Measurement and Desert: Why Grades Cannot Be Deserved

One of the most ubiquitous and uncontroversial ideas concerning desert—both in ordinary discourse and in the philosophical literature on desert—is the idea that students can, and often do, deserve some grade or other. And furthermore (typically), their grade is deserved on the basis of their performance in the class, broadly speaking. A student that performs well on their exams, completes all of their assignments, is active in class discussion, etc., deserves a better grade than a student who performs poorly in the class, for example. Or so the thought goes.

Now, one can imagine various ethical arguments against this idea. Perhaps, for instance, students don’t deserve grades because the practice of grading, itself, is harmful, or because they conflict with the proper aims of a good educational system. But the case I want to make here is that, whether or not these ethical arguments succeed or not, there is a more basic, metaphysical problem with the idea that students deserve grades. And so, what I want to argue is that grades can’t be deserved, regardless of whether the practice of grading, ultimately, is justifiable. (And so, if the argument is good, then when a student insists that they deserve a better grade, teachers have a very obnoxious, but accurate, response available to them: “I’m sorry but that’s impossible.” At this point, the student can simply reply that they ought to get a better grade, in which case, I have nothing to offer.)

The basic problem stems from what I call the “Measurement Constraint” on desert, which is, roughly, the idea that a thing cannot be both what is deserved, and a measure of how much one deserves it:

Measurement Constraint: If M is a measurement of a desert basis B, M cannot be

deserved on the basis of B.

I will argue for the Measurement Constraint shortly, but for now, let me simply explain and illustrate the idea. A “desert basis” is some fact about, or property of, someone that explains why they deserve something. So, for instance, if Walter (my cat) deserves a treat for being such a good boy, his being such a good boy is the desert basis for his deserving the treat. What the measurement constraint entails is that if I had some device which could measure just how good a boy Walter is—let’s say the device reads “100% good boy”—then Walter could not deserve the reading of “100% good boy” on the basis of his being such a good boy. He deserves the treat, but not the reading. 

Or consider another context where we measure in order to determine what one deserves. Suppose you are competing in a sprint, and you finish the race in 10.5 seconds, which ends up being the fastest time. In that case, we will say that you deserve the top prize. The Measurement Constraint does not conflict with this idea. But the question is not whether you deserve the prize, but whether you deserve that the clock read “10.5” on the basis of the speed of your run. Intuitively, this seems strange to me. You might fail to get what you deserve if the clock is inaccurate, and the event managers might have professional obligations to try to measure your time accurately, but none of this would show that you deserve that the clock read “10.5 seconds.”

Intuitions aside, the Measurement Constraint entails that you do not deserve that the clock show your time on the basis of your run—it measures the speed of your run, which is the basis on which you might deserve a prize. Now, the point here is that the analog to grades is not the prize, but the clock. After all, grades are, essentially, measures of performance (where “performance” can be understood as broadly as one likes, to include whatever we might want to measure with a grade). But grades are taken to be deserved on the basis of student performance, which is precisely what they are a measure of. Plug this into the Measurement Constraint, and we get the result that grades cannot be deserved on the basis of student performance, because they are a measure of student performance.

So far, I have just been illustrating the Measurement Constraint. Let me now actually give an argument for it. There are two crucial observations the argument requires. First, desert bases have an explanatory and evidentiary function. If someone asks, “Why should I think that Walter deserves the treat?” I will immediately respond by appealing to the desert basis: “Because he’s such a good boy!” The desert basis both explains, and is our best evidence that someone deserves something. Next, we need to consider the evidentiary role of measurement in cases where we are measuring a desert basis. In the context of a sprint, for example, we use a clock because it is our best evidence concerning the speed with which someone’s sprint. Similarly, we measure a student’s performance with a grade because once we grade, that is our best evidence for the quality of the student’s performance (whether it be on something like an exam, or an entire course). 

Now, suppose, for reductio, that a student deserves an A in a course. She deserves this, of course, on the basis of her excellent performance in the class. But how do we know that her performance was so excellent? We look to her course grade—after all, that is our measurement of the quality of her performance in the class. But then, given that her excellent performance is our best evidence that she deserves an A, and her grades are our best evidence that her performance is excellent, we arrive at the idea that our best evidence for thinking that the student deserves an A is that she got an A. But this doesn’t seem like a good justification at all. The fact that the student gets an A would only be good evidence that she deserves an A, it would seem, if we already assume that she got what she deserves. But we can’t make that assumption here, because we are trying to figure out if she got what she deserves. Thus, we have a contradiction: the fact that a student got an A both is and is not good evidence for thinking that she deserves an A. I propose, then, that we should reject the idea that grades are deserved, and accept the measurement constraint, which we can get by generalizing from the above argument.

Supposing this argument is correct, and students can’t deserve grades, two questions remain: First, if students can’t deserve grades, then why is it so natural to think that they can? Second, does it matter if students can’t deserve grades?

As to the first question, I think the explanation is simply that, given the stakes that are involved with grades, grades can quite naturally feel like a reward or a punishment, which are paradigm objects of desert. When a teacher gives a student a bad grade, they are, very often at least, quite literally harming a student both in the long and short term, while a good grade is a benefit to a student in the short and long term. Even worse, the goodness or badness of the grade (and thus the magnitude of the apparent reward or punishment) is proportional to the goodness or badness of the quality of the student’s performance. This kind of proportionality is a feature that we often see in other contexts where a reward or punishment is thought to be deserved, as with criminal punishment, or the desert of income. It’s quite natural, then, that getting a bad grade should feel like a punishment, and getting a good grade should feel like a reward.

Now, if students can’t deserve grades, does this matter for the practice of grading, itself? Strictly, no. All that follows is that we should refrain from saying that a student deserves some grade or other. It can (and plausibly is) still the case that students ought to be given grades which accurately reflect their performance.

On the other hand—and here I’m speculating—I’d like to hope that removing desert-talk from the context of grading would subtly encourage a more sane and healthy view of grades. Being deserving of something is, of course, a moral property, and thus, it is natural to think that deserving some grade or other is a reflection of some deep part of one’s character. And so it might be that desert-talk in the context of grading contributes to a student’s internalizing that they are good or bad or exceptional or poor at some subject area (or at school generally), where these judgments are meant to reflect some static aspect of their intellectual or moral character. I take it for granted that these kinds of judgments tend to be unhealthy, both for those who get good grades, but especially for those who don’t. But in fact, grades are not rewards or punishments, but are simply measures of performance, where one’s performance can be affected by any number of variables, most of which are not fixed facts about one’s character or intellectual abilities.

Toby Napoletano

Toby Napoletano is a lecturer at the University of California, Merced. His specialization was originally in the philosophy of linguistics/language, but he now mostly works on issues surrounding desert, meritocracy, and human rights.