Suppose a student fails a math test. Casual observers will often announce, “He can’t do the math.”
Or suppose a country has a horrible corruption problem. Casual observers will often announce, “The government can’t solve this corruption problem.”
In each case, I detect a casual logical fallacy.
Namely: If person X actually does Y, we can legitimately infer, “X can do Y.” But if person X does not do Y, you cannot legitimately infer that they can’t. Maybe they don’t do Y because they can’t do Y. Maybe they don’t do Y because they choose not to do Y.
What’s the real story? Figuring that out requires further investigation. Before you declare that, “X can’t do Y,” start with this simple checklist:
Step 1: See if the actor in question even tried to do Y.
Step 2: If the actor tried, examine how hard he tried.
Step 3: Look at how successful comparably-able actors are when they try their very hardest.
Thus, before you say that a kid who fails a math test “can’t” do it, you should examine (a) whether he even tried to pass, (b) if so, how hard he tried, and (c) the pass rate for comparable students who try their very hardest.
Similarly, before you say that a country “can’t” solve its corruption problem, you should examine (a) whether the country even tried to do so, (b) if so, how hard it tried, and (c) the success rate for comparable countries that try their very hardest.
You could respond, “Running such an investigation sounds awfully difficult. How do we know whether someone tried? How hard they tried? How do we find comparable actors? How do we know whether the comparable actors ‘tried their very hardest?'”
I fully agree. Knowing what someone can’t do tends to be hard. Which is precisely why you should make such claims with care.
The most egregious misuses of “can’t,” however, come when even token efforts reliably produce success. When someone says, “He can’t stop drinking,” or “He can’t stop cheating,” common sense revolts. Maybe you aren’t smart enough to pass a math test. But what skill does it take not to consume a beverage? To abstain from sex? Just don’t do it – and it is done.
READER COMMENTS
Phil H
Jan 15 2020 at 10:14am
I happily welcome Bryan Caplan to the progressive fold. Because he’s absolutely right. The fact that a person doesn’t do something doesn’t prove that they can’t. It shows that under present circumstances, they don’t. So if the thing is a thing that we want to happen (e.g. ending corruption, passing the test), the correct response is to change the circumstances. We could make the tests fairer and the teachers nicer. We could pay the government workers more, to reduce the incentive to be corrupt.
Or, like rant about it on Twitter. Also a well-loved solution to social problems!
KevinDC
Jan 15 2020 at 4:38pm
I know this comment of yours is at least partly tongue in cheek, but you do realize how much of the progressive worldview has to already have been taken on board in order for those solutions to even seem like they follow from that premise, right? I ask out of genuine curiosity! It seems like in a lot of the time, on a lot of issues, people make arguments that would be compelling only to people who already share their worldview anyway. (A lot of theology comes to mind!) Michael Huemer is rare in this respect, as he usually employs a strategy of “start from premises that seem obviously true to pretty much everyone regardless of ideology and work from there.”
But anyway. Back to my question. Looking at the comment you just left, do you think anyone who doesn’t already share your worldview and underlying assumptions would have any reason, based on that comment, to feel like this should give them reason to reassess things?
Phil H
Jan 15 2020 at 8:49pm
Hi, Kevin. You’re right that the comment was tongue-in-cheek, and I didn’t offer any direct argumentation. The point of the comment was to show that progressive approaches, far from being some alien approach or a different worldview, clearly grow out of the same concerns that libertarians are wrestling with. Because if you start with the “facts” that Caplan starts with (many people do not do desirable thing X, many people do not seem totally incapable of doing desirable thing X), and think through the possibilities systematically, you will come up with two major types of solution: 1) change the people; 2) change the world (/society).
(The reason I can never get on board with conservatism is because it becomes so inconsistent at this point. Conservatives love to talk about human nature being immutable, until we start talking about poor people, at which point they suddenly take Caplan’s turn and start “blaming” the poor.)
(There’s another class of solutions: 3) change the desirability of X. Not discussed here.)
Caplan is interested in using better incentives to change the people, and I completely agree that this is a logical response to the problem he’s identified. Moreover, it often works! Setting the right incentives can be very powerful.
