John Searle Book

Confession of Stupidity:

Lately, I've been had long and agonizing conversation with my friend about the categorical imperative. I was insisting that it didn't make sense; my friend insisted that it did, and that I merely misunderstood it. After much deliberation, I found to my embarrassment that he was right: I had misunderstood it. I had misunderstood it badly. Now, fortunately, I think I've got a hold on the concept, which indeed is not terribly complex (though, for my brain at least, a bit too much).

Having thought a lot about it, I wish to give a fairly pedantic examination of the theory (forgive me!). But first, I'd like to explain what it is not, and the various ways that I managed to misapprehend it.


My Mistakes:

I was under the impression that the categorical imperative was this: "Before you do an action, consider whether it would work as a universal law; if it would, it's okay; if it wouldn't, then it's forbidden." But I thought to myself "I could will almost literally anything as a universal law. I could will universal suicide or a universal fight to the death, just so long as I was willing to commit suicide or fight to the death myself." The thought experiment Kant instructed us to perform seemed completely arbitrary; he might as well say "before you do an action, imagine if it could be performed on a spaceship."

Also, I thought "if I give enough qualifications, almost anything could work as a universal without anything catastrophic happening." For example, I could say "if you are tired, going to work on a Tuesday morning, hate your job, are six foot three inches tall, and need to urinate, it's okay to be push people on the street." The conditions given for this action are so specific that nothing would really change. Similarly, I could say "if you're really really desperate, it's okay to steal," and it could work. In fact, I bet that's already the case.

The second formulation also confused me: "treat people as ends, not means." For one, I couldn't see any connection between this formulation and the previous one: what does treating people with respect have to do with willing universal laws? What's more, the command seemed preposterous. I thought, "but I treat people as means all the time. When I order coffee I don't do it for the sake of the person selling the coffee." I also thought that there was a contradiction between doing an action for the sake of duty and doing it for the sake of another person; what's the real 'end', the person or the duty?

All of my objections managed to completely and totally miss the point. My friend got frustrated because I was bringing up all these irrelevant objections, and I felt very confused.

Hope came when I took a long walk, and decided that I would attempt to start from Kant's assumptions (which I knew roughly from his Critiques) and see if I could get to something that resembled the categorical imperative. Here is what I found.


My Attempt to Derive the Categorical Imperative:

When we look at nature, we often find determinism. Equations determine the movement of particles and the temperatures of stars; chemical structures determine the qualities of materials; instincts honed by natural selection determine animal behavior. Sometimes, we also see random chance. We run into an old friend in a distant country, or we accidentally drop our mug of beer. But freedom is incompatible with either determinism and chance: to be free, we cannot be said to be determined by anything else, nor can we attribute our actions to some random process. Nonetheless, we cannot help but suppose ourselves free; otherwise, we can never decide what to do'since all decision-making presupposes freedom.

We can relieve this tension in one of two ways. One way would be to declare freedom illusory. We presuppose freedom when we decide, but this is just a feeling of freedom; we are just as determined by natural laws as anything else in nature, and just as subject to random processes. And here we might ask ourselves, what is freedom, anyway? Well, maybe it's easier to answer: when are we not free?

When we are compelled to follow a law or directive foisted on us by somebody in power, we aren't free because we aren't determining our own actions. But, when a drug addict sells their property to get a fix, we also say they aren't free, even though they aren't following some external directive, because their desires are determining their actions. Last, we don't hold accountable a person whose house was destroyed by a hurricane, and is reduced to penury, because the hurricane might be said to have struck by chance.

So we say a person is free when they make coolly rational decisions, not forced by some outside party, not overwhelmed by some strong desire, and not affected by some random process. But is this justifiable? Is this really freedom? And do we have it? It seems that, even when we're making coolly rational decisions, we're still subject to the laws of nature, to random events, and are still guided by our wants and needs. So is freedom'at least in the fundamental sense of an action being undetermined by all previous events, nor at all random'is this freedom possible?

Kant thinks it is; but he has a job to do in proving that it is possible. We can attempt to resolve these conflicts by hypothesizing that there is a part of us that is neither determined nor subject to chance.

