In the context of morality, then, the question of what is “true” about the “nature” of right and wrong becomes, “for what purpose is morality designed, what description of it best serves that purpose, and does that purpose actually help us?” Hobbes would say that morality is designed to allow people to come together into societies, and so a view of morality in which the primary arbiter of rightness is the degree to which it encourages society would be the most “true” interpretation. It could be freely admitted, of course, that such a morality might lead to decisions that many in our society would consider “wrong” — putting an innocent man to death, for example, might be acceptable if to admit his innocence would destroy the legitimacy of the judicial system — but in that case either morality’s original intent was different, it was flawed to begin with and should be discarded in favor of something else, or it has been assigned duties and values (like individualism) over time that it was never meant to carry.

So the point, it seems, is that asking for definitions of truth or goodness is not quite as “natural” an activity as one might think. In fact, it’s fairly specific in its origins and applications, and indeed the fact that we can't answer questions such as “what is goodness?” doesn't mean that there's no such thing as goodness. It simply means that goodness does not have a metaphysical “essence” that can be defined. When it really comes down to it, we can't give a definition for anything else either. The question “what is a tree?” can be answered with “a tree is a large plant with leaves”, but that definition applies to many bushes and cacti as well, and not to, say, bonsai. To fully answer the question, we might need to make an appeal to science and say “Oh, a tree is any species within such-and-such an order within such-and-such a phylum.” But even these boundaries between tree and non-tree do not claim to have isolated and discerned the Fundamental Treeness. They are, if not arbitrary, than at least fuzzy. The same thing applies to terms like truth and goodness. We can give “paradigm cases of x-ness”, but if the question is repeated over and over we’ll finally be forced to admit that we can't just say “goodness is love” or “goodness is hope” because none of those things encompasses the whole of goodness. A full answer to the question would require the identification of the properties which are necessary and sufficient to render something right or wrong, and these properties simply may not exist. We may, in the end, be asking the wrong question.

What do we mean, then, when we say something is good or bad? Henry Aiken puts forth the idea that these terms are used on at least four distinct levels, levels he terms the expressive-evocative, the moral, the ethical, and the post-ethical level. By identifying and exploring these levels, Aiken shows with great clarity where the moral discussion begins and ends, and helps to weed out as purely emotive questions like “how can something be good if it doesn’t make anybody feel good?” which seem on the surface to present a dilemma, but really does not because of the overloaded meaning of “good”.

The expressive-evocative level is the level with which cultural relativists are so well acquainted, and is the realm of our second usage of the word “good” in our example. Here, the terms “good” or “bad” refer to an entirely subjective, emotive response to something. This response may be based on whether the something in question causes pleasure or pain, happiness or sadness; whether it is seen as beautiful and attractive or ugly and repulsive; whether past experiences or biases stir up fond memories or resentful ones; or any other emotive reaction of that sort. This is the level of discourse that can be assumed to be in play if I were to say to you, “The term is over, and I haven’t finished my essay on morality,” and you were to respond, “That’s too bad.” It is entirely possible that the reason for this conclusion is that you realize it will be bad for me, as I will doubtless be caused grief by its incompleteness, but the comment would (presumably) be due to an emotive reaction of sympathy on your part.

The next level of discourse is the moral one. This is the level of basic, practical decision-making using predefined moral rules derived from a set of ethics. To say “stealing this stereo system would be wrong” is to speak on the moral level. The immediate answer for why stealing the stereo system is wrong is that stealing is wrong, which is a moral rule.