Though aware of the impossibility of it, he would have liked to find some acid test to subject works of art to which would tell the scientific investigator whether they were good or bad. Survival, he typically decided, was the only measure of greatness, but of course this leaves the problem of what causes an author to last. He devised a test for characters in fiction: a character in a novel “passes” if you can hold an imaginary conversation with him. In his own novels, only Big Brother, probably, would meet that eccentric requirement. He was a Sherlock Holmes fan and a lover of puzzles and brain-twisters, also of the odd fact of the “Believe It or Not” variety. His book reviews often turn on the methodical solution of a puzzle (“What’s wrong with this picture?”), as when he discovers—quite astutely—that the fault of Koestler is “hedonism,” something that is not apparent to the untrained eye. He was not a natural novelist, having no interest in character or in the process of rising or sinking in ordinary society or in a field of work—a process that engaged the sympathies not just of Proust or Balzac but of Stendhal, Flaubert, Tolstoy, Jane Austen, George Eliot, Conrad, Zola, Dickens, Dreiser. He would have been indifferent both to success and to failure. It is hard to imagine the long family-chronicle novel in several volumes he was planning to write just as the war was breaking out. Maybe he did not have enough human weaknesses to be a real novelist.