But what would this part of us be? I can find two possibilities, not mutually exclusive: consciousness and rationality. Humans are distinguished from other creatures by our self-consciousness and by our ability to reason. First, let us suppose it is consciousness only that makes us free. But what are we conscious of? Hunger, thirst, exhaustion, desire, and various other things in our surroundings.

If something external to our bodies forces us to do something, we are obviously not free, just as a dog is not free when being trained by its master. Consciousness seems to make no difference in that case. But we also seem not to be free when following some desire. For example, a dog is probably conscious of hunger, too, yet we do not usually think that dogs have free will when they pursue food. Perhaps you can say you are free because you can chose which desire to satisfy; but then what is the criterion by which one makes such a decision? Another desire?

Clearly, something extra is needed: rationality. Our ability to use reason is what sets our decision-making apart from that of dogs and cats. Using reason, we can establish criterion that are not themselves desires. We can reign in desire for fast food if we realize that it will have negative long-term effects; we can abstain from buying that expensive new luxury car by considering how it would affect our children's futures. Ah, but that's not quite enough! Because, even when we refuse to eat fast food, all we're doing is balancing our desire for something salty against our desire for long life. In a sense, we're still in the position of a drug addict balancing his desire for a fix against his desire for a coat. So not only must reason be the criterion, but reason must be the motivation, for free decisions. We must both be determining our own actions and not pursuing some desire.

Now we are in a position to ask ourselves: what is morality? To be moral is to decide to do the right thing; it requires decision-making, and therefore can only apply to rational creatures. Not only can morality only apply to rational creatures, but morality can only apply to creatures insofar as they are rational.

Anything non-rational, therefore, cannot be moral. Animals and inanimate objects cannot reason, so morality cannot apply to them. We have previously determined that things like hunger, thirst, and other desires are non-rational; so such things are not the basis of morality. Neither is morality concerned with achieving any particular goal in the world, because all goals derive their value from desiring them. Phrased in a slightly different way, all goals are contingent: they are only operative when the desire for them is operative; and we know that our desires are ever-changing. Nor can morality even have anything to do with human nature, since all other rational creatures'human, alien, or angel'would be equally subject to it.

So morality, being derived from rationality and only applicable to creatures insofar as they are rational, must not have anything to do with empirical reality; it is, in other words, a priori.

Now, morality deals in oughts, commands, or imperatives'what we should do. Since morality cannot take into account states of fact, the commands of morality must apply under all conceivable conditions. Also, since every rational creature is equally subject to the commands of morality, all moral imperatives must apply equally to all rational creatures. In short, morality is equally operative no matter who you are or what you're doing. It is not dependent on any circumstances: it is a categorical imperative.

From this alone we can draw the conclusion that any action which makes an exception of the actor cannot be moral. In other words, any action which could not be universalized is immoral (since the categorical imperative applies to everyone equally at all times). Also, since morality applies to all rational agents equally, any actions which treat a rational agent as not deserving of equal respect is immoral. This is to say, any action which treats a rational agent as a non-rational part of nature is forbidden; there is no valid reason for doing so.

This test is a negative test. The categorical imperative cannot tell you what to do; it can only tell you what you may not do. You may not make an exception of yourself; you may not treat another rational agent as a part of nature. In other words, act only on maxims that can be willed as universals; never treat other rational agents as means only, but as ends in themselves deserving of respect.


The Categorical Imperative in a Nutshell:

So Kant does a very clever thing here. Kant essentially makes morality and freedom synonymous. You are only free if you are motivated by reason; and when you are motivated by reason, you are abiding by the categorical imperative, and are thus moral.

Rationality is, for Kant, the basis of free will. So when rationality fully determines the will, it is the will giving a law unto itself. This removes the paradox of freedom. We are not free when we are following a law from outside ourselves, nor when we are following our own desires; we are only free when we are following the laws we created for ourselves (you can see the Rousseau influence here). And not only must we abide by these self-made laws, but we must abide them purely for the sake of abiding by them, because only then are we free and moral.


Some Implications:

Before examining whether Kant's premise holds, let us take a moment to ponder out some of the implications of his conclusion. In Kant's system, many things commonly regarded as immoral are forbidden: lying, stealing, raping, murdering. Stealing, for example, treats people as ends and not means; to steal makes an exception of yourself from a general rule; it cannot be willed as universal. This consonance with popular opinion is (at first sight, at least) an encouraging sign.

