A recent post by Brian Weatherson over at the Arche Methodology Project Weblog raises an interesting idea: can philosophical thought experiments be treated as a genre like, say, science fiction? This idea is also explored in Jonathan Weinberg’s article “Configuring the Cognitive Imagination” in New Waves in Aesthetics. Weinberg spells out the idea, without endorsing it, on page 214:
Yet, what if philosophical thought experiments were a genre—at least in the sense that engaging in them successfully requires mastery along the same lines as I have sketched for the mastery of literary genres? There are rules to engaging properly with a hypothetical scenario, after all. To make just some of the more obvious generalizations about our imaginative practices with thought experiments: one should embellish as little as possible; generally it is a practice conducted in an affectively `cool’ manner; and our inferential systems must often be brought to bear in this particular sort of imaginative project as well. And there are surely other, and more subtly articulable, rules for the proper performance of thought experiments still to be detailed.
While I was initially attracted to the idea—especially given my interest in imagination, fiction, and genre—I now think that it won’t do, on the more interesting interpretation. Roughly, the idea is to treat philosophical thought experiment as a genre relevantly similar to other genres of fiction. I have two worries with the idea, interpreted thus. In this post, I’ll press the first worry: we engage with philosophical thought experiments relevantly differently from the way we engage with fictions in other genres.
There are two ways to interpret the idea of treating philosophical thought experiment as a genre. The less interesting interpretation is just noting that there are some descriptive norms shared by the various philosophical thought experiments. If one has a decently permissive notion of what constitutes a genre, it would be hard to see the interest of this interpretation. Look! Austen novels constitute a genre, novels with Mr. Darcy being a main character constitute a genre, etc. Genres are easy and cheap to get.
The more interesting interpretation is that, not only are there descriptive norms shared by the various philosophical thought experiments, but that we engage with thought experiments in relevantly similar ways as we engage with fictions in other genres, such as science fictions. In other words, with respect to how we engage with them, the idea is to treat philosophical thought experiment as a genre relevantly similar to other genres of fiction. This more interesting interpretation is the one suggested by Weinberg and, I believe, intended by Weatherson.
To see why it is a further claim, of the more interesting interpretation, that we engage with philosophical thought experiments in relevantly similar ways as we engage with fictions in other genres, consider the case of the encyclopedia genre. I take it that most people would grant encyclopedia status of a genre: it is a grouping of works with well-defined conventions and is well-accepted in our society. However, we clearly do not engage with encyclopedias in relevantly similar ways as we engage with fictions in other genres. For example, whereas we tend to believe what is said and implied in The Stanford Encyclopedia of Philosophy, we tend to imagine what is said and implied in The Hitchhiker’s Guide to the Galaxy. Even if we do derive some knowledge from engaging with The Hitchhiker’s Guide to the Galaxy, we clearly do not do so in relevantly similar ways as we derive knowledge from The Stanford Encyclopedia of Philosophy. Hence, different genres could have importantly different modes of engagement. (This is a point that I think Weinberg would agree with.) Combined with the fact that genres are easy and cheap to get, the more interesting interpretation must include a further claim that we engage with them in relevantly similar ways as we engage with fictions in other genres. And that is the substantial claim that I will challenge.
To reiterate, the issue is with modes of engagement, not with fiction versus non-fiction, in the sense of stories’ correspondences with reality. There are non-fictions, such as memoirs, that we engage in relevantly similar ways as we engage with other fictions. However, philosophical thought experiments appear to be different. With philosophical thought experiments, we typically are primarily in the business of judging—judging whether the scenario presented is indeed possible, or judging whether the relevant modal claims are actually true. We do imagine, but only in the service of judging. On the other hand, when reading The Hitchhiker’s Guide to the Galaxy, we typically imagine the scenario therein without even thinking about judging whether they are possible. (We might incidentally make other judgments, however, such as how gripping the plot is or how complex a character is.) Although we might make an analogous judgment about the story’s plausibility, or whether it conforms to the genre’s conventions, we tend to do so only when something has gone wrong, or only when we judge the story to be implausible. Such judgments certainly are not typically required in our engagements with fictions. That is a relevant difference: philosophical thought experiments require us to judge and require us to draw real-world conclusions, but fictions do not. In fact, that is what makes thought experiments useful in philosophical inquiry.
Let me try to put the point differently. Different genres not only have different rules of import, or principles of generation, but they also have different rules of export. Philosophical thought experiments require exports that are relevantly different from what fictions of other genres typically require. The requirement to export information about possibility makes engaging with philosophical thought experiments relevantly unlike engaging with fictions in other genres. In this respect, I am inclined to agree with Williamson that philosophical thought experiments are more like, or are species of, cases of counterfactual reasoning, which also typically require judgments and drawing real-world conclusions.
The foregoing discussion explains my first worry against the idea of treating philosophical thought experiment as a genre: we engage with philosophical thought experiments relevantly differently from the way we engage with fictions in other genres. Even if this first worry can be assuaged, there remains a more fundamental worry. My second worry is that treating philosophical thought experiment as a genre does not obviously help in vindicating a non-skeptical epistemology of intuitions. Spelling out that worry and making some positive suggestions, however, will have to wait.