The importance of thought experiments

Thought experiments are crucial to the field of philosophy. They seek to reveal the principle(s) underlying the reasoning for a position so that such a principle(s) can either be applied ubiquitously in one’s life or thrown out all together. Or, if the thought experiment is really good and/or really precise, so that such principle(s) can be augmented for a given context. This process is so important I have a hard time imagining too many philosophers disagreeing with the usefulness of thought experiments (though many will reject the validity of various ones on varying grounds).

The reason I bring this up is because of Michael Hartwell’s post about utilitarianism from a couple of months ago. I have already responded to the bulk of what he had to say, so that can stand for itself where it is. However, I only briefly touched on one aspect of what he said and I want to address that now. Here is the relevant portion:

Of course, it’s never that simple in real life. These fables (thought experiments) assume godlike knowledge of the situation. What if the cave was only going to flood knee-deep levels and there were small holes to breath from? What if the five people on the train tracks weren’t oblivious to the train or were planting a bomb?

They also assume a dichotomy of actions. Do nothing, or kill. There’s no option to swim out of the cave, wait for rescuers or warn the people on the tracks.

There are two major issues with this. First, Michael is attempting to apply the idea of thought experiments specifically to utilitarianism. I have little doubt that he knows that there are plenty of other thought experiments which are used for other ethical theories, but none-the-less, he is applying certain ones solely to utilitarianism. That is, in his post he references the Trolley Problem as if it is a utilitarianism center-piece, a bit of logical exploration which is unique or primary to that ethical framework. He is wholly wrong to do this.

The Trolley Example is used by a number of ethical theories in order to arrive at particular moral answers. Libertarian Judith Jarvis Thomson famously extended the problem and concluded, as she often does, that there is a right to not be unjustly harmed. That had little to nothing to do with utilitarianism. There are dozens of other uber-famous thought experiments people of all ethical persuasions use. People may design their scenarios with a particular framework in mind, but nothing stops any other philosopher from applying entirely different ideas to them.

Second, the whole point of a thought experiment is to present a scenario with controlled parameters. The goal is to unveil a principle behind the reasoning for a position. (This is especially important when a given position is intuitive but has no good underlying principle from the viewpoint of the thinker.) It may be of interest to ask something like, Is it moral to take a risk when the negative consequence is significant? Does the positive consequence need to be equal? Bigger? But that is still searching for underlying principles – and it is still doing so with controlled parameters. That is, even if there is a factor of randomness thrown into a scenario in an effort to better mimic real life – “There is a 5% chance everything will be fine if you do X instead of Y” – it is still controlled.

A good thought experiment gives enough information to illicit a certain type of response. In the traditional trolley example, it’s you, a lever, and a few people scattered across a couple of tracks whilst unable to communicate with you and certain to die or live given your choices. That gets at particular principles of right and wrong. If we were to change the experiment to say that there was a small chance that the people on one track turn around in time and survive whereas there was no chance the person on the other track would turn around in time, we would shift the focus from principles to risk/reward analysis, getting into a much more subjective area of human psychology. That could help us in real life, but only insofar as we find ourselves in similar enough situations – which is unlikely. That’s fine if that’s the type of response one wants, but it doesn’t do much to illustrate principles – at least not in the way the original Trolley Problem does.

I’ve written a few times about thought experiments in philosophy. I’ve never been that extensive on their importance because I just sort of assumed people recognized how useful they are. But I guess I know what happens when I assume: