SOURCE: "Fact, Value, and Nonnatural Predication," in Ought, Reasons, and Morality: The Collected Papers of W. D. Falk, Cornell University Press, 1986, pp. 99-122.

In the following essay, Falk attempts to clarify Moore's distinction between "good" and natural properties.

I

Twentieth-century views on value are broadly divided between non-naturalism and noncognitivism. The choice is between saying that 'x is good' ('is good as such', 'good to experience', 'have', 'behold', 'ought to be', 'ought to be done') asserts some fact or truth, but one which is knowable only in some extraordinary and unique way; and saying that 'x is good', and so forth, does not primarily assert anything at all, but is a way of speaking commendingly or directively. These are opposed views, but they share a common bond. They agree that if 'x is good', and so on, stated any fact or truth, it could only be...