Talk:Uto-Aztecan languages

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Link Destinations?[edit]

Some of these links go to language pages and some of them go to pages on the peoples who spoke them. Should they all be directed to a language page at the expense of having more red links? Maybe following the language with the peoples who spoke them would be a good compromise. For example:

Ute-Southern Paiute language - Ute, Southern Paiute

What do you think? Toiyabe 23:52, 22 August 2005 (UTC)

I would like a link to something rather than nothing. Of course, links to language articles are prefered, but since most languages do not have articles I linked them to the people (which are also lacking). I guess I was hoping that when people linked the wrong page, they would be motivated to create a language page. So, I guess I like my way better than your suggestion, but I wont object if you change it. right now, I am just making language family pages, after that I will create pages for the individual languages. It is a slow process though: most authors write about other topics... peace – ishwar  (speak) 00:50, 2005 August 23 (UTC)
Fair enough. You seem to be putting in most of the effort in this area, so I'll defer to your lead. I guess I could be helpful and start some stubs, but this is an area I know little about. Toiyabe 16:22, 23 August 2005 (UTC)
I think we all agree that the links should direct to the language articles, unless they do not exist. Alexander 007 14:13, 8 December 2005 (UTC)

Tlalchiyahualica [edit]

Hopiakuta 22:40, 23 September 2006 (UTC)


I really must object to the Tanoan family not being present in this chart. Tanoan peoples are most certainly Uto-Aztecan. The Kiowa for example have known many other Uto-Aztecan peoples as their cousins, visited them, and travelled as far as central Mexico where they were able to converse with the people there. A single rogue linguist trying to make fame for himself by denying long established realities does not make a good reason to toy at such historical revisionism as removing Tanoan from the Uto-Aztecan family in an encyclopedia. Xj 07:12, 19 December 2006 (UTC)

There is no single rogue linguist. Tanoan has never been demonstrated to be related to Uto-Aztecan. This appears to be the consensus among the published family classifications. The hypothesis is old, however, and has been popularly repeated, probably due in large part to Edward Sapir. Although they may be related, this remains to worked out through linguistic reconstruction. —The preceding unsigned comment was added by Ish ishwar (talkcontribs) 08:36, 19 December 2006 (UTC).
Ish Ishwar is correct - The Aztec-Tanoan proposal is not established by a longshot, at this point it is just one more of Sapirs hunches. Whorf and Tragers establishment of Aztec-Tanoan was rebutted in 1979 by Kenneth Hale (hardly a rogue linguist) who wrote: "..a cautious view must lave the question open. If Uto-Aztecan and Kiowa-Tanoan are elated the time depth is extremely great". You shoulld read Campbells summay of the poposal in American Indian Languages (1997) pages 269-273. He concludes that the evidence hands falls short of establishing the likelihood of a relationship beteen the two families although further research is warranted. Further more as Ishwar says hardly any works on Uto-Aztecan historical linguistics have included Tanoan since the fifties. Maunus 09:38, 19 December 2006 (UTC)
Thirded. Sapir was a great linguist, but several of his hunches have not really withstood careful scrutiny. I would add that even if Aztec-Tanoan were agreed to be a correct grouping, Tanoan would still not be within Uto-Aztecan, but cognate within a higher grouping. · rodii · 00:50, 21 December 2006 (UTC)
Totally disagree with all this. Kiowas and other Tanonan peoples are known historically and factually to be closely related to a large number of Uto-Aztecan tribes. They were able to converse with one another each using their own languages. All the novel new theories the latest linguist are spewing to impress other linguists is stuff they don't even know what they are talking about. They are highly ignorant of tribal history and their claims that Tanonan is not in the Uto-Aztecan family is just a big white supremecist intellectual circle jerk outside of ethnic realities. Xj 01:30, 15 January 2007 (UTC)
You are completely wrong. The Tanoan langauges aren't even intelligible among themselves - Kiowas couldn't speak with Tiwas in their own language. And much less so with Uto-Aztecan peoples. However Most plains people could speak to unrelated tribes by using Plains Indian Sign Language. As for your accusing an entire science of racism that is just petty and ignorant. An article about the Uto-Aztecan language family must be based on linguistic data because it is only through linguistic research that the familyrelation ship has become known. Tribal ethnohistory while useful for something and certainly always interesting cannot be the basis for linguistic classification.Maunus 09:21, 15 January 2007 (UTC)
There is not one single, solitary linguist, who is a verifiable expert in either Uto-Aztecan or Tanoan languages, who accepts the relationship between Tanoan and Uto-Aztecan languages. It was based on a single article published in the 1930s. The evidence was problematic from the beginning and despite further efforts in subsequent decades, no strong evidence was ever found for a linguistic relationship. This unanimity is not "some rogue linguist". I am an expert in the Numic languages and I can't make heads nor tails of any Tanoan language. Laurel Watkins, the authority on Kiowa, would agree with this (although she can make sense of Tanoan). Kiowa and Comanche communication was either through bilinguals (the Kiowas usually learning Comanche) or they communicated through sign language. I would be interested in Xj's "evidence" and his credentials as a historical linguist. (Taivo (talk) 02:41, 5 February 2008 (UTC))

