Reconstruction talk:Proto-Germanic/kankatjaną

RFD discussion: June 2020–January 2021
The OE suffix was productive, leaving no reason to assume this word dates back to PG. -- 22:40, 29 June 2020 (UTC)
 * Is the entry any better? Seems like an ablaut to, and as easily 🇨🇬, the only descendant given, is an ablaut to the Old English descendant of . So of course the comparisons with 🇨🇬 and 🇨🇬 are coincidental. It’s written on both pages since 2008 and looks like something Pokorny would write. Fay Freak (talk) 00:10, 30 June 2020 (UTC)
 * Yeah it should also be burned with fire. I'm pro deleting any proto entry with a single descendant. -- 01:01, 30 June 2020 (UTC)
 * A proto entry with apparent (genetic) cognates outside the group has some validity, and looks highly plausible.  'Not proven' has to be a verdict in some cases.  I'm not sure if one can assign a numeric probability to its being a coincidence. --RichardW57 (talk) 10:00, 5 July 2020 (UTC)
 * 🇨🇬 is another entry with only one descendant, but cognates outside the group. Leasnam (talk) 04:08, 22 July 2020 (UTC)
 * I'm very much opposed to the blanket deletion of reconstruction entries with a single descendant because, as has been pointed out, having multiple descendants is not the only possible justification for assuming an unattested word existed in any given language. If a word cannot have been formed or borrowed synchronically in a given language (such as appears to be the case with ), it must be inherited from a parent language in which it could have originated, justifying an entry in that parent language because the word must have existed in that language. That said, delete *kankatjanan - canc + -ettan seems like a solid etymology and there is no need to assume it formed in a parent language of OE. — Mnemosientje (t · c) 10:15, 27 July 2020 (UTC)
 * RFD deleted. — Mnemosientje (t · c) 11:07, 11 January 2021 (UTC)