Talent and Bias

By Sam Duncan

In this post I want to dig more deeply into some points about the deeply worrying effects of our belief in talent on our students made by Allison Gopnik and Eric Schwitzgebel. As Gopnik notes there’s good reason to think that there is some causal relationship between the fact that philosophy puts so much stock in talent and our lack of gender and racial diversity. And Schwitzgebel paints a very plausible picture of one way that this likely plays out in practice; as he notes it’s a lot easier for a young white guy to get away with various ploys that allow him to seem smart than it might be for a woman or a member of a minority group (It’s also a lot less risky). However, I think there are deeper ways that a belief in talent sets up bias in our field.

Let me begin with a question: When you look at student papers or consider how to respond to student comments, questions, and challenges in class do you think about whose paper it is or who made the comment? I’m betting that you do. If you make much of an effort to know who your students are it’s hard not to do this, and to a great extent this is part of getting to know our students. But this also opens up a danger of bias and believing that talent determines success in philosophy makes this danger much worse.

When we think about who wrote what we’re reading or who’s speaking it is all too easy to treat students unfairly without meaning to, and a belief in talent makes this much easier to do. If we sort our students into “talented” and “untalented” boxes (or even something more fine grained) then it is very hard indeed to hold them to the same standards. When a student we deem  to be “talented” student writes or says something that seems wrong or just surprising, one gives them much more of a benefit of the doubt and tries much harder to make sense of what they’re saying than one does when an untalented student does so. When students say or write something truly surprising that we are initially unsure what to make of, there is all too much temptation to interpret it in a charitable light when it comes from talented students. 

On top of all this one must also take account of the fact that students themselves will no doubt react to the subtle, and often not so subtle ways, that they are treated differently based on judgments about talent. Students deemed talented and treated charitably become more willing to speak in class, more enthusiastic about the material and willing to contribute positively, and more willing to put in effort, while those deemed untalented shut down in discussion, stop putting in work, or even try to trip up the professor or disrupt class in other ways. 

All this is bad enough but let us take a step back and think about how we philosophers often decide who is talented and who isn’t. We judge the students who give the “right” answers on papers and tests or make “good” or “insightful” points in discussion to be talented and those who don’t do these things to be less talented or untalented. 

Even if one thinks there is such a thing as philosophical talent that plays the role that many people think, there is every reason to believe that these judgments are quite fallible and often wrong. For one thing, they are often made very early on in the semester on the basis of very limited evidence. More importantly though, it just isn’t nearly as clear as we would like to believe what a right answer or “good” or “insightful” comment is. I have changed my own mind about just how powerful a philosophical argument is or how good various objections to it are a number of times. For me at least, this happens especially often with objections against and supposed counter-examples to philosophical arguments. On several occasions I have come around to the view that student objections I had dismissed as misguided, “missing the point”, or even just plain dumb were much more powerful than I had given them credit for.

For example, take Thomson’s “A Defense of Abortion.” For a long time, I pretty haughtily dismissed a number of common objections about just how bizarre and unrealistic the violinist and people seed cases were to be thoroughly mistaken. The “talented” students were the ones who focused on the questions and objections I’d had when I first read the piece, which were mainly about Thomson’s claims about the duty to rescue. The untalented ones were the ones who got hung up on the fact that say both the violinist and “people seed” cases were at odds with basic biological facts. However, I’m much less convinced that these are necessarily bad objections. They might not be fatal but I no longer think that we can simply overlook them. (The main reason for my change of heart is that I realized that it seemed to be inconsistent to treat the fact that Nozick’s arguments ignored important facts about history and economics as damning objections to his arguments while dismissing similar objections to Thomson).

Now I deliberately chose this example for its politically and culturally loaded nature. I worry very much that philosophers’ judgments about what are and aren’t good arguments as well as good objections to those arguments are determined by our political views and our background experiences and beliefs. This is a hard problem with no easy solution, but focusing on talent makes it much much worse. Students with backgrounds and values very different from our own are likely to approach standard cases in philosophy classes differently than we do. I see this constantly in teaching students in the community college system who are more diverse in pretty much every way than most student bodies. When they do there will always be a temptation to dismiss them rather than giving them the hearing that the principle of charity demands. Judging a student to be untalented (or let’s be blunt, dumb) makes this so much easier to do; there’s no point in being charitable. Of course not thinking in terms of talent will not magically make us treat other viewpoints fairly much less charitably. Those are hard things indeed and can take a lifetime of practice. What it will prevent though is a sort of snowball effect where being unfair or uncharitable to a student’s views in one case makes it all the easier to do so in another.

