What Is It to Be Rational?

Like any word worth its salt, “rational” has many commonly accepted meanings or senses, or has a compound (or polysemous) meaning. Here is an attempted inventory of its meanings that I myself accept:

The logical sense: To be rational is to think logically, to reason validly. This means, in formal terms, that one is able to, or actually does in a particular case, or tends to in general, draw conclusions that conform to the recognized rules of logic, which is to say, conclusions whose truth, regardless of the subject matter, is guaranteed by the truth of one’s premises. Thus, it is rational to believe that whales are mammals if one’s cited evidence or reason is that whales give birth to live young and only mammals give birth to live young. The logical form of this  argument or inference is:

p (Whales give birth to live young.)

If p then q (If something gives birth to live young, then it is a mammal.)

Therefore q (Whales are mammals.)

            Of special note is that rationality does not thereby guarantee truth. Rationality pertains only to the hypothetical: If the premises were true, then the conclusion would also be true. Thus in the example I gave, one of the premises happens to be false, since there are animal groups besides mammals who give birth to live young. But on this conception of rationality, a person would still be perfectly rational if, believing falsely that only mammals give birth to live young, they concluded that whales are mammals (which, by the way, is true).

The statistical sense: To be rational is to base one’s beliefs on what is probably the case. This is an important sense of rational since many or maybe most or even all conclusions regarding empirical matters are only probable, not certain. Is this a dagger which I see before me? Maybe, maybe not. The rational person will decide on the basis of what is the more probable, given the circumstances, etc. Thus, if daggers are not the sort of thing that suddenly materialize out of and then hover in the air, then it is probably not a dagger one sees before one – at least, not a real as opposed to an hallucinatory one. Macbeth was bonkers precisely because he could not or did not reason in this way.

            A few months ago a little black spider suddenly appeared (or I suddenly noticed it) in the upper right of my visual field. I might at first have started and tried to move away from it. (I don’t recall.) But from previous experience I quickly inferred that it was (much more probably) a floater in my eye. I could easily imagine, though, that uneducated people would not really know what to think, and might even rationally conclude that they were being followed around by a little black spider ghost (that no one else could see or feel). The probability that grounds this sort of rationality is, therefore, and like the validity that grounds logical rationality, hypothetical: It depends on one’s premises. If one’s existing beliefs included ghosts and one were simply ignorant about eye anatomy and optics, then the ghostly nature of the spider could indeed be probable.

Note then that I am rejecting the idea that probability is objective … or at least that my conception of probability is objective, since I certainly acknowledge that there is also an objective conception. To me it seems silly to suppose that there “is” such-and-such % chance that the spider is a floater: Either it is a floater (or a ghost) or it isn’t. What is being labeled “probable” is only a hypothetical assessment based on my beliefs, which are subjective in the sense of being of unknown truth to me.

The integrated sense: To be rational is to act or even feel consistently with one’s beliefs. Thus, if you are asked whether you believe that whales are mammals, then, other things equal (for example, you don’t believe you will be shot if you say anything), if you believe they are, then, if you are rational, you will say “Yes.” Similarly it would, other things equal, be rational to feel relieved on the basis of your belief if you learned that new maritime laws forbid the harvesting of mammals and you loved whales especially of all creatures in the sea.

The practical sense: To be rational is to want the means to conduce to the ends. Thus, if I desired x and believed that the only way to obtain x was to do y, then, other things equal, I would, if I were rational, want to do y and, other things equal (e.g., an earthquake does not intervene), do y. Thus if someone had a toothache and wanted very much to get rid of it, and believed that only going to the dentist would resolve the matter, and any aversion they had to seeing the dentist was far less than their aversion to continuing to suffer toothache, and yet they were thoroughly unmotivated to go to the dentist and hence didn’t, then surely they were being irrational. Indeed, it is hard even to imagine how there could be such a case.

The substantive sense: To be rational is to possess beliefs that are widely accepted. Thus, a person who believes that everyone (themselves included or not included) is a Martian spy robot is irrational … not in the first instance because of a faulty inference but simply in virtue of the (nonconformity of the) belief itself. Indeed, such a belief could be based on a valid or even a probable inference, but then it could only be because at least one of the premises was itself nutsy.

            I am not so sure that I want to say the same thing about desires or behaviors. That is, if a desire or an action were intrinsic and not based on an inference from a nutsy belief, I would be hesitant to label it irrational just because it was not widely accepted. So if someone possessed a strong yearning to eat thumbtacks, despite knowing the consequences of doing so, and for no further purpose or reason, then I am not convinced that labeling the desire irrational would be anything other than expressing one’s strong disapproval of it or strong sense of strangeness, or just one’s aversion – as “brute” as the thumbtack eater’s desire.

            Of course if the thumbtack eater themself considered the consequences of doing so less desirable than of not eating them, they would be irrational (in the practical sense). But in the case where they would indeed derive more subjective satisfaction from eating thumbtacks than from abstaining and avoiding the noxious consequences of doing so, then I think “irrational” is just not the right label (although I would be as tempted as anyone to call them “nuts”).

            I admit also to some hesitancy about this entire category of (ir)rationality. After all, I (and probably you) consider many beliefs that are widely accepted to be not only false but irrational, and other beliefs that are almost universally rejected to be in fact true and even rationally held. For instance, I think the religious beliefs of most people are stark raving crazy. And most people would probably find my (dis)belief in morality (objective value) and at least skepticism about the self and so on to be equally insane.

            So it could be that in the end rationality (at least in my estimation) is always about a relation of some kind, whether it be between premise(s) and conclusion, means and end, or belief and behavior (or feeling). Nothing is in itself rational or irrational. The rational is relational.