Mentioned in a footnote in that book is the following, which I have always rather liked: "A science is said to be useful if its development tends to accentuate

Mentioned in a footnote in that book is the following, which I have always rather liked: "A science is said to be useful if its development tends to accentuate the existing inequalities in the distribution of wealth, or more directly promotes the destruction of human life." (He mentions it in a footnote mostly to clarify that he doesn't really quite believe it -- it was a "conscious rhetorical flourish" in something he wrote in 1915, and in the main body of the text he gives a less cynical account of what it means for something to be useful.)