Genetic classification[edit]

Although the previous version said it was "based on the authoritative classification given in Lyle Campbells 'American Indian Languages'," in fact there were many discrepancies between Campbell's scheme and the one presented. I've tried to give a general consensus version based on Campbell, Goddard, and Mithun. If making changes in this, please cite your sources and give your reasons for rejecting the ones used here. RhymeNotStutter 22:25, 31 December 2006 (UTC)

I've changed some of your broken links. Southern Tepehuan is not extinct see the ethnologue references here: Tepehuán language. I don't think a "consensus" classification is a good idea since it borders on Original Research - we should faithfully reproduce one classification and only incorporate any later substantial changes to that classification giving specific references. Maunus 22:51, 31 December 2006 (UTC)

According to both Campbell (1997) and Mithun (1999), Southern Tepehuan is extinct. If there's reason to prefer the testimony of the Ethnologue reference, though, I don't think we should mechanically follow the first two on this point. The same reasoning applies to the matter of classification. Where there are contradictions between the authorities, it isn't doing "Original Research" to decide on the most credible between them on any particular point. Nor is it standard encyclopedic practice to uncritically follow one authority throughout; quite the contrary. If you prefer, though, we could switch over to presenting one authority's classification scheme (but which one?), and then note the discrepancies with other treatments in footnotes or text discussions. (I'm not sure it's worth the effort to do that, though, rather than trying to present a concise general picture that encompasses the general view in the article.) RhymeNotStutter 01:09, 1 January 2007 (UTC)

The problem is that Tepecano isn't the same thing as southern Tepehuán but southern tepehuán is sometimes called Tepecano. The original Tepecano language however is extinct. The Ethnologue counted over 15,000 speakers of southern Tepehuán in 2000 - I would prefer this info over Campbell et al. since counting speakers is what the ethnologue specializes in and campbell doesn't. As for the classification I think we should follow Campbell or maybe the ethnologues classification. And I do think it is OR to decide which authority is wrong on a specific point, such a decision cannot be made without arguments. This is why I find it better to reproduce "uncritically" a classification and let any errors in it be the errors of its original author and not ours.Maunus 06:15, 1 January 2007 (UTC)

I think there's a need for greater consistency here. You say that critically choosing among the contradictory testimony of experts on specific points is prohibited as Original Research, yet that's exactly what you're doing in the case of southern Tepehuán. You say that we should uncritically reproduce one classification, yet you've again inserted elements into the classification that are supported by NEITHER Campbell NOR Ethnologues (nor other sources that I've seen, and you don't cite any). As to Campbell's distinction between Southern Paiute and Ute/Chemehuevi, I strongly suspect that this is mostly a typo. I think that all other authorities recognize that Chemehuevi is a dialect of Southern Paiute. Some may make a distinction between Southern Paiute and Ute, but if so they would group Chemehuevi with Southern Paiute, not Ute. As to Alliklik, see the discussion in the articles on Tataviam and Tataviam language. It may well be that "Alliklik" is better reserved as a term for a Chumashan language or dialect, but the equation of Alliklik with Tataviam is well established in the literature, and an encyclopedia should alert its readers to that fact. RhymeNotStutter 16:31, 1 January 2007 (UTC)

Which unsourced statements unsupported by Campbell and the ethnologue have I inserted? Both Campbell and the ethnologue acknodledges southern tepehuan as being alive, and both say that Tepecano is a separate extinct language(and the ethnologue provides the additional information that southern and northern tepehúan is sometimes also called Tepecano).I am using only Campbells classification right now, but yes I am working it into the scheme already present without changing the entire layout at once. And this does amount to OR because I am deciding which discrepancies of your "consensus" classification I disagree with. But as we haven't agreed to follow one classification only yet at the moment I am just trying to make the classifcation presneted into something that I can agree with, and which doesn't present obvious arrors such as declaring a language with 20,000 speakers extinct.
Campbells classification of Southern Numic is:
Southern Paiute
Ute, Chemehuevi

Whether grouping Chemehuevi with Ute is a typo I don't know, but I know thats the way it is in the book and that I would want to see a good and referenced counter argument to it in order to change it. Laso if it is it is worth at least a note to say that we have assumed it to be a typo in Campbells classification. Why he distinguishes Ute and Paiute I don't know - even ethnologue who are notorious splitters don't distinguish them (The distinction goes back to Sapir though so maybe that's why he distinguishes). But again that's what he writes. (we could mention the conclusion of J.A. Jones article from Anthropological Quarterly 1954 here it is argued convincingly that "Ute-chemehuevi" is the best nomenclature for the linguistic group, and Ute/Souther Paiute as cultural distinctions)

