College Week at Slate
by Jon Shea
15 November 2005
K. Anthony Appiah is my new best friend:
I start with two problems. One is most evident with humanities majors: Many of them don’t know how to evaluate mathematical models or statistical arguments. And I think that makes you incompetent to participate in many discussions of public policy.
Nov 15, 04:04 PM
Excellent! I like especially that he followed up the quotation you cite here with the following:
“I favor making sure that someone teaches a bunch of really exciting courses, aimed at non-majors in the natural and social sciences, which display how mathematical modeling and statistical techniques can be used and abused in science and in discussions of public policy.”
So, he’s not just whinging about a lack of math knowledge—he’s got an idea for how to fix it. Granted, his idea is desperately in need of an implementation, which is harder than most people realize, but at least it’s not just a mindless rant.
(And I know from mindless rants :-)
As an aside: Jon, I though you said this thing used Textile. How come block-level Textile markups like bq. don’t work?
Nov 15, 04:10 PM
I’m not a huge fan of his idea, which I suspect a few Dartmouth departments attempt. It isn’t enough to shunt the math off to a course or two. You should have to apply models and statistical reasoning in pretty much every clase in the social sciences. Math is what separates science from bullshit. Necessary, not sufficient requirement.
I’m not sure broke bq., especially because basic Textile still works in comments. It was probably something I did when I tried to adjust the CSS so that the comment font wasn’t so small.
Nov 15, 04:30 PM
Math is an excellent tool, and I agree with you that practically everybody should learn more than we typically do.
But I disagree that math separates science from bullshit. In fact, the worst bullshit I’ve seen in science is cloaked in mathematics. What scientists and engineers always seem to forget is that most of their mathematical models are incomplete at best, and misleading at worst. Scientists often use math as a tool, but math isn’t what makes their work science.
Certain parts of many real-world problems can be partially analyzed with mathematical tools—but only partially. Probability and statistics are a great way to capture certain aspects of social and economic behaviour; just like partial differential equations can tell you part of the truth about the strength of a bridge or the flow of fluids around a sailboat’s hull. But math is just an idealization, and it necessarily discards many salient features of any complex problem.
So I don’t think mathematics is a necessary or sufficient criterion for good science. The models are nice and clean and seductive, and more or less always wrong. That doesn’t mean we shouldn’t study them, or that math is crap just because it only solves part of the problem. Mathematics is a beautiful and subtle instrument, practically the only coherent system of rational thought we have. But a theorem of mathematics is not necessarily a theorem about real problems in the real world.
Nevertheless, that’s no excuse not to study math. I’m just ranting about the obscene worshipfulness of some scientists about the supremacy of mathematics. I think it’s a dodge.
Nov 19, 01:50 AM
I agree with Michael. Everything we learn in my Epidemiology course can be quite useful in clinical practice… but on the other hand, for every equation and rule there’s about a million exceptions and confounders left out. Practical experience definitely has a place too. For medicine anyway, if we just went with math alone, that’s when it would be bullshit.