But the other class of solutions is clearly very important, too. If we accept Caplan’s premises that large numbers of people *can* do X but *aren’t* doing X, then it may be because something is preventing them. Now we have an engineering problem.
So the reason I made the comment was really to counter this “worldview” thing you’re suggesting. I don’t think progressives have a very different worldview. We’re looking at the same set of problems, and offering some different solutions.
Thaomas
Jan 16 2020 at 5:33am
Kaplan is clearly right, but not wholly relevant to policy making. Drivers CAN drive down twisting mountain roads without driving over the precipice. It would be silly to thank they CANT. But it still makes sense for a highway department that wants to reduce fatalities in a a cost effective manner to experiment with speed limits, rumble strips on the road edge, DWI laws, and barriers.
Don H
Jan 15 2020 at 10:23pm
Why do presume the test is unfair?
Phil H
Jan 15 2020 at 10:51pm
Life is unfair. The chances that this test has managed to magically become the one thing in the world that is absolutely fair are so remote that they are really not worth considering.
robc
Jan 16 2020 at 8:19am
It has been nearly 25 years since I finished school, and I am sure there were tests I whined about at the time, but I only remember one unfair exam (actually a set of them from the same professor in the same class). And it wasn’t math. I could imagine an unfair math exam (Statistics questions on a Calculus exam, for example), but I never had one.
The unfair exams were in an Intro to Psych class. Most of them exam consisted of 50 multiple choice questions that were obviously created by the book authors. If you read the textbook, you wouldn’t have any problems with these questions. Lazy way to create an exam, but not unfair. Then the professor added on 10 open ended questions based on her lectures. This was a fine approach, no problem with the concept. The unfairness was in the actual questions. They weren’t questions about the concepts she was teaching but insane, trivial details.
Phil H
Jan 16 2020 at 10:02am
I think the kind of fairness you’re talking about there is fairness between yourself and the gatekeeper: “fair” in your sense means, “has the gatekeeper (professor) set a path to exam success which a normal human being can negotiate as they learn the subject”. This kind of fairness is definitely important, but it isn’t really the one I had in mind. I was thinking about fairness between you and other students: does that student have ADHD; was that student favoured by the instructor in class; do some live further away from the school… and of course, the structural “-ism” disadvantages.
We’re never going to make things perfectly fair, but I think we can make a big difference, and the process of trying to make things fairer seems to me to be valuable in itself.
Philo
Jan 15 2020 at 10:52am
The country case (corruption) is not quite comparable to the individual case (math test). The individual is a genuine agent, who really can try to do many things (not everything: one cannot try to do what one knows to be impossible). But the country is not a genuine agent: it is a collection of agents. The notion of collective trying is metaphorical; and I, for one, need to have it explained how the metaphor is to be cashed out.
Obviously, if each person in the country tried to behave non-corruptly, all or almost all would succeed, and corruption would be eliminated. But, probably, when one speaks of “the country’s trying” one means something weaker than every person-in-the-country’s trying to do his part. What, then, does one mean?
nobody.really
Jan 15 2020 at 11:49am
Of course! Which explains why we so rarely see chemical additions, and why we so rarely see obesity. Oh, wait….
The remedy is obvious: All that is required is self-discipline. And, sure, there’s a raft of literature about how self-discipline is a finite resource that can be depleted, and thus people living under stress have a harder time exercising that discipline over time. Indeed, it doesn’t have to be MUCH stress. Simply ask a group of people to remember a string of numbers, and their ability to resist eating a doughnut declines.
I invite Caplan to read Thinking Fast and Slow, and then outgrow this curious need to ridicule people for their finite capacities of self-discipline. As a wise man said, just don’t do it — and it is done.
Christophe Biocca
Jan 15 2020 at 12:13pm
“Ego depletion” as it is known in the psychology literature, is under serious scrutiny as it has failed to replicate in large scale studies and may simply be the result of publication bias and p-hacking.
nobody.really
Jan 15 2020 at 1:37pm
Ha! Good catch. I (provisionally) owe Caplan an apology.
