Is and Ought

In Book III of his Treatise of Human Nature, David Hume argues that reason and morality are completely independent:

Reason is the discovery of truth or falsehood. Truth or falsehood consists in an agreement or disagreement either with the real relations of ideas, or with real existence and matter of fact. So anything that isn’t capable of this agreement or disagreement isn’t capable of being true or false, and can never be an object of our reason. Now, our passions, volitions, and actions are basic facts and realities; they are complete in themselves and aren’t in any way about other passions, volitions, and actions; so they aren’t capable of either of those sorts of agreement or disagreement; so they can’t be sorted into ‘true’ and ‘false’, and can’t be either in conflict with reason or in accord with it.

It is true that an action is not true or false, nor can it be the conclusion of an argument. And in that sense there is some separation between action and reason, as Hume says here. Nonetheless, Hume intends to imply something more, namely that reason cannot tell us what we should do. If reason can tell us what we should do, then doing something else instead is an action in conflict with reason, even though it is not a false statement.

Later, Hume clarifies his position:

I can’t forbear adding an observation that may be found of some importance. In every system of morality I have met with I have noticed that the author proceeds for some time reasoning in the ordinary way to establish the existence of a God, or making points about human affairs, and then he suddenly surprises me by moving from propositions with the usual copula ‘is’ (or ‘is not’) to ones that are connected by ‘ought’ (or ‘ought not’). This seems like a very small change, but it is highly important. For as this ‘ought’ (or ‘ought not’) expresses some new relation or affirmation, it needs to be pointed out and explained; and a reason should be given for how this new relation can be—inconceivably!—a deduction from others that are entirely different from it. Authors don’t ordinarily take the trouble to do this, so I recommend it to you; and I’m convinced that paying attention to this one small matter will subvert all the vulgar systems of morality and let us see that the distinction between vice and virtue is not based merely on the relations of objects, and is not perceived by reason.

Hume suggests that it is “inconceivable” that a statement involving the word “ought” would derive from statements without that word. This can be taken as a mere point of logic: the conclusion of a syllogism will not contain a term which is not contained in the premises. But if we understand it in this way, his point is true but rather unimportant, at least in relation to his argument about morality. For the same thing is true of all terms. In the statement, “Trees have leaves and branches,” the word “trees” cannot possibly be logically derived from statements that do not mention trees. It does not follow that statements about trees are somehow cut off from all the rest of reality. It does not even follow that people cannot explain what they mean by a “tree.” Someone can explain this by referring to various fairly common experiences that people have.

So as a logical point, this proves nothing in particular about morality, although it is a partial explanation for the difficulty of philosophy. I pointed out in the discussion of Spinoza’s Ethics that the fact that philosophy wishes to speak about reality in general implies that it must use a practically unlimited number of terms, and consequently that it cannot be built up as a formal system like geometry.

In his Nicomachean Ethics, Aristotle discusses ethics as a kind of art which aims at an end:

Every art and every inquiry, and similarly every action and pursuit, is thought to aim at some good; and for this reason the good has rightly been declared to be that at which all things aim. But a certain difference is found among ends; some are activities, others are products apart from the activities that produce them. Where there are ends apart from the actions, it is the nature of the products to be better than the activities. Now, as there are many actions, arts, and sciences, their ends also are many; the end of the medical art is health, that of shipbuilding a vessel, that of strategy victory, that of economics wealth. But where such arts fall under a single capacity- as bridle-making and the other arts concerned with the equipment of horses fall under the art of riding, and this and every military action under strategy, in the same way other arts fall under yet others- in all of these the ends of the master arts are to be preferred to all the subordinate ends; for it is for the sake of the former that the latter are pursued. It makes no difference whether the activities themselves are the ends of the actions, or something else apart from the activities, as in the case of the sciences just mentioned.

If, then, there is some end of the things we do, which we desire for its own sake (everything else being desired for the sake of this), and if we do not choose everything for the sake of something else (for at that rate the process would go on to infinity, so that our desire would be empty and vain), clearly this must be the good and the chief good. Will not the knowledge of it, then, have a great influence on life? Shall we not, like archers who have a mark to aim at, be more likely to hit upon what is right? If so, we must try, in outline at least, to determine what it is, and of which of the sciences or capacities it is the object. It would seem to belong to the most authoritative art and that which is most truly the master art. And politics appears to be of this nature; for it is this that ordains which of the sciences should be studied in a state, and which each class of citizens should learn and up to what point they should learn them; and we see even the most highly esteemed of capacities to fall under this, e.g. strategy, economics, rhetoric; now, since politics uses the rest of the sciences, and since, again, it legislates as to what we are to do and what we are to abstain from, the end of this science must include those of the others, so that this end must be the good for man. For even if the end is the same for a single man and for a state, that of the state seems at all events something greater and more complete whether to attain or to preserve; though it is worth while to attain the end merely for one man, it is finer and more godlike to attain it for a nation or for city-states. These, then, are the ends at which our inquiry aims, since it is political science, in one sense of that term.

