"If there is no God, the labels “good” and “evil” are merely opinions. They are substitutes for “I like it” and “I don’t like it.” They are not objective realities."
(Dennis Prager, in a recent piece in National Review Online)
The argument is wrong twice over. The existence of a god does not solve the problem of justifying right and wrong and there are solutions that do not require a god.
The existence of a god does not solve the problem because we need some reason to conclude that the god is good, that his will defines what we ought to do. The existence of a very powerful, perhaps all powerful, being who created us and the universe does not, by itself, imply anything at all about right or wrong. He could be a devil, he could, like gods in many religions, be no more morally perfect than humans, he could be a moral nihilist with no views at all on good and evil and a wicked sense of humor. To get from a god to God in something like the Christian sense, you need some further basis for moral beliefs, some way of deciding that the god is good.
If the problem is soluble, it is soluble without a god. One solution, the one that strikes me as the least unsatisfactory, is to posit the existence of moral truths analogous to physical truths, perceived by a moral sense analogous to physical sight or hearing. That describes the world as almost everyone actually perceives it—there are not many people who do not see torturing small children for fun as wicked. And that view of moral reality can be confirmed in the same way we confirm our view of physical reality, by subjecting it to consistency tests. If there is a moral universe out there, there ought to be a reasonably good correlation across people in their fundamental moral perceptions. It is arguable that there is, providing we think of moral perceptions at a sufficiently fundamental level.
The reason the claim of moral consistency across people and cultures seems wrong is that we are used to talking about moral beliefs in terms of general moral principles, about which people quite often disagree. But people also quite often disagree about questions of physical reality viewed at that level—for instance what the sources are of global warming and what its consequences are, or whether Obama's deficit spending did or did not reduce the unemployment level over what it would otherwise have been. If we consider moral perceptions at a more fundamental level, evaluations of fully described situations with all factual disagreements resolved, they look a lot more consistent. Executing witches seemed right to King James, wrong to us—but then, he believed in witchcraft and we do not, a disagreement about facts not about morality.
For a more thorough defense of this approach to moral philosophy, referred to as "intuitionism," I recommend Michael Huemer's book on the subject. I was first persuaded of its plausibility when, as an undergraduate, I got into an argument with Isaiah Berlin and lost it. His essential point was not that the evidence for moral reality was stronger than I thought but that the evidence for physical reality was weaker. If we applied the same standards to testing morality reality—rough consistency at the most basic level of perception—the case for it does not look that much worse.
Prager also writes:
If there is no God, the labels “good” and “evil” are merely opinions. They are substitutes for “I like it” and “I don’t like it.” They are not objective realities.
Every atheist philosopher I have debated has acknowledged this.
I do not acknowledge it and, although not a professional philosopher, I am an atheist. I have emailed him offering to debate him on the subject but have not yet received a response.