This was fun to read — not only does Liberman lay out very sound methods for determining the histories of words, with occasional appropriate ridicule of the storytelling that some etymologists have engaged in, but he peppers his prose with wordplay and wit. He introduces some ideas that were a bit unexpected to me. For example, he thinks the role of "sound symbolism" is quite important — that is, people either alter existing words or create ones with an appropriate sound to the subject, like pig, pug, pod, pad, etc. being appropriate to something "swollen". Also, his idea of etymology hopes to get to the origin of the word — either when it was coined or when it split off from related words in a sort of speciation event. I hadn't considered being that daring. But he's appropriately cautious about the subject of the origin of language itself, of course.