A common objection to reductive moral naturalism (aka ‘ethical naturalism’) is the so-called “naturalistic fallacy.” This fallacy comes into flavors: logical and semantical.
The Logical Form
This version of the naturalistic fallacy is normally referred to as the is-ought fallacy, the fact-value fallacy, or, in honor of its author, Hume’s Law. The source of this form of the naturalistic fallacy is the following passage by Hume.
In every system of morality which I have hitherto met with, I have always remarked, that the author proceeds for some time in the ordinary way of reasoning, and establishes the being of a God, or makes observations concerning human affairs; when of a sudden I am surprised to find, that instead of the usual copulations of propositions, is, and is not, I meet with no proposition that is not connected with an ought, or an ought not. This change is imperceptible, but is, however, of the last consequence. For as this ought, or ought not, expresses some new relation or affirmation, it is necessary that it should be observed and explained; and at the same time that a reason should be given, for others, which are entirely different from it. (A Treatise on Human Nature, Book III, Part I)
Although this passage is often used to support a distinction between facts and values, the passage does not accomplish that. As Charles Pidgen writes, “For Hume is making a simple logical point. A conclusion containing ‘ought’ cannot (as a matter of logic) be derived from ‘ought’-free premises.” Hume’s logical point does not refute the ontological claim of reductive moral naturalism.
To understand why, consider an analogy with the biological concept of species. There is a similar gap between conclusions about species and premises that make no mention of them. A conclusion containing ‘species’ cannot be derived from ‘species’-free premises. Yet nobody proposes a fact/species distinction or alleges that there is a realm of irreducible species facts.
Therefore, metaphysics does not create either gap. Rather, as Pidgen writes, “It is the conservativeness of logic that creates the gap.” But this entails that the gap is logical, not ontological.
The Semantical Form
In his historic book, Principia Ethica, G.E. Moore introduced and defended the theory of moral anti-reductionism (aka ‘ethical nonnaturalism’) over and against moral reductionism. Moore argued that moral reductionism is fatally flawed since moral facts and properties cannot be reduced to other types of facts and properties. To support his claim, Moore advanced what he called “the open-question argument.” According to Moore, no matter which naturalistic definition of good we consider (X), it is still meaningful (and hence an ‘open question’) to ask whether X is good. So, for example, if an ethical naturalist claims that goodness is identical with pleasure, the nonnaturalist could reply, “Rape gives rapists pleasure, but is it good?” Does Moore’s argument succeed in refuting ethical naturalism?
Moore’s argument may well be successful in showing that goodness is not strictly a synonym with any non-moral word. Thus, semantic naturalism would be false; moral words would not be synonymous with non-moral words. But, even if that were the case, moral words and non-moral words might both stand for the same property. To use an example outside of meta-ethics, consider the relationship between water and H2O. As Pidgen explains:
Water and H2O are identical. Yet ‘water’ is not synonymous with ‘H2O’ even though they stand for the same stuff. ‘Water’ expresses a pre-scientific concept accessible to children and savages – roughly, the colourless, tasteless fluid that falls from the sky and is found in lakes and rivers. ‘H2O’, by contrast, expresses a scientific notion. You can’t understand it fully without a modicum of chemistry. People did not find out that water is H2O by meditating upon meanings. Empirical enquiry did the trick. Semantic naturalism might be false but synthetic naturalism might be true. That is, moral properties might be identified with natural properties by means of empirical research rather than conceptual analysis. Thus semantic autonomy, which says that moral words do not mean the same as any others, does not entail ontological autonomy – that moral properties are not identical with any others.
Thus, neither the logical nor the semantic forms of the naturalistic fallacy succeed in refuting the ontological claim of reductive moral naturalism.
 Charles R. Pidgen, “Naturalism” A Companion to Ethics (ed. Peter Singer, Oxford: Blackwell, 1991), p. 423.
 Pidgen 1991, p. 424.
 Pidgen 1991, pp. 426-27.