> How is a new dictionary supposed to source definitions without using current dictionaries? Aren't existing dictionaries essentially the "canon"?
No, because that isn't how dictionaries work. Dictionaries, and this goes back to the very first dictionaries, have always paid people to look at current usage and write definitions based on that. Dictionaries are descriptive, not prescriptive; they reflect usage, not control it.
The OED is well-known for its citations backing up its definitions, and they have a very strong bias towards actual usage as opposed to other dictionaries.
Edited to add: Some dictionaries are prescriptive to an extent. Those are the ones that define terms used in a controlled vocabulary, such as technical terms used in a specific field. When the CRC handbook says that this is what 'butyl' means, for example, it is pretty well taken to be prescriptive among chemists.
The wikipedia article doesn't say, but I'm curious if "esquivalience" or "dord" (another example) have been used in error to support their fictitious definitions, thus becoming suitable for inclusion in newer dictionaries.
Very similar to the effect when an otherwise "non-notable" wikipedia page is cited in the public press, and then cites the article citing it as a way to claim notability.
> I'm curious if "esquivalience" or "dord" (another example) have been used in error to support their fictitious definitions, thus becoming suitable for inclusion in newer dictionaries.
This would be interesting: Can adding a word to a dictionary be a method of introducing it into the language as it is actually used? I don't know, and, you're right, nobody seems to talk about that. This is the kind of thing you need corpus linguistics to figure out:
> Corpus linguistics is the study of language as expressed in samples (corpora) of "real world" text. This method represents a digestive approach to deriving a set of abstract rules by which a natural language is governed or else relates to another language.
Finding out how (or whether) a given word is used is right up that field's alley.
"Dictionaries are descriptive, not prescriptive; they reflect usage, not control it."
I've heard rumors that there are lexicographers have other ideas about this, but I've never actually found anything about lexicography other than what you've described.
Thanks for this post, I sometimes feel as if this is the world's least well made distinction. So many people seem to have constructed the idea that dictionaries instruct the use of language not vice versa.
How is a new dictionary supposed to source definitions without using current dictionaries? Aren't existing dictionaries essentially the "canon"?