Today, in a letter to John C. Wright, I fell into a digression on neologisms, and one of the possible reasons why some of them catch on and others fail. I thought it might be as well to repeat it here, and throw it open to my 3.6 Loyal Readers for discussion or demolition:

One wants names for things, not for un-things. One may need new words to express new facts, but a lie, to be effective, must be tricked out in language that the intended victim already understands.

If I discover a species of rabbit previously unknown to science, I may point at it and say, ‘That is a zeffle.’ I have done well: I have made a new name for a new thing. If anyone asks ‘What is a zeffle?’ I can appeal to the facts by showing them the animal. But if I point at a plain old-fashioned domestic rabbit, and say, ‘That is not a rabbit, but a smeerp,’ my words will not convince even the most gullible, because there is no fact to appeal to. They have no standard of ‘smeerp-hood’ in their minds, so the word does not communicate any ideas to them, not even false ones.

If I said, ‘That is not a rabbit, but a horse,’ I would at least communicate a meaning. If I were to say, ‘That is not a rabbit, but a hare,’ I would move into the realm of the plausible, where all lies must have their being if they are to prosper.

It is for this reason that the most skilful liars work not by inventing new words, but by distorting and perverting the meanings of old ones.