Talk:Language family

Misleading part of the map
Parts of the map are very misleading. It seems to be supposed to show the present-day language of the majority population, but this principle is not observed for Canada, which is shown as almost entirely Algic and Na-Dene-Yenisseian, even though it is observed for the US, which is shown as Indo-European. --82.137.111.223 (talk) 02:17, 31 January 2019 (UTC)


 * This actually isn't that misleading, unless the point was that the indigenous people have been culturally assimilated to the point where they lost their language. Most settlers settles in the PIE area (unless the map changed, I didn't check the history of the page.) Robust21 (talk) 17:29, 4 May 2023 (UTC)

Merge proposal
I propose merging Genetic relationship (linguistics) into this page. The two pages are about exactly the same concept: what it means for languages to be in the same family is that they have a genetic relationship. We don't need two different pages about this topic. AJD (talk) 16:41, 7 July 2023 (UTC)
 * Weak support. It would indeed seem more logical to combine these two pages, but I do not have a very strong opinion on the matter, keeping them seperate would also be fine with me.
 * , do you have any views on this? --Glennznl (talk) 09:45, 2 October 2023 (UTC)
 * I don't see a point in having 2 articles, assuming they're merged well. — kwami (talk) 22:17, 2 October 2023 (UTC)
 * If we can get a little more support for this I will perform the merge. --Glennznl (talk) 16:18, 3 October 2023 (UTC)
 * I'd like to see Austronesier's take. — kwami (talk) 23:27, 3 October 2023 (UTC)


 * your view on this might help to move the discussion on. Klbrain (talk) 21:46, 27 December 2023 (UTC)
 * I just performed the merge. If anybody suddenly decides to oppose it can be reverted of course. --Glennznl (talk) 14:35, 19 January 2024 (UTC)

Wiki Education assignment: Linguistics in the Digital Age
— Assignment last updated by Fedfed2 (talk) 00:53, 9 December 2023 (UTC)

duwaxa 94.200.93.38 (talk) 01:58, 27 October 2023 (UTC)


 * بيام داد ب شماره ٩٣٧٢٣٠٤١١٩ 2.147.225.229 (talk) 01:42, 2 November 2023 (UTC)

misleading?
the following line in the opening para is, i think, misleading: Linguists therefore describe the daughter languages within a language family as being genetically related.

i am not a linguist, and arrived here via recent reputable articles on Romeyka which said that modern greek has been considered an isolate (i.e. within IE). if both usages are current (‘genetically related’ meaning siblings from an immediate parent language, versus any relatives at all) i think it would be helpful to say that explicitly. similarly, if a historic usage is no longer preferred (which i think(?) seems to be the case), it would improve the article to say that clearly. 121.200.6.68 (talk) 13:48, 3 April 2024 (UTC)


 * I'm not 100% sure what you mean here. A language that is not an isolate may sometimes, I guess, be described as a relative isolate within its family if it's more distantly related to the other languages in the family than they are to each other. But nobody restricts "related" to mean only related to siblings from an "immediate" parent language, whatever that means; the nearness or remoteness of the ancestor has nothing to do with whether or not languages are related, and I don't think the sentence you quote implies that. AJD (talk) 15:47, 3 April 2024 (UTC)
 * Yes, I wasn't very clear. I think that people do indeed sometimes (mistakenly, it seems) describe a language as an isolate if it doesn't have "siblings", e.g. the recent article I mentioned on Romeyka Greek:
 * https://www.theguardian.com/science/2024/apr/03/endangered-greek-dialect-living-bridge-ancient-world-romeyka - 'As a result, Sitaridou has concluded that “Romeyka is a sister, rather than a daughter, of modern Greek”, a finding she says disrupts the claim that modern Greek is an “isolate” language, meaning it is unrelated to any other European language.'
 * This post on the modern Greek subreddit suggests that it is a common error: https://old.reddit.com/r/GREEK/comments/yonhbc/are_there_any_languages_mutually_intelligible_to/ivg59y3/
 * My impression from your reply is that this is considered incorrect, and that 'isolate' is not generally used this way by linguists. I had assumed that was the case, but I came here to confirm. When I found this page on Wikipedia, the sentence I mentioned made me uncertain. Re-quoting with the italics used in the article:
 * Linguists therefore describe the daughter languages within a language family as being genetically related.
 * I cannot see what the italic text is achieving other than implying that the 'daughter' language relationship specifically is what makes a relationship genetic. Removing the italics and perhaps clarifying that this is just one example might help, e.g.:
 * Linguists therefore describe the daughter languages (or parent languages, or child languages, etc.) within a language family as being genetically related.
 * But I have no expertise here and given that, was reluctant to alter the article. 121.200.6.68 (talk) 07:28, 4 April 2024 (UTC)
 * ...I don't know actually that people would usually use the term "genetically related" to describe the relationship between parent and child languages. Is Italian "related to" Latin? I think people would say it's "descended from" Latin. "Genetic relationship", as the term is generally used, refers to the relationship between daughter languages of the same parent language. Italian is related to Romanian, to Welsh, to Hindi, etc., but it would be weird to say it's "related to" Latin. AJD (talk) 07:38, 4 April 2024 (UTC)
 * Hmm fair enough. I think I would say that Italian is related to Latin, in the same way I would say that a child is related to their parents (i.e. it sounds slightly funny but is strictly accurate). In that case I would suggest that the first, defining sentence of the 'Language isolate' article may be the problem and source of my confusion:
 * A language isolate is a language that has no demonstrable genetic relationship with another language.
 * It seems that currently Wikipedia has a 'Language isolate' article that defines a language isolate as a language with no genetic relationships, and a 'Language family' article which implies that genetic relationships are sibling relationships, ultimately leading to the implication that language isolates are languages without siblings (even if they are known to be part of a broader language family).
 * If you are a linguist and editor, perhaps you could help there? 121.200.6.68 (talk) 08:04, 4 April 2024 (UTC)
 * No, you're still misunderstanding something. "Languages without siblings (even if they are known to be part of a broader language family)" is an incoherent concept; what I've been trying to say is being part of a language family is the same as having siblings. Maybe you're getting hung up on the idea that "sibling" implies a particularly close relationship? That's not how I've been using it in this discussion; I haven't been distinguishing between "sibling" and "cousin" relationships between languages. Italian and Romanian are "siblings" because they're both descended from Latin; Italian and Hindi are "siblings" because they're both descended from Proto–Indo-European.
 * The definition "Isolates have no known genetic relationships" is a good reason to exclude parent/child languages from the category of "genetic relationships". Basque is an isolate, and the fact that obviously it has a parent language that it's descended from (earlier forms of Basque) doesn't make that any less true. AJD (talk) 15:21, 4 April 2024 (UTC)