Friday, November 20, 2009

More on Thought Experiments in the Classroom

This issue has been addressed before on ISW, but I would like to raise it again. I find it difficult to get students to see the relevance of thought experiments, even the less esoteric ones such as Singer's drowning child analogy in his argument for famine relief. Even if they come along for the first part of the ride, when I start adjusting the drowning child case to handle their objections to the analogy between it and the starving child, I start to lose them.

David Boonin and Graham Oddie's applied ethics anthology has a helpful discussion of the role and relevance of thought experiments (specifically, arguments from analogy) in the introduction. They discuss how to understand such arguments, how to criticize them, and the technique of appealing to variant cases. I have found summarizing and explaining their points to be somewhat helpful, but I'm interested in how others motivate and justify this form of argument in applied ethics to students who seem skeptical of the method.


  1. Hi Mike. I generally try to talk through the "so what" of the thought experiment and situate it within the theoretical issue. But I also encourage students to try to articulate problems they might have with the method, or specifically with nagging problems they have with specific thought experiments (say, Thomson's violinist).

    About adjustment: maybe the students are worried that there's a sort of ad hoc problem? (I can imagine that being a fair objection in some cases--e.g. how far can we get from reality before the particularly fantastic thought experiments seem lose their analogical force?)

  2. Judging from my own experience, I think another problem with adjustment is that they have the impression that you can arbitrarily make up anything to deal with anything and that in the end, when you've done so, you've just made up a story, not proven anything. I happen to think that they're often right in this regard: a good thought experiment in philosophy is one that can be cashed into a rigorous argument based on plausible premises, just as a good thought experiment in physics is one that can be cashed into testable mathematics, and there are plenty of thought experiments that turn out not to have 'cash value'. People can grasp the illustrative character of a thought experiment quite easily; skipping the 'cashing out' into straightforward argument and simply arguing through the thought experiments themselves is not something non-philosophers can ever be expected to be used to and will tend to regard as mere hocus-pocus. (Sometimes rightly.) So I think at the undergraduate level it is crucial to use thought experiments in such a way that they see that the thought experiments being used do actually have 'cash value' in terms of real arguments, by giving the corresponding arguments whenever possible. Which, admittedly, is often harder to do than it sounds.

  3. Hi Mike,
    From you post, I am not totally sure where students are falling off the boat, so to speak since I'm not sure how you are using the Pond / Fountain case.

    As I see it, the Fountain case is a case designed to elicit a particular moral judgment, viz. that anyone in that situation would be morally obligated to save the child. This is also the point of the Dora and Bob cases in "The Singer Solution to World Poverty," with the stakes / costs getting progressively higher.

    Things get interesting, however, when we ask people WHY saving the kid is morally obligatory. I ask my students this, give them a few minutes to write out answers on their own, and then we write their answers up on the board.

    I point out that some of these answers are better than others (worse ones: I'd feel guilty if I don't save the kid; I'll get a reward if I save the kid; better ones: stuff about the kid being more valuable than material possessions and comforts, Golden-rule type explanations, etc.).

    Then I mention Singer's moral principle (IF we can prevent something bad w/o sacrificing something of comparable moral significance, then we're obligated to do that) and note that it's similar to what many of them just said, in defense of *their own* explanation for why saving the kid in the foundation is required.

    Then it's just a short step to point out that this principle applies to the "real world" too, and seems to have interesting implications.

    Objections amount to pointing out disanalogies and other reasons to think some premise of the argument is false. (BTW I set the conclusion of the argument as that each of is obligated to donate $.33 a day). I try to show them that most of these objections or critical responses, when presented as valid arguments, nearly all have at least one false premise, which can be seen with counterexamples, often from the original Fountain case, as well as some more fun counterexamples.

    Some of my handouts and stuff I use on this issue are here:

    HEre are some thoughts about analogies from Richard Feldman:

    In his (correct!) view, the "point" of an analogy is to use a case, or cases, to extract and justify a moral general principle from. That's what I do, which works much better than asking whether case 1 is "like" case 2 or not, since everything is like and unlike everything else in some way(s)!

  4. First, thanks for In Socrates' Wake. I lurk-read most of your posts - great stuff!

    In introducing thought experiments to students I find it very useful to first spend some time talking about experiments in general. What are experiments? Why experiment? What types are there? And so on. Next, comparisons to "clean room" tests seem very useful in explaining the point of certain ethical thought experiments: we deploy them with the aim to track the pro tanto moral force of ONE moral factor. To do so effectively we construct a "clean" situation that block out other putative moral factors.


If you wish to use your name and don't have a blogger profile, please mark Name/URL in the list below. You can of course opt for Anonymous, but please keep in mind that multiple anonymous comments on a post are difficult to follow. Thanks!