For this reason, Rand's assertion that "Reason is man's only means of grasping reality and acquiring knowledge" is empirically empty. Since Rand provides us no way of distinguishing between what she regarded as valid and non-valid reasoning, there is no way to evaluate the extent to which her conception of reason clashes with the "natural reasoning" discovered by cognitive science.
By studying how people actually reason to solve problems of everyday life, cognitive science has discovered that deductive logic has little to do with the inferences people actually make:
[F]ormal logic is not a good description of how our minds usually work. Logic tells us how we should reason when we are trying to reason logically, but it does not tell us how to think about reality as we encounter it most of the time....
Logic enables us to judge the validity of our own deductive reasoning, but much of the time we need to reason non-deductively — either inductively, or in terms of likelihoods, or of causes and effects, none of which fits within the rules of formal logic. The archetype of everyday realistic reasoning might be something like this: This object (or situation) reminds me a lot of another that I experienced before, so probably I can expect much to be true of this one that was true of that one. Such reasoning is natural and utilitarian — but logically invalid....
Our natural reasoning is thus a kind of puttering around — the intellectual equivalent of what a child does when it messes around with some new toy or unfamiliar object. After a spell of mental messing around, we may put our reasoning into logical terms, but the explanation comes after the fact, the actual process took place according to some other method... [The Universe Within, p. 132-136]In my JARS article, I quoted psychologist Paul E. Johnson's description of how experts in medicine, engineering, and other skilled professions think when solving professional problems. Here's the entire quotation:
I’m continually struck by the fact that the experts in our studies very seldom engage in formal logical thinking. Most of what they do is plausible-inferential thinking based on the recognition of similarities. That kind of thinking calls for a great deal of experience, as we say, a large data base. If anybody’s going to be logical in a task, it’s the neophyte, who’s desperate for some way to generate answers, but the expert finds logical thinking a pain in the neck and far too slow. So the medical specialist, for instance, doesn’t do hypothetical-deductive, step-by-step diagnosis, the way he was taught in medical school. Instead, by means of his wealth of experience he recognizes some symptom or syndrome, he quickly gets an idea, he suspects a possibility, and he starts right in looking for data that will confirm or disconfirm his guess.I further suggested in the JARS article that the way in which professionals solve problems does not accord with Rand's notion of reason. Seddon responded to this by claiming that I equate Rand's view of logic with deductive logic. But if he had read more attentively the passage above, he'd see that his claim misses the point. When medical specialists "quickly get an idea" based on their wealth of experience, they are reasoning deductively, as follows:
Disease X is known to produce various symptoms, including A, B, and C.
This patient has symptoms A, B, and C.
Therefore, (I suspect) this patient has disease X.
This is clearly not an inductive inference. It's a deductive inference, but an invalid one: it commits the fallacy of the undistributed middle.
Now whether Rand's "reason" includes invalid deductions or not, I will leave for Rand's apologists to decide. The only point I wish to make is that human intelligence apparently cannot do without making use of invalid inferences. Indeed, the whole question of "validity," in everyday life, is of no very great concern. What is important is: (1) the experiential data base from which the invalid inferences are made; (2) the extent to which our conclusions are subjected to criticism (particularly empirical criticism, when that is possible).
This is not to say that we must always make use of invalid inferences, coupled with criticism of conclusions. There are fields of inquiry, particularly in the sciences, where strict deductive logic remains an important tool of analysis. But this does not apply across the board. Many of the problems of everyday life cannot be solved, in part or whole, by only using logically valid forms of reasoning. As John R. Anderson, a "leading theoretician of thinking" (according to Morton Hunt), puts it: "Naturalistic reasoning doesn't use the rules of general inference, but tends to be 'content-specific' — it uses rules that work in a particular area of experience." Morton Hunt sums up the position as follows:
[W]e are pragmatists by nature; what feels right we take to be right; were this not so, we would long ago have disappeared from the earth. Our pragmatism, our natural mode of reasoning, is not anti-intellectual but is the kind of effective intellectuality that was forged in the evolutionary furnace. (The Universe Within, 138)