So what’s really clear, you can see it in the videotapes of the experiment, is: people give a reason. When that reason is stripped from them, they give another reason. When the new reason is stripped from them, they reach for another reason. And it’s only when they reach deep into their pocket for another reason, and come up empty-handed, that they enter the state we call “moral dumbfounding.” Because they fully expect to find reasons. They’re surprised when they don’t find reasons. And so in some of the videotapes you can see, they start laughing. But it’s not an “it’s so funny” laugh. It’s more of a nervous-embarrassment puzzled laugh. So it’s a cognitive state where you “know” that something is morally wrong, but you can’t find reasons to justify your belief. Instead of changing your mind about what’s wrong, you just say: “I don’t know, I can’t explain it. I just know it’s wrong.”
So the fact that this state exists indicates that people hold beliefs separate from, or with no need of support from, the justifications that they give. Or another way of saying it is that the knowing that something is wrong and the explaining why are completely separate processes.