The Gettier argument against the thesis that justified true belief is sufficient for knowledge has been stated from the first with artificial examples. It is possible and important to naturalize the argument. Then we can appreciate the extent to which it is a skeptical argument. We thereby get a better fix on the points at which the skeptic can attack the possibility of knowledge. We cart also see that some recent defenses of the thesis do not come to grips with the essentials of the argument.
Let me elaborate upon the skeptical potential of the argument. As far as the argument is concerned, a belief may be not only true but justified. This is what the skeptic is usually thought to deny, while the point of the argument is rather that, even when joined, truth and justification do not guarantee knowledge. But when we consider what else is needed -or, better, what else is needed that can sometimes be had -the answer comes, once the argument has been naturalized: a little luck. And to require luck for knowledge might naturally be thought to lead to skepticism in any case. I do not see that it needs to. An argument must first be given. Still, luck is involved, whether in getting justification in the first place, or in getting from justification to knowledge in the second. If skepticism is fear of bad luck, the thesis that justified true belief is sufficient for knowledge will not ward it off. And I think that it is better natural history of concepts to locate the need for luck in the gap between justification and knowledge.
Every evening for years, Sneed and his wife went their separate ways, he in his car, she in hers, he to his club, she to her bingo game. She always returned late, expecting him to have preceded her home. His car was always there, always in the same place. And lights were always on in the house, too, always the same ones. "Sneed is home," she would conclude. "It was always thus." And it was.