Section II of the Groundwork takes a more philosophical starting point. Section I began with the seemingly ordinary concept of a “good will” and derived from it the “supreme principle of duty.” The concept of “duty” was defined in Section I as the “necessity of action from respect for the moral law.” The leading question of the more philosophical Section II is: what makes any action necessary?
Imperatives. Grammatically, an imperative is a commanding expression. Logically, an imperative is neither true nor false, unless it is expressed with special terms such as “ought,” or “necessary.” An expression like “Close the door!” cannot be either true or false; but imperative expressions like “You ought to close the door,” “You should close it,” or “It is necessary for you to close the door” can be. Imperatives express the necessity of action, and Kant’s leading concern in Section II is about what makes it necessary to act, or what makes any imperative true. It is almost obvious that an imperative is true when it expresses the necessity to act a certain way in order to achieve one of your goals—provided the action is necessary for achieving the goal. For example, when your goal is to do well on an upcoming exam, and when you need to study for the exam in order to do well, then the imperative “You should study for the exam” is true.
Instrumental reasoning. We typically see our actions as means to our ends. Actions can therefore be referred to as instruments, or as instrumental for bringing about ends. Reasoning tells us that when we have an end for which an action is a means, especially a necessary means, then the action is something we should do. This type of thinking is called instrumental reasoning. It was important for Kant to discuss instrumental reasoning because the philosophers he disagreed with in ethics believed that all reasoning related to action, all “practical” reasoning, is instrumental reasoning. They thought, in other words, that actions are morally necessary because they are means to a moral end sometimes known as “the highest good.” They did not always agree on what that moral end is, or how we can know what it is, but they pretty much agreed that practical reasoning is a matter of identifying the actions that are the necessary means to that end. Some, for example, thought that human beings were created in order to serve God’s purpose, and that they are truly happy only when they do. It follows that God’s purpose in creating human beings is our ultimate or highest end; and so in order to act morally we must do whatever is necessary to achieve that end, and omit whatever is not consistent with it. If lying is not consistent with the purpose for which God gave us the ability to communicate, then the imperative “Thou shalt not lie” is true.
Presupposed ends. Kant introduced a technical term for the imperatives that tell us to take actions as means to ends (goals). He called them “hypothetical” because they presuppose that we have the ends to which the actions they command are means (Kant 1997a, 24-27/4:413-16). Since for most ends it is not necessary that we have them, when we don’t, the hypothetical imperatives presupposing that we do are not true. Politeness in dealing with others is necessary in order not to offend anyone—but if I don’t care whether I offend anyone, then the imperative “You should be polite” is not true—at least when it refers to me. It is an “imperative of skill” for speaking Spanish that “You should place your adjectives after your nouns.” But it is not true for you if you do not want to speak any language other than English. There is one end that every rational human being necessarily has, according to Kant. That is the end of his or her own happiness. So it may at first seem that there are no hypothetical imperatives presupposing this end. Remember, an imperative is hypothetical if it is not necessary that you have the end it presupposes. But it is Kant’s view about happiness that even though everyone has it as an end, necessarily, it is not possible for anyone to tell exactly what makes up their happiness. Happiness is largely a condition of the future, and we cannot see very far into the future in order to tell what we will care about then, or what things we will need and wish we had. So for this reason Kant calls hypothetical imperatives presupposing the end of happiness “counsels.” They are, at best, good advice. “Save your money for spending it in your old age” is good counsel. But it will turn out not to have been such good advice if you do not live long enough to enjoy spending your savings.
