In late 2008 I put myself through a crash course in the works of Willmoore Kendall, the “wild Yale don,” as Dwight Macdonald called him, who had been one of the founding senior editors of National Review. This was research for an essay that would appear in The Dilemmas of American Conservatism. I’d read some Kendall before—a desultory stroll through The Conservative Affirmation in America, at least—and hadn’t profited much from the experience. But the second, more attentive perusal was different. Kendall himself had told of how R.G. Collingwood had taught him at Cambridge to read a book by asking what question the author was trying to answer. I didn’t find that approach too insightful, but I picked up something else from Kendall’s own methods—the habit of asking “What conditions would have to be true in order for this author’s arguments to make sense?”That’s a more productive thing to ask of a serious work than simply, “Do this author’s arguments make sense?” The latter invites the reader to supply a misleading context: the author’s arguments may not match up with reality, but they must match up at least with his own view of reality, and that’s something worth figuring out and contrasting against whatever the reader thinks he already knows.