But consider further. Because Kant has divorced morality from all consequences, and founded it purely on consistency, all moral actions are equally moral, and all immoral actions are equally immoral. This is apparent at once, when one considers that one can either be consistent or inconsistent, not half consistent; one can either treat someone as an end or not, not half as an end. Therefore, lying and murder are equally immoral and equally forbidden. The white lie you told your wife puts you on a level with the murderer in prison. This is a chilling conclusion, as any punitive system which doles out punishments in proportion to the crime's consequences (such as ours) is itself immoral, or at least ammoral.

Another odd implication of Kant's conclusion is that non-rational creatures are completely exempted from the system, as they do not (according to Kant) have free will, and therefore cannot be bound by morality. This means that all bets are off regarding animal cruelty. Because animals are non-rational, there is no restrictions on how one must treat them. To pick a grim example, slowly torturing a squirrel to death can certainly be willed as a universal without contradiction; the act doesn't treat a rational agent as a means; thus, it is permissible.

Kant says so much himself:
Beings whose existence depends not on our will but on nature's, have nevertheless, if they are irrational beings, only a relative value as means, and are therefore called things; rational beings, on the contrary, are called persons, because their very nature points them out as ends in themselves, that is as something which must not be used merely as means, and so far therefore restricts freedom of action (and is an object of respect).
One wonders whether this exemption from the strictures of morality applies to young children and the insane, who are also not capable of reason. If so, infanticide is permissible, as is the mistreatment of the mentally ill. Another chilling conclusion.

But perhaps the most striking thing about this chain of reasoning is that, as a result of Kant's disdain for empirical facts, a moral person has no reason to expect happiness. In fact, a person acting in accordance with the categorical imperative may reasonably expect to be miserable; their unerring code of behavior would make them easy prey for anyone who wished to take advantage of them.

This is not a theoretical objection to Kant. But one may reasonably ask, "then why be moral?" The only thing Kant can say is, "to be free." And if you ask, "why be free?" Kant's famous response is "to be worthy of happiness." But I'm sure many would rather take happiness than worthiness.


But was Kant Right?

Kant's argument rests on the premise that, when one acts rationally, one is not determined by anything else. Rationality, for Kant, is not part of the world of nature, and is therefore the basis of freedom.

I am extremely skeptical that this is the case. I do not see how anybody could make an absolutely free decision, independent of the normal laws of nature. We cannot, so to speak, take ourselves out of the stream of causation. It therefore seems more likely that freedom is an illusion, or a particular kind of ignorance. In Spinoza's words, "men are mistaken in thinking themselves free; their opinion is made up of consciousness of their own actions, and ignorance of the causes by which they are conditioned. Their idea of freedom, therefore, is simply their ignorance of any cause for their actions."

Thus, acting in accordance with Kant's principles would not make a person more or less free. Refraining from stealing based on the categorical imperative is just as "free" a decision as eating lunch because of hunger, or sleeping because of exhaustion. We are always both subject to random processes and to deterministic laws, and all our decisions are just as motivated by desires as the drug addict's. (Even the strict Kantian is motivated by his desire to abide by the categorical imperative.)

Kant makes the subtle and interesting argument that even if rationality doesn't actually make us free, the categorical imperative is still operative because, in order to act, we must assume we're free. In other words, Kant says that, even if freedom is an illusion, his conclusions still hold. But if freedom is an illusion, acting according to his principles might be literally impossible for sentient creatures (as I suspect is the case); so striving after some "ideal of reason" (as Kant calls it) hardly seems like the sensible thing to do.

Moreover, because we are not capable of completely free decisions, and because morality apparently does have its basis in empirical fact'if it can be said to exist at all'it behooves us to take into account things like human psychology, empirical conditions, cultural and historical forces, and consequences. A moral system that treats lies as equivalent to murder is impracticable; and a moral system that only binds rational agents may lead to inhumane acts. Finally, no person can be reasonably expected to abide by a moral system that will not lead to their own happiness.


Parting Thought

As I reread this book, a feeling suddenly took hold of me: admiration. I found myself almost in awe of Kant'both of his boldness and his genius. Even if I don't believe his premises are correct, I can't help but think it would be a beautiful thing if such a kingdom of ends were possible. It just so happens that the world isn't as beautiful as Kant's mind.