 but abandoned the effort in 2014. What did you learn from them, and how is this proposal different?

Shieber: When Princeton started, there was already a long history of grade inflation, but they hadn’t gotten to where we are today, with grade inflation producing dramatic grade compression.

As grades inflate, they eventually bump against the wall of the 4.0. You can’t have a higher GPA, so you end up with more and more 4.0s.

As they pile up against that wall, we get less information, less signal, and more noise. When grading is substantially noise, it’s unfair to make comparative decisions — for honors and prizes — based on GPA. At a certain point it becomes immoral.

The other big difference was Princeton limited all A-range grades, including A-minuses, to 35 percent.

But the cap was not applied to individual courses. It applied to all the courses in a particular department and was based on a running average over multiple years. So it wasn’t really a cap, it was more of an exhortation, and then you’re still subject to the same challenges.

 What do you say to those who say this will increase student competition?

Greene: Some worry about intense competition among students vying for scarce A’s. But our understanding of the Princeton experience is not that students were backstabbing each other. And that makes sense from an evolutionary perspective.

In my GenEd class “Evolving Morality,” I emphasize that competition is what creates cooperation. Under natural selection, molecules form cells, cells form organisms, and organisms form societies, because teamwork is an essential strategy for surviving and flourishing.

In this context, if the whole class can’t get A’s, you’ve got more reason — not less — to form a study group.

When people say the Princeton policy made students feel highly competitive, my interpretation is they felt like they couldn’t get the grades that they wanted or deserved. In crafting this proposal, we took that very seriously. Asking Princeton undergrads who have been A students their whole lives to suddenly get mostly B’s is asking a lot. (Although that was once the norm at Harvard.)

Our proposal is less draconian. With a limit only on A’s, theoretically all letter grades at Harvard could be A’s and A-minuses.

How did you decide on 20 percent? And why did you recommend capping it at once, instead of phasing it in?

Holland: When people see the headlines, it sounds radical. But the main thing to understand is that the 20 percent plus four results in roughly 30 to 35 percent of the grades in the College being straight-A grades, because we teach a lot of small classes. And there’s no cap on A-minuses. That’s a far more moderate — though still substantial — change.

We discussed starting higher and lowering the cap by a certain percentage each year to reach the target, but there’s no way to easily communicate that kind of grading system to the world of employers and postgraduate programs.

And if you’re a student experiencing four different caps in four years, it would be a mess. We wanted to make a single change that we could communicate in an easy and transparent way that feels substantial, sustainable, and reasonable.

With more A’s available per student in smaller classes, will the APR balance enrollment between seminars and large classes — where a B-plus or A-minus might yield a better relative rank — in addition to providing more “signal” for determining honors?

Greene: Right. Our overall goal is not to send students whiplashing back and forth between small and large classes, but to make students not think about it at all and just take the classes that they want to take because they’re excited about the material. These countervailing incentives should nullify each other and put the emphasis on the intellectual content of the course.

The proposal quickly made waves in the Ivy League. Yale College Dean Pericles Lewis said he doesn’t want “an A at Yale to be seen as a lesser A.” What are the wider implications for this proposal at a fraught moment for higher education?

Greene: I hope we’re showing what good governance looks like. Harvard has its critics — many operating in good faith, others not so much. Some of our critics will gleefully point to grade inflation and say, “See, this is a deteriorating institution.” In any case, we can’t just ignore what they’re saying and put a smiley face on our problems.

To govern with integrity is to acknowledge that we have real challenges and to study them as thoroughly as possible from as many perspectives as possible to find the best solutions. We have to face up to tough tradeoffs, engage in persuasion with people who disagree, and try to reach a consensus. So, yeah, it’s about grades and education, but it’s also about what it means to have good governance in challenging times.