Let me modify my remarks to say that the question of whether self-control is a finite resource is an open question. In concur with Caplan about the wisdom of using caution in concluding can’t. I urge a similar caution in concluding can — even when the can merely requires the exercise of self-restraint.
KevinDC
Jan 15 2020 at 1:31pm
I’m not sure this works well as an “oh wait” moment. In order for that to work, we’d have to establish that Caplan’s model of “can’t vs won’t” predicts low levels of addiction or obesity or [fill in the blank]. But nothing in Caplan’s outline here predicts what levels we should expect to see of such things. Instead, it’s more a matter of how confidently we should interpret the levels we do see, and how confident we should be in saying that a person “can’t” lose weight. If Caplan believes that a large fraction of obese people “can” lose weight and if they don’t it means they “won’t” take the necessary steps, nothing about that belief leads to saying “and therefore we should expect very few people to be obese.” In general, you’re not going to convince people to change their mental models by pointing out things that are perfectly consistent with those models.
That said, I do agree with you that finite capacities for self discipline are a real issue. I left a comment saying as much on Bryan’s reply to Scott Alexander, but Eliezer Yudkowsky made the point better in his reply. (He often does that!) Bryan seems to genuinely think that if a person is capable of doing something in the moment, under extreme duress (like having a gun to your head!), they must therefore be capable of doing it all the time, forever. I think that is just obviously false.
Jens
Jan 16 2020 at 4:36am
Exactly. Saying “no” to a bottle, is different to saying “no” to all the bottles one might encounter on the way. There is nothing wrong with declaring a pile of sand “a pile of sand”. Even though 2 grains of sand are not a pile, and 3 grains are not a pile, and so on. “Thinking on the margin” may confuse once in a while. It can be useful to talk about piles.
Hazel Meade
Jan 16 2020 at 5:28pm
Maybe he’s not wrong to think that they are “capable” in some physical sense of doing it (not doing it) all the time. But their personal circumstances makes the option of not doing it (doing it) locally optimal most of the time. I.e. you know you need to lose weight, but you’ve got a report due in a couple hours and the hunger is preventing you from concentrating. It’s not totally wrong to say that is a “preference” but it also doesn’t take into account that all “preferences” are relative to a set of alternatives. Not all alternatives are equally accessible to everyone, either in terms of physical or mental resources. (i.e. when your mental resources are stressed, ignoring your hunger gets harder, raising the cost of resisting the donut).
Hazel Meade
Jan 16 2020 at 2:13pm
But what skill does it take not to consume a beverage? To abstain from sex? Just don’t to it – and it is done.
Something not obvious, clearly. Maybe there are some “skills” which we are not quite used to calling “skills” that have to do with emotional competence. For instance, perhaps the ability to resist temptation is a kind of skill that some people learn with practice. Maybe they’ve learned that thinking about the thing you are trying to resist is counter-productive, or that a good way to resist temptation is to distract yourself with some other activity, or that dropping your old friends and finding new ones can help getting out of drug/alcohol addiction.
Or, it could be something deeper relating to the ability to control one’s own thought processes and purposely ignore or focus on specific interests and desires. Some people may have an easier time ignoring cravings and focusing on something else, and other people may find the craving impossible to ignore to the point that the only way to stop thinking about it is to satisfy it. If you spend all day thinking about it to the point that you literally can’t focus on your job or get anything else done, you might just decide to satisfy the craving to get rid of it (at least temporarily). The more other stuff you have going on that you have to focus on (and can’t) the harder it’s going to be to resist the temptation to just temporarily abate the craving by giving in.
But if you’ve got a higher level of mental self-control you may be able to suppress the craving and focus on something else.
Hazel Meade
Jan 17 2020 at 12:49pm
Relevant article I just stumbled across:
https://www.theguardian.com/society/commentisfree/2020/jan/15/a-new-life-being-diagnosed-with-adhd-in-my-40s-has-given-me-something-quite-magical?CMP=share_btn_fb&fbclid=IwAR2vKbqCBsEKKFSGD0LeW_GUF6AmsXBNQXi5ArJzBZixpooOBhFjCZTCxMQ
Comments are closed.