Just to be clear I am not advocating a suspension of judgment when it comes to every single argument students might make. When students make claims we know to be factually wrong we can and should correct them. And some arguments are bad ones and some objections are misguided. However, even in these cases we should avoid any judgments about talent as far as we can. After all, who among us hasn’t made a bad argument or gotten a point of fact wrong? We would not want our colleagues, students, and other conversation partners to dismiss anything we might say in the future because of that. Even when students do say stuff that’s clearly wrong or mistaken we should do everything we can to treat them with the charity we want others to extend to us. Not sorting them into a mental box them makes them easy to dismiss or ignore in the future is one small step in that direction.

Comments

[Tammo]

I think discussing the role the idea of talent plays in philosophy is extremely important. There was a study a few years ago that found that the expectation that a successful researcher needs to be intellectually talented is by far the greatest in philosophy (https://www.science.org/doi/10.1126/science.1261375). It also found that this kind of expectation correlates with a smaller share of women in a field.

I think the role of this expectation cuts both ways. On the one hand, I think you are right to point out that we can fault students by sorting them in "talented" or "untalented" boxes. I also think there's an issue with whether students perceive themselves as talented. If you think that you are talented and you think that talent is important to being a good philosopher, you'll be more likely to be active in class discussions, more likely to be in an 'open' or playful made of thinking during these discussions, more likely to have some initial confidence in your own ideas, more likely to bounce around these ideas with your peers or your instructor, and because of all of that you might well write a better paper. (There's an issue with over-confidence too, but I find that to be a far less common problem.)

I think your example of that is great: maybe just saying that Thomson's violinist is biologically unrealistic (but not saying why that might be a problem) is a bit flat-footed. But a confident student might turn that into a paper about whether philosophical thought experiments need to be realistic, or whether they can be a bit more far-fetched. So we should encourage a student who has that concern to keep pursuing it.

(I don't have all the answers about how to fix all this, but I think active learning goes a long way in making students who consider themselves untalented build up some confidence.)

[Sam Duncan]

Tammo,
I think these are all good points. I do think its important to ask whether students perceive themselves as talented or not. But even there I think that professors' do a lot to set up students' own judgments both of their own levels of talent and how important talent is in a field. On the latter issue I think of myself and math. None of my teachers ever flat out said "talent is the deciding thing in math" but they just really conveyed that impression all kinds of subtle ways. And I picked up on it and decided that since math wasn't easy for me I must be untalented and therefore hopeless at math and so there was no point in "wasting" effort on math.
I think your last point about overconfidence is really interesting too and I think overconfidence can be a massive problem. It's much rarer but the really overconfident student can really wreck discussions and a classroom atmosphere. I honestly struggle to do with those students. How does one convey to them that things aren't as easy or clear as they think and that they need to take dissenting voices more seriously without getting nasty? Then again when it gets to a point where they're stepping all over other students and monopolizing discussions getting nasty may be the lesser of evils. I dunno. Maybe we should have a discussion here about what people do with overconfident students. I'd like to see those suggestions myself.

[anonymize papers]

"Let me begin with a question: When you look at student papers or consider how to respond to student comments, questions, and challenges in class do you think about whose paper it is or who made the comment? I’m betting that you do. If you make much of an effort to know who your students are it’s hard not to do this, and to a great extent this is part of getting to know our students. But this also opens up a danger of bias and believing that talent determines success in philosophy makes this danger much worse."

It cannot be done for comments made in class, but papers should really be graded without knowing the identity of the author, especially in lower-level classes. In almost all cases, any benefit of knowing the identity is greatly outweighed by the benefit of limiting bias.

[Sam Duncan]

anonymize papers,

When possible I agree with this. But for most of my classes it's not feasible. In pretty much all of my classes I make students do multiple drafts of their papers. This makes it much much harder to do, especially since a fair number of them want to talk to me about the first draft and my comments. I've went back and forth on requiring students discuss first drafts with me in office hours (it has its advantages but takes sooo much time). Anyway, when I do that it's pretty much impossible to anonymize in any meaningful way since much of what I'm grading on is whether they took my comments and the overall discussion into account and how they responded to them. I know both of these things are a little unusual at four four year schools but they're kind of par for the course in humanities classes at community colleges like mine.

[anonymize papers]

Hi Sam,

This might be one of the exceptions. But it might not: the halo effect is a real effect in a lot of areas. For example, if I recall, the evaluation of something known to be associated with Harvard actually goes up when the word 'Harvard' appears and down when it doesn't. I have no idea whether the halo effect is operative in grading things like multiple drafts, but it might be. And it's still possible to anonymize assignments with multiple drafts. For example, perhaps each draft can be given three bullet points as suggestions, and these need to be included at the top of the subsequent draft. Again, this might be an exception to anonymization, but I don't want us to be too quick to think it is.