Instrumentalism. According to some views of practical reasoning circulating today, the only rational thing to do, ever, is take the means to your ends. We should take our desires or preferences as informing us about our must fundamental ends, and recognize that the only actions necessary for us to do are those that contribute to realizing those ends. This is not necessarily a self-interested or purely “egoistic” outlook. Most of us care about other things beside ourselves. We have loved ones and we care about what makes them happy, too. The point of instrumentalism is simply that an action’s serving one of your personal ends is all that can ever make it necessary for you to do it, or give you a reason to do it. According to instrumentalism, then, all imperatives are hypothetical in the sense that Kant explained, even moral imperatives (see Foot 1972). But Kant and most of those who tend to agree with him reject instrumentalism. For Kant thought that some actions are necessary, and so are commanded by true imperatives, regardless of what ends a person may have. In other words, Kant’s ethics supposes that not all imperatives are hypothetical: that the necessity of some actions does not presuppose any ends (see Kant 1997b, 19-24/5:22-6).
The Hypothetical Imperative. The point is sometimes put this way, although it is a bit subtle. Kant wrote, as quoted above, that whoever wills an end, if she is rational, wills the means within her power that are necessary to achieve it (Kant 1997a, 28/4:417). This is often taken to mean that someone unwilling to take what she discovers to be a necessary means to her end is irrational—unless she then gives up her end. Someone in that situation should therefore either take the means or give up the end. But notice that this is an imperative: “You should take the necessary means or give up the end.” That seems to be a true imperative; and it seems to be true for every rational person, regardless of what ends they may have. Some Kant scholars refer to this imperative not as a hypothetical imperative, but as The Hypothetical Imperative (Hill 1973). It seems to be a basic principle of practical reasoning, and seems to be something an instrumentalist about practical reasoning would agree with, also. But how can an instrumentalist agree? For instrumentalism is the belief that no imperatives are true unless they presuppose one of your ends; and The Hypothetical Imperative seems to say that something is necessary for you to do regardless of what ends you have: take the means or give up the end. Kantian critics of instrumentalism therefore point out that it looks like there is more to practical rationality than simply taking the means to your ends. That is what Kant thought, too; since he thought that practical rationality is the basis for moral action, and acting morally is more than a matter of acting in order to satisfy your desires or preferences. It seems to me, however that an instrumentalist could respond to this type of critique by saying the following: some people, probably most, prefer not to have ends that frustrate them. So, The Hypothetical Imperative is true for those people, in general, because it serves their preference. Despite its capital letters, therefore, it is just another ordinary, hypothetical imperative: people with a low tolerance for frustration should try to give up any ends for which they are unwilling to take the means. It might seem irrational to prefer to have ends like this, but I think an instrumentalist could say that this depends on the ends, and on a person’s capacity to live with frustration. People can find satisfaction in the mere fact of their having certain ends, even if they may be unwilling to pursue them. That does not necessarily mean they are irrational; it means only that they prefer having the end they are not willing to pursue to not having the end, or to trying to get themselves to stop having the end.
The problem with hypothetical imperatives. From Kant’s point of view, although imperatives in hypothetical form tell us what we ought to do, they are not moral imperatives. Remember that the “good will” is good because of willing to do what is right, or morally good. If moral imperatives were hypothetical, then the good will would be good merely because of its willing the end these imperatives presuppose; it would not be good because of willing what is morally right or good to do. For example, suppose that world peace is the end or ultimate purpose morality is to achieve. In that case, an action that would contribute to causing world peace, like donating to an international charity, could be morally right. Given this picture, the good will would be one that wills to achieve world peace. It would not be one that wills to do what is right, or what is morally good to do. That may seem like a very subtle difference, but in Kant’s mind it is significant. The good will, which is the only thing good without limitation, wills its object through a conception of its object’s morality. Morality is the good will’s target, in other words, not world peace. That of course does not mean that peace isn’t morally good. It means that what a good will would care about is the moral goodness of peace. So the hypothetical imperative, “Donate to an international charity” (because you care about world peace), is not the imperative followed by the good will. Rather, the good will would observe the imperative: “Donate to an international charity” (because it is morally good to do). That imperative is not hypothetical but categorical. It does not command the action because you desire to achieve some end; it commands doing the action because of its morality. That’s really what it means to say that the good will acts from respect for the moral law, and that “duty” can be defined as “the necessity of action from respect for law.”