Do the inferences in this chapter occur anywhere outside the psychological laboratory? In fact, they do crop up from time to time. I was sitting at the exit of a chain store, and a young man was sitting next to me filling in a form, which I knew was to apply for a job in the store. (I’d seen it lying on the next chair when I had first sat down.) At one point, the young man got stuck, and turned to me and asked: “what’s an optimist?” I said: “Someone who always expects good things to happen.” I almost added my favorite definition of an optimist, “An optimist is anyone who believes that optimists exist,” but his accent betrayed that he was not a native speaker of English. My unspoken coda, however, has an interesting property.
Is it true that optimists believe that optimists exist? Everyone to whom I have put this question, thinks for a moment, and then assents to the proposition. As one person said, “If you didn’t think there were any optimists, you wouldn’t be very optimistic.” I asked a class of some fifty students whether they accepted my definition. They all did. Yet, it can’t be true, and our inability to grasp its defect illustrates the problem we have in making inferences that depend on recursion—the repeated use of a general premise. Ask yourself: do you think that optimists exist? Of course, you do. I believe that there are optimists, and so do you. So, it follows from my definition that we are both optimists too. If others realize that we’re optimists (because we believe optimists exist), they are in the same position as us: they think that optimists exist—namely, us—and so they are also optimists. The definition is a mental infection. It spreads optimism among the population like a happy plague. You pass on the definition, and others succumb at once to it and become optimists. This consequence of the definition is not obvious. Unlike superhuman reasoners, we do not grasp at once the consequences of repeated uses of general propositions. We prefer to think about particular individuals, and then perhaps to use general propositions to update our models. This bias is in keeping with the fundamental principle of the model theory: models represent individuals.
Why is the definition of optimists false? Optimists do think that optimists exist, but so do many pessimists. Hence, the definition doesn’t apply only to
optimists. The proper definition of an optimist is: a person who expects good things to happen. And not everyone is an optimist.
What falsity? The definition includes 'anyone who believes that optimists exist', which includes pessimists.