Philippa Foot's article, "Morality as a System of Hypothetical Imperatives" should be listed as one of the most important articles in philosophy. (The Philosophical Review Vol. 81, No. 3 (Jul., 1972), pp. 305-316).
Before this, as she reports, philosophers took for granted Kant's claim that moral command represented categorical imperatives, distinct from the the hypothetical imperatives of practical reason. After this, the possibility of morality as a system of hypothetical reasons could not be easily dismissed.
To understand the issue, we must first understand the distinction between hypothetical and categorical imperatives.
Hypothetical imperatives are contingent on the wants of the individual.
Foot notes the advice we may give a traveler and its relationship to his interests.
Suppose, for instance, we have advised a traveler that he should take a certain train, believing him to be journeying to his home. If we find that he has decided to go elsewhere, we will most likely have to take back what we said: the "should" will now be unsupported and in need of support. Similarly, we must be prepared to withdraw our statement about what he should do if we find that the right relation does not hold between the action and the end-that it is either no way of getting what he wants (or doing what he wants to do) or not the most eligible among possible means.
Moral claims, however, seem to lack this hypothetical nature. If we say that a person ought not to take money from a co-worker's desk, we would not retract the statement if we were to learn that taking the money will allow him to buy the television he wants and he doesn't like the person whose money he is taking. Regardless of how it serves his interests, he should not take the money.
It seems clearly the case that Kant is right on this, and the task that remains for moral philosophers is to figure out the nature of these categorical imperatives.
Then Foot pointed out that moral rules are not the only rules that are independent of an agent's interests - that act like categorical imperatives.
She argued that the rules of etiquette as an example.
For instance, we find this non-hypothetical use of "should" in sentences enunciating rules of etiquette, as, for example, that an invitation in the third person should be answered in the third person, where the rule does not fail to apply to someone who has his own good reasons for ignoring this piece of nonsense, or who simply does not care about what, from the point of view of etiquette, he should do.
When presenting the rules of etiquette, we do not start by asking the agent what his interests are the way we ask for a destination when giving advice on which train to take. The rules of etiquette are independent of these types of considerations. They are, in this sense, much like categorical imperatives.
Foot also mentions club rules, where a person does not ask the agent (at least not in a way that suggests genuine interest) what his intentions are before he reminds the member that the rules require that men wear a suit and tie.
Foot suspects that many moral philosophers would want to classify the "should" of etiquette and club rules as types of hypothetical imperatives. The challenge is for them to do so in a way that does not also make the "should" of morality hypothetical imperatives - since they have so much in common.