If we understand ethics in this way, this explains what is meant by saying that something “ought” to be done, because we know what is meant when we say, for example, “If you want to attach those boards with a nail, you ought to use a hammer.” This implies that a hammer is useful, or perhaps even necessary, in relation to the goal in question. In fact, “ought” is sometimes replaced with “must,” which is a word directly expressing necessity.

Of course, “you must use a hammer” does not imply that there is a physical necessity that you use a hammer, so that you cannot avoid using one. The necessity is a hypothetical one: given that you are going to obtain the end, you must use a hammer. If you do not use a hammer, you will not obtain the end.

This explains, in a certain way, why Hume has difficulty with the point. A repeated concern in his work is that he cannot see a way to understand the idea of necessity. He says for example that he has a hard time understanding any “necessary connection” between cause and effect:

Having thus explained how we reason beyond our immediate impressions, and conclude that such and such causes must have such and such effects, we must now retrace our steps and pick up again the question that first occurred to us, and that we dropped along the way (near the end of section 2). The question is: What is our idea of necessity, when we say that two objects are necessarily connected? As I have often said already, if we claim to have such an idea we must find some impression that gives rise to it, because we have no idea that isn’t derived from an impression. So I ask myself: In what objects is necessity commonly supposed to lie? And finding that it is always ascribed to causes and effects, I turn my attention to two objects that are supposed to be related as cause and effect, and examine them in all the situations in which they can occur. I see at once that they are contiguous in time and place, and that the one we call ‘cause’ precedes the one we call ‘effect’. In no instance can I go any further: I can’t find any third relation between these objects.

He does go on to say that there is something else, but in turns out to be something like the vividness with which the mind expects the effect, or in other words, a property of the mind, not of the causes and effects. Similar concerns lead him to deny the possibility of probable knowledge about the future, and here to effectively deny the possibility of the knowledge of morals. If moral knowledge consists in knowing that something must be done, even for the sake of an end, then according to Hume, it is not possible to know this. If we cannot even know that the sun will rise tomorrow, we also cannot know that we need to use a hammer in order to attach two boards.

Aristotle’s account of ethics is, I think, the correct account. But people find it troubling, mostly because they suspect that it has unlikely or unpleasant implications. If saying that you “ought” to do something only means that you must do it in order to obtain a certain end, what if you do not care about that end? There will no longer be a need to do that thing. Likewise, if you do want certain ends, then perhaps you ought to do certain things which we really think you ought not do. For example, if you want to inherit money, perhaps you ought to murder your elderly relatives. But this consequence is absurd: of course you ought not to murder your relatives.

In order to be sure e.g. that you ought not to murder your relatives, regardless of what ends you have in mind, people prefer something like Immanuel Kant’s categorical imperatives. I will not discuss Kant’s moral philosophy in detail at the moment, nor am I necessarily asserting that the usual understanding of his philosophy is correct. But the usual idea is that the rule is absolute: “you must not kill” is not relative to an end, but stands in itself.

This seems to me to be a mistake, and one where Hume’s criticism would be valid. There is obviously no physical necessity that you abstain from killing. It is quite possible to kill people, and it is sometimes done. So if you say that you “must” not kill, and you do not refer to hypothetical necessity, to what do you refer? There does not seem to be anything left here. And this is probably the motivation behind error theory. The obvious interpretation of the theory is that it is saying, “I do not believe in right and wrong.” In reality, however, it seems to be based on the mistaken assumption that right and wrong refer to something like categorical imperatives, and then it proceeds to rightly deny the existence of these things. For no one except an actual follower of Hume would deny the existence of hypothetical necessity, namely that if you want to obtain certain ends, you must use certain means. In this sense, error theory is mistaken, but it is mistaken much in the way that Richard Dawkins is mistaken to suppose that theists intend to speak of something complex when they speak of God.

The objection to Aristotelian ethics, then, is that it is too flexible, and may ultimately imply that people may do whatever they want. I would respond to this by making two claims, although I will not establish their truth in this post.

First, the truth about ethics is in fact more flexible than people suppose. If you take extremely complex situations, the moral truth about those situations will be complicated, not simple.

Second, the truth about ethics is less flexible than people tend to suppose would follow from an Aristotelian account. It does not follow in any meaningful way, for example, that it is sometimes fine to murder your relatives in order to obtain an inheritance.