Language desk
< January 21	<< Dec \| January \| Feb >>	Current desk >

Welcome to the Wikipedia Language Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

January 22

Monogenesis hypothesis

According to Proto-human language, it didn't gain much traction in academia. But, if I understand correctly, initial speakers of every language family must have spoken the language of their predecessors to develop a new language. With that in mind, correlation with early human migration routes must produce a chronological chain of emergence of every language family and language isolates from another ultimately leading to the proto-human language, probably somewhere in northern Africa (similar to evolutionary universal common ancestor). Does such reasoning validate proto-human language? 212.180.235.46 (talk) 20:05, 22 January 2024 (UTC)[reply]

No, it doesn't, for two reasons. First, that assumes that language developed before humans spread, but that is not necessarily the case. But the stronger argument is that signed languages do not necessarily follow the pattern you describe - in fact, we have solid recent evidence of the genesis of at least one such: Nicaraguan Sign Language. ColinFine (talk) 21:46, 22 January 2024 (UTC)[reply]

The origins of some spoken creole languages are almost as extreme as the origin of Nicaraguan Sign Language... AnonMoos (talk) 23:27, 22 January 2024 (UTC)[reply]

Interestingly, pidgins and creole languages have had their own monogenetic theory, now generally discarded by linguists. --Lambiam 10:14, 23 January 2024 (UTC)[reply]

The Saramaccan language is known to some linguists as a language which has zero mutual comprehensibility with any of its source languages, and where it's often difficult to even trace the origins of words unless you're a specialized linguist (as opposed to many other creoles, such as Haitian Creole, or Tok Pisin, where no matter how limited the mutual comprehensibility with the source language, it's easy to guess the origins of many words). AnonMoos (talk) 23:50, 24 January 2024 (UTC)[reply]

Language almost certainly predates modern Homo sapiens by several rungs on our evolutionary ladder, based on the evolution of breathing and mouth structures. (by some definition of speech this was always believed to some extent, but to a sapiens level of articulation the review is newer: Boe et al 2019; lay writeup in [The Atlantic 2019-12-12.) The relative complexity of such language is debated, as there is little evidence of symbolic understandinding or spiritual reflection in non-sapiens hominids. BMC Q&A 2017

I agree however that there's nothing to be found through current historical linguistics techniques beyond an order of 10,000 years. Save for discovering new archaeology, like for example if that idea of extracting recorded audio from pottery engraving had been fruitful. SamuelRiv (talk) 20:18, 23 January 2024 (UTC)[reply]

SamuelRiv -- Some have interpreted archaeological behavioral modernity as the final transition to fully-modern human language from something that wasn't yet fully-modern human language (though doubtless still far more complex than anything chimpanzees can manage). AnonMoos (talk) 23:50, 24 January 2024 (UTC)[reply]

I'm not in the field, so my opinion's worthless, but I'm more and more inclined to hypothesize that once hominins had finished selecting the hardware for complex speech, hominin language would be in many ways virtually indistinguishable in complexity from today -- even if they're/we're not yet inclined to make art and bury the dead. A lot is unanswerable, at least until you grow yourself some clone of a hominin, or create a plausible model of non-complex non-symbolic human-like language. One could also build a convincing case (with a great deal more neurosci and neuropsych research backing it) that symbolic and abstract expression in art, music, planning, etc is connected to our fundamental language mechanisms. SamuelRiv (talk) 04:05, 27 January 2024 (UTC)[reply]

212.180.235.46 -- Linguistic reconstruction produces useful results only to about 10,000 years ago at most, so your assumptions can't be factually cross-checked. AnonMoos (talk) 23:27, 22 January 2024 (UTC)[reply]

To test the hypothesis that two languages are related need not involve an intermediate step of reconstructing a common ancestor. One can craft a sophisticated metric for measuring the similarity of two languages. Applying it to a wide variety of pairs of languages that are not known or strongly suspected to be related will result in a probability distribution. Two reconstructed proto-languages, say Proto-Afroasiatic and Proto-Indo-European, could then (conceivably) turn out to have a greater similarity than can reasonably be ascribed to coincidence, while the footprint of any shared ancestry is too thinly spread out over the languages to aid in further reconstruction. --Lambiam 09:50, 24 January 2024 (UTC)[reply]

I was including that in the 10,000 year round number (which some linguists would consider excessively generous even so). The Amerind languages as posited by Joseph Greenberg have not achieved scholarly respectability as a historically based grouping, as also the proposals of Sergei Starostin... AnonMoos (talk) 23:54, 24 January 2024 (UTC)[reply]

P.S. Proto-Indo-European and Proto-Afroasiatic were not geographically adjacent, while Proto-Afroasiatic (as opposed to Proto-Semitic) is not very securely reconstructible, which are big problems right off. Proto-Indo-European and Proto-Uralic may have been adjacent. AnonMoos (talk) 23:57, 24 January 2024 (UTC)[reply]

My main point was that showing statistically significant similarity need not require reconstruction. Here is a trivial example. The following seemingly random strings cannot have been generated independently; there are more commonalities than can be ascribed to coincidence. Yet it is impossible to make a reasonable guess at a common source.

OJHFEUOTIYZYVFOFUVNZGCZRYMLASBMVESKMBIDZKVVRLCECGSRUUOCCFCOCXGCAJ

OGTDINSCBJJALKEUZKKBGAURAOJYSFMWPFZMHEBQGVVRDWERYJQQUVFKOHUFBSCPD

^ ^ ^ ^ ^ ^ ^^^ ^ ^ ^

There also does not have to be a generally accepted reconstruction for the language families to be compared; one can (theoretically) develop similarity measures between whole families. The cradles need not have been geographically close if the founders of the proto-languages could have trekked the distance from a common ancestor site in a limited number of generations. Not being geographically close is in fact an advantage; otherwise similarities may be ascribed to contamination by contact; cf. Sprachbund. --Lambiam 17:24, 25 January 2024 (UTC)[reply]

I should have said Proto-Finno-Ugric, not Proto-Uralic. As for your method, that's basically known in linguistic circles as mass comparison... AnonMoos (talk) 00:26, 26 January 2024 (UTC)[reply]

To expand, if you're not doing any reconstructing, then supposed "comparing for similarity" can be problematic in many cases. English "wheel" and Sanskrit "chakra" don't have one single sound in common, yet are historically related, while Modern Greek "mati" and Malay "mata" (both meaning eye) are not historically related. AnonMoos (talk) 20:13, 30 January 2024 (UTC)[reply]

What does M/S mean in front of Indian company names?

I was reading through some Indian railway literature and noticed that almost every time a company name is mentioned, it is preceded by M/S, i.e. "Three years ago, new electric equipment made by M/S Acme India Ltd. was installed." What does the M/S stand for? --188.23.206.25 (talk) 20:47, 22 January 2024 (UTC)[reply]

According to M/S: Messrs., especially in India as a prefix to a firm or company name. Your example doesn't really capture that. --Wrongfilter (talk) 20:52, 22 January 2024 (UTC)[reply]

Thanks, didn't see that article since I typed it as m/s which redirects to a different article. Interestingly, it seems to be used also used for large corporations while for me Messrs. would sound more like a mom and pop business. --188.23.206.25 (talk) 21:03, 22 January 2024 (UTC)[reply]

Only if mom and pop were both misters. It's rather obsolete in Britain, with the exception of legal firms. Alansplodge (talk) 22:51, 22 January 2024 (UTC)[reply]

It might depend on the surroundings, but I guess a pop and pop business might pop. 惑乱 Wakuran (talk) 11:16, 23 January 2024 (UTC)[reply]