Here are three ways to color a map of the world:

1. Put dots on all the places (including seas) whose names you happen to like. That will produce a random-looking pattern. If people want to know whether the dots are in the correct locations, or why they are where they are, they must ask you, because the only truth is in your head–it is your set of preferences.

2. Color according to a rule, principle, or algorithm. A simple example would be: color everything above the half-way line. More complicated rules would produce much more complex patterns, even fractals. The colored map might be useful as a visual representation, but if there were any questions about what should be shaded or why, one would consult the rule. The words and/or numbers would be more precise than the map.

3. Color significant areas on the map, such as North America. This region could be described in words (a continent of approximately 25 million square miles wholly situated in the Western and Northern hemispheres. Its eastern littoral is predominantly a plain bordered by a mountain range that parallels the coast. A peninsula descends from the bottom. Etc.). It could also be defined by a kind of rule: any location reachable by land from Chicago, IL, without crossing an isthmus, plus some neighboring islands. Such rules and descriptions can be illuminating, but they only partially describe the truth, which is in the world. The fullest description is a detailed map.

Now imagine that instead of coloring a map of the world, we are looking at a map of all the actual and potential actions (or situations), and our task is to color the good ones (or the right or just ones). Our choice of methods resembles the three above.

1a. If you are a moral subjectivist or relativist, you think that the choice of what to color is a matter of private opinion. Even if people tend to cluster their dots within regular-looking shapes, and even if many people color the map the same way, the truth is in their heads, not on the map.

2a. If you subscribe to one of the classical philosophical schools, such as Kantianism or utilitarianism, you believe that the map should be colored using a rule or set of rules. Indeed, the map is not terribly useful because the rules will give you more accurate and reliable answers. One would check the rule to see if the map were accurate, not vice-versa.

3a. If you are a particularist, you believe that there is an objective difference between right and wrong–like the very important distinction between North America and the Atlantic Ocean. (There may also be some borderline cases and objectively gray areas.) The difference between right and wrong can be described in words, but any combination of words simplifies the reality, which is in the world, not in our heads. The shape of the good is complex because it consists of various institutions, norms, concepts, practices, and ways of life that have evolved over long history without conscious design. You must explore reality to know what is good and bad, but other explorers’ accounts and maps are valuable.

For what it’s worth, I think the reality is a combination of 2a and 3a. Rules mark valid moral distinctions in basic, elemental situations–for instance, is it right to kill?–but in dealing with evolved institutions, the rules no longer mark the important boundaries. In a similar way, part of the border between the United States and Canada is defined by a horizontal line (the distance from the North Pole), but part of the border follows the winding course of rivers and lakes. These bodies of water lie where they are because of the uniform laws of physics, but their evolution is so old and so often influenced by exogenous factors that the laws cannot tell us where we will find them. Note that defining our borders in two ways produces no contradiction. The boundary is straight in some places, winding in others.

Compare Wittgenstein, Philosophical Investigations, 18: “Our language can be seen as an ancient city: a maze of little streets and squares, of old and new houses, and of houses with additions from various periods; and this surrounded by a multitude of new boroughs with straight regular streets and uniform houses.”

Implications: To know the good, it is helpful to study and critically investigate moral principles, such as those analyzed in philosophy. That method may be necessary, but it is not sufficient. We must also explore and describe reality and form moral judgments of it. That is something that people do in ordinary life, but its academic corollaries are in fields like history, anthropology, and literary criticism. (For an argument that those disciplines should be more explicitly moral, please see my Reforming the Humanities.)