But at the time when I fell under his influence he stood more aloof; and this made him the more impressive to a youthful atheist. He had a keen sense of language and its imperial influence on men; language contained all the great and sound metaphysics, he was wont to say; and a word once made and generally understood, he thought a real victory of man and reason. But he never dreamed it could be accurate, knowing that words stand symbol for the indefinable. I came to him once with a problem which had puzzled me out of measure: What is a cause? why out of so many innumerable millions of conditions, all necessary, should one be singled out and ticketed "the cause"? "You do not understand," said he. "A cause is the answer to a question: it designates that condition which I happen to know, and you happen not to know." It was thus, with partial exception of the mathematical, that he thought of all means of reasoning: they were in his eyes but means of communication, so to be understood, so to be judged, and only so far to be credited. The mathematical he made, I say, exception of: number and measure he believed in to the extent of their significance, but that significance, he was never weary of reminding you, was slender to the verge of nonentity. Science was true, because it told us almost nothing. With a few abstractions it could deal, and deal correctly; conveying honestly faint truths. Apply its means to any concrete fact of life, and this high dialect of the wise became a childish jargon.