In his book, Ethical and Religious Thought in Analytic Philosophy of Language, Quentin Smith defends an argument for moral realism which he calls the argument from veridical seeming.
(1) Ordinary ethical sentences and commonsense first-level moral beliefs imply moral realism (or “Moral realism tacitly seems to be true in ordinary commonsense moral attitudes”).
(2) There are no empirical or a priori reasons to believe that first-level moral beliefs are all false.
(3) Therefore, it is more reasonable to believe moral realism that not to believe this.
(4) There is no reason to believe that the conjunction of (1) and (2) is a defective reason to believe moral realism.
(5) Therefore, the belief in moral realism is indefeasibly justified.
In this post, I’m going to sketch a brief objection to (4) based on what I will call “naturalistic evolution.” According to this objection, naturalistic evolution furnishes naturalistic evolutionists who are also moral realists with a defeater for their belief in moral realism, a defeater which cannot be defeated.
Let us begin by reviewing Smith’s definitions of key terms.
“moral realism” = df. “the metaethical theory that human life has an objective ethical meaning,” viz., “moral facts obtain independently of whether humans believe they obtain.”
“objective ethical meaning” = df. “ethical sentence have truth-value and sometimes correspond to moral facts that obtain independently of our beliefs about whether they obtain.”  “If human life has an objective ethical meaning, then there is a class of intrinsic goods, a class of properties and relations that possess the property of goodness.”
“first-level ethical belief” = df. the belief that “something is good or evil or that something is of equal or greater value than something else, for example, that philosophical understanding is at least as valuable as aesthetic enjoyment.”
“second-level ethical belief” = df. a belief “about some or all first-level ethical beliefs. The belief that ‘the intuition that the proposition that philosophical understanding is at least as valuable as aesthetic enjoyment is true does not absolutely justify belief in the proposition’ is an example of a particular second-level ethical belief, and the belief that “life is meaningful but absurd’ is an example of a general second-level ethical belief.”
To Smith’s definitions I will add the following definitions (all taken from Paul Draper):
“hypothesis” = df. a proposition which we do not know with certainty to be true or false
“metaphysical naturalism” = df. the hypothesis that the universe is a closed system, which means that nothing that is not part of the natural world affects it
“genealogical thesis” = df. complex life evolved from simple life
“genetic thesis” = df. all evolutionary change in populations of complex organisms either is or is the result of trans-generational genetic change
“evolution” = df. the genealogical thesis conjoined with the genetic thesis
The objection goes as follows. If evolution is true, then human beings have developed from non-human animals as a result of natural selection, genetic drift, etc. As many writers have observed, evolution provides a plausible reason to expect (1) even on the assumption that moral realism is false. Theism provides a strong antecedent reason to trust the reliability of our metaethical intuitions (i.e., our second-level ethical intuitions), namely, God, as a morally perfect being, would want to ensure that all moral agents had moral intuitions which corresponded with what we might call “moral reality.” In contrast, if metaphysical naturalism is true, there is no God overseeing our development and orchestrating the course of our evolution, including the evolution of reliable metaethical intuitions. In short, “blind nature” provides us with no antecedent reason at all to believe that our metaethical intuitions are correct.
The potential unreliability of naturalistic metaethical intuitions does not prove that moral realism is false. (To suggest otherwise would be to commit the genetic fallacy by confusing moral epistemology with moral ontology.) For all the metaphysical naturalist knows, it could be the case that both metaphysical naturalism and moral realism are true. Nevertheless, a metaphysical naturalist’s belief in naturalistic evolution seems to undermine a metaphysical naturalist’s belief that moral realism is true. In other words, it could be the case that both metaphysical naturalism and moral realism are true, but if one knows the former, one cannot know the latter. At least, that’s what this objection claims.
This argument is similar to Alvin Plantinga’s famous Evolutionary Argument Against Naturalism, but it has an important difference. As I understand it, Plantinga’s EAAN seeks to show that the naturalist has what I will call a ‘global defeater,’ i.e., a defeater for all of her beliefs, including her belief that naturalism is true. In contrast, the objection I’ve sketched above only claims there is a ‘local’ (or localized?) defeater, i.e., a defeater just for the belief that moral realism is true.
 Quentin Smith, Ethical and Religious Thought in Analytic Philosophy of Language (New Haven: Yale University Press, 1997), 171-72.
 Smith 1997, 159.
 Smith 1997, viii.
 Smith 1997, 159.
 Smith 1997, 11.
 Smith 1997, 18.
 Smith 1997, 19.