Natural science, which deals with universal generalizations, must never be confused with natural history, which deals with particular events.

Think of how legend might tell the story centuries hence if in our time, as in Walter Miller’s novel A Canticle for Leibowitz, there had been a nuclear war:  “From a scientific point of view, of course, we know there never was a Flame Deluge.”

According to the “Strong AI” theory, if a digital computer were ever to imitate a human mind so well that we could not tell that we were speaking with a computer, then we ought to admit that the computer too has a mind.  To put it another way, if a computer could ever be programmed to produce the right outputs in response to the right inputs, then it should be regarded as understanding what it is doing in the very same sense that we understand what we are doing.

Philosopher of language John Searle has provided what I consider a wholly convincing refutation of this hand-waving view.  Searle speaks English and does not know Chinese.  Suppose, he says, that he is locked in a room with a set of rules, written in English, for correlating certain sets of Chinese symbols, which are passed into the room, with other sets of Chinese symbols, which he passes out.  And suppose that he masters the rules so well that when the people outside the room pass in Chinese questions, he returns the right Chinese answers.  The people outside the room may be convinced that he understands the conversation, but in fact he doesn’t.  According to Searle, this is all that happens when a computer executes a program.

Notice that this argument does not show that the human mind cannot be reduced to the operation of a machine; it shows only that it cannot be reduced to the operation of that kind of machine, a machine executing algorithms by digital processes.  Whether it can be reduced to the operation of some other kind of machine is an open question.

But now Searle engages in some hand-waving of his own, because he does not regard the question as open.  “We know,” he says, “that thinking is caused by neurobiological processes in the brain, and since the brain is a machine, there is no obstacle in principle to building a machine capable of thinking.”

Just how do we “know” this?  It is not an argument, but a declaration of faith.

No doubt the mechanism of natural selection can explain some things, such as why certain skin colors are more prevalent in climates in which they are adaptive.  But to suppose that natural selection can explain everything about human nature is absurd.  On this hypothesis, the only genes that are consistently passed on are the ones for traits that help us to pass on our genes.  Any genes which don’t should eventually disappear from the genome.  But is it really true that all human traits are adaptive in this sense?

It might be adaptive to enjoy exploring my environment, because I will be more likely to know where to find things like food.  But there is no adaptive value in seeking to know the meaning of life; how would it contribute to reproductive success to look for something which, on the natural selection hypothesis, isn’t even there to be found?  It might be adaptive to prefer rhythmic over arhythmic sounds, because rhythmic ones resemble Mama’s heartbeat.  But there is no adaptive value in being the sort of creature who is awed, humbled, and transported by the music of J.S. Bach.

I dare you to tell me the adaptive value of believing in God.  One eminent sociobiologist claims that we have genes for believing in God because belief in God unites the social group.  Hasn’t anyone ever told him that differing beliefs about God can tear apart the social group?  Besides -- why not just have genes for social unity?

Both legislators and judges should consider natural law and justice, but in different ways.  Legislators should consider them in order to frame the statutes; judges should consider them in order to understand the statutes.

According to some conservatives, this view is terribly wrong because it invites judicial activism.  They agree with the first part of the statement -- that legislators should consider natural law and justice in order to frame the statutes -- but they say judges should not consider them at all.  If judges are allowed to do so, they fear, then judges will inevitably substitute their own opinions about right and wrong for what the statutes mean.  This is much like the Reformation’s idea about how to prevent distortions of sacred text, but instead of sola scriptura, the rule is sola statutum.

Perhaps solum statutum would work if it were possible for judges not to consider anything outside of the statutes.  But it isn’t, because the statutes are not self-interpreting.  No texts are.  No matter how earnestly the interpreter defers to the meaning of their words, this meaning depends on innumerable things beyond and outside of them – yes, on things like justice.  If you get those other things wrong, you will not grasp the law either.

If I am right – if not to consider those other things is literally impossible – then if you push them out the front door, they will sneak in through the back.  But in this case, the demand not to consider them does not reduce the danger of judicial activism:  It increases it, because it is much more difficult to scrutinize considerations we pretend not to be thinking of than considerations we admit we are thinking of.

Some judges say that they are forbidden to look beyond the text of the statutes by their very oath of office.  It would be a pretty paradox if the oath did require impossibilities, but fortunately, it doesn’t.  Look it up:  They promise to administer justice and to uphold the laws.  Both duties are binding; neither may be violated; neither may be subordinated to the other.  The statutes do not take the place of justice; justice illuminates the meaning of the statutes themselves.

A highly placed federal judge once suggested to me that if construing the meaning of the statutes does require understanding the meaning of justice, then the legislature has done a bad job.  It should use clearer words which don’t carry such a moral charge.  He had a point, for one can certainly replace words which are more likely to be abused with words which are less.

But he also overstated his point, for one cannot replace words which need interpretation with words which interpret themselves; there are no such things.  Ultimately, all language points beyond itself to the real world.   If you cannot fathom the world, then you cannot fathom the language, and the just and unjust are features of the world.

It always surprises me when social scientific colleagues express the conviction that in principle, human behavior is predictable.  They seem to make two assumptions:  First, that all human behavior is causally determined, second, that in principle, whatever is causally determined can be predicted.

But even leaving aside the question of whether all human behavior is determined, determinism does not entail predictability.  We can form reasonable expectations about many things in the future, but it would be impossible to predict human behavior with certainty even if determinism were true.  The main reason is that the very act of prediction changes the system which the scientist is trying to predict.  He may think he can predict the effects of his predictions, but this is like the dog’s belief that if only he chases fast enough, he will eventually catch up with his tail.

For example, suppose the social scientist predicts that certain people will do P.  What is to prevent them from doing the opposite of whatever he predicts, just because they are bent on refuting him?  In this case, even though their doing not-P was causally determined by his prediction that they would do P, he was unable to predict it.  Suppose he predicts that on the following day, as he is going about his ordinary activities, his research assistant will shoot him.  Isn’t it likely that he will alter his ordinary activities to keep from being shot?

If he takes his response to his prediction into account, then he must make a different prediction, and then, of course, he must have to take into account his response to that prediction.  Like the dog, he can never catch up.