How to talk about odd ideas
Say you get into a conversation with someone who insists that there is a giant teapot nestled somewhere on the other side of the galaxy. Or with someone who insists that there is an invisible, intangible goblin living in his garage. Or with someone who insists that the Illuminati are behind every major event in modern world history. What do you say?
Certain people seem to think* that you should solemnly demand “Show me the evidence,” and, if said evidence fails to appear, you should dismiss the claim and move on with your life. But this misunderstands the nature of our relationship with other rational persons. Certainly, if a random thought pops into my head and I decide to look into whether it’s true or not, I should ask myself, among other things, whether I can produce any empirical evidence for or against it. But someone else saying something is not analogous to a thought randomly popping into my head. People don’t say things without reasons. If someone makes an eccentric claim, and I have a desire to treat him as a neighbor and potential friend, rather than as a datum, I should ask him: “Why do you say that?”
Well, maybe he has empirical evidence. Maybe he read it somewhere. Etc. There’s no way to determine in advance which English sentences can function as valid reasons and which cannot. The test is whether or not I can accept the reason he gives for his belief as a reason for myself to believe as well. The goal of every conversation should be to find common ground, which of course requires that each of the participants precisely locate their own ground also.
We’re all pretty sure that there is no teapot, no goblin, and no Illuminati (well, maybe we’re not so sure about the last one). But if we can’t imagine a reason someone might give for their existence, the examples are useless and misleading. There is no point to deciding what to say in response to a person we expect to never exist. We should not feel as if we need a mantra (“Show me the evidence”) that will serve equally well as a response to every possible statement our interlocutor might make.
Luckily, we can, in fact, imagine what kinds of reasons a person might give for believing in the teapot, the goblin, the Illuminati.
The teapot-positor, that is, the pseudo-scientist, is the most difficult to sympathize with, for he’s the most obviously erroneous. Obviously wrong pseudo-science is indefensible (though non-obviously wrong pseudo-science can be very seductive). We can perhaps imagine him spouting some nonsense about sixth senses, and his ability to feel gravitational waves at immense distances, and how, while he understands that not everyone has this ability, he just happens to be able to sense large objects at great distances, and he senses the teapot’s presence. So then we ask, “Why do you believe yourself to have this ability?” He responds, “What kind of a question is that? Why do you believe that you can see?” You respond, if you want to use a fancy word, “Because my vision allows me to make empirical predictions. Can your sense do this?” At this point, he either acknowledges that it cannot, or he claims that it can, in which case we demand: show us what you used as evidence for this hypothesis. Whatever happens, we have now agreed to consider his statement to be Scientific, and to evaluate it according to scientific criterion.
The goblin-medium, that is, the occultist, may go this way, but he seems to me more likely to go a different way. After all, obviously wrong pseudo-science is indefensible, but obviously wrong occultism can still be tricky. We can perhaps imagine him saying, “Whaddaya mean? How can you tell that I’m a person and not just an automaton? Or I tell about you? I tell you, there’s a goblin in the garage, and it’s in pain! (angry with you!) (offended that you’re ignoring it!)” The correct response here is not “Show me the evidence.” That way madness lies.** The correct response is, “How can you tell that it’s in pain? (angry?) (offended?)” If he objects, “What kind of question is that? How can anyone tell what anyone is thinking and feeling?” we can say, “Well, you look annoyed, and you’re talking about certain things. Does the goblin use words, or otherwise express itself?” At this point, he either acknowledges that it does not, or he doesn’t. If he does, then we say: “how can a mind be both disembodied and in the garage?” and either he gives up, or we end up talking Metaphysics. The case in which he doesn’t acknowledge it is more difficult. Say he instead gestures to the arrangement of the furniture of the garage: “This is him being offended.” A breeze knocks over a rake that was leaning against the wall: “Now he’s really offended.” At this point we could observe that by “goblin” he seems to mean “mind embodied by the garage itself,” and he will likely agree. So now we demand: show us how such a form of life makes sense. If he can do so, we will agree to consider his statement to be Anthropological. If not, we will consider it to be Aesthetic.
The Illuminati-theorist, that is, the paranoiac, will perhaps adopt a line of reasoning that will expose him to one of the preceding demands. But we can also imagine him saying, “Of course they deny it, but I can tell when they’re lying.” At this point things get complicated. The topic of lying–and every paranoia involves the idea that most people are not just ignorant, but actively deceived–immediately makes the matter a moral one. We also know that, while there are probably no teapots nor goblins, people lie all the time. The Illuminati just might exist. And a conspiracy need produce no evidence; we can imagine one according to which, going back to ancient Egypt, the single most powerful person in the world has always informed his successor of a secret plan for how history will unfold, which he thenceforth secretly intended to bring about, without noticeably departing from how he would have acted were there no such conspiracy. So the best response may well be, “Who’s lying?” And if he responds, “Well, at the moment, all living Presidents,” we now demand: show us why we should trust you rather than them. Whatever he says, we have agreed to consider his statement to be Ethical, and to evaluate it as either good or bad.
When we make these demands, we cannot know before they answer that their answer will not satisfy us. Perhaps the teapot-believer really is a freak of nature, and possesses a sixth sense. Perhaps garages really do in some sense have emotions, intentions, thoughts, and everything else that goes with being a person, that is, having a face that others can read. Perhaps there really is just something a bit too honest about the President’s smile… and do you hear the faint quiver in his voice whenever he mentions the Pharaohs?…
*: The ideas for this post originated in an email exchange between me and my brother about the validity of Eliezer Yudkowsky’s epistemology.
**: This is essentially a Wittgensteinian/Cavellian point. A cartoonish argument for it would be: since we get at evidence through language, and we get at language through a community of language-users, we can’t demand evidence for someone being a language-user; for if we did determine who was a language-user through evidence, that would undermine the evidence itself. Our observations can give us reason to think that something is or is not a language-user, but when this happens, they’re not functioning within some quasi-formalized system for drawing deductions based on evidence.