His classification of Pimic is:
Pima Bajo
Northern Tepehuan and Southern Tepehuan 
Campbell does not say that southern Tepehuan is extinct, he says that tepecano is extinct as does the etnologue. But he counts southern and Northern Tepehuán as one language just as the present article did before you changed it.
As for Tataviam Campbell writes that he does consider Tataviam to be UA in accordance with Bright and in contradiction to Beeler and Klar, but that he considers Alliklik as identified by kroeber to be a Chumashan language in accordance with Beeler and Klar. He clearly distinguishes between Tataviam and the language called Alliklik by Kroeber. The current disageements should be enough to show that we should stick with one classifcation scheme because there is no "neutral consensus" we can't even agree on which languages are which. I reallly propose that we change the strategy and state specifically which classification says what and present arguments in favour and in contra of those differences. Allowing the readers to see which linguists say what and why instead of showing which classifications we personally believe the most. (I am assuming that you haven't personally done comparative research on Uto-Aztecan, I know I haven't and am only presenting Campbells data and arguments for face value)Maunus 17:47, 1 January 2007 (UTC)

Okay, I think the best would be to follow Campbell's classification explicitly and completely, with footnotes added to document the points on which other recent experts disagree with him. I'll work on changing over to the straight Campbell system in the next couple of days, unless you want to do so. (Places where you've introduced changes contrary to Campbell include Sonoran as a major grouping of all non-Aztecan southern languages, distinguishing Southeastern and Southwestern Tepehuan languages, and moving Tepecano into a subdivision of Southern Tepehuan.) For reasonable consistency, I don't think dialects should be listed; almost all of the major languages have documented dialects, and listing them all would overwhelm the classification. Where Campbell lists more than one language on a line without putting them in parentheses or hyphenating (as in the case Northern and Southern Tepehuan; also Cahuilla and Cupeño, which are certainly considered separate languages; etc.), I don't think he's saying that these are dialects, only that they belong to a common unnamed subgroup. - RhymeNotStutter 20:00, 1 January 2007 (UTC)

Agreed. Sonoran was already introduced in the classification before and I nearly moved it to the level where Campbell mentions it. He says that it is a grouping of Taracahitan, Pimic and Coracholan that isn't fully accepted - thats how I've introduced it. I did mean to list tepecano along side Southern tepehuan and not as a dialect of it. I also think we shouldn't list dialects. Please go ahead with the implementation of Campbell throuhgout and I'll assist where I can.Maunus 21:16, 1 January 2007 (UTC)

Nobody separates Southern Paiute from Chemehuevi and Ute as dialects of a single language. Lyle Campbell has told me that this is a typo in his book. It makes absolutely no linguistic sense whatsoever to group Ute (on the eastern end) with Chemehuevi (on the western end) as opposed to Southern Paiute (in the middle). I'm going to fix the Numic section to be accurate. It is not "original research", but simply recognizing the UNANIMITY of opinion among Numicists on this point. This is absolutely noncontroversial in either Numic or Uto-Aztecan studies. (Taivo (talk) 02:54, 5 February 2008 (UTC))


Nice map, Maunus, thanks. I don't see Guarijío or Pima on it. Lavintzin 16:44, 2 January 2007 (UTC)

I'll put in Guarijío (it wasn't on any on my source maps either for some reason) and Pima Bajo is in the northern map for some reason. Also the locations are slightly impressionistic since I don't have coordinates or anything I have just done it impressionistically from Campbells maps and some others. I hope they aren't too far off. Also it is a problem that some locations are precontact and some are current. I do think it is an OK additio for the time being but hopefully they can become more precise.Maunus 17:35, 2 January 2007 (UTC)

Brian Stubbs?[edit]

I know that Brian Stubbs has in the past couple decades done some work on Uto-Aztecan and Semitic languages, and found something like a thousand correspondences, which manifest consistent sound shifts. I'm a little suspicious of the conclusions because he's a Mormon (and so has reason to find such a correspondence). The reasearch has been mentioned favorably by other linguists, but they're also diffusionists. I don't know if any mainstream linguists or anthropologists have resonded to this research. Does anyone know? Should this at least be mentioned in this article? CaliforniaKid 19:45, 12 February 2007 (UTC)

This should not be mentioned since it is not taken seriously by other linguists. I have personally asked Brian about this and his opinion is that it's good enough for the Mormon faithful, but "not yet ready for primetime" and wider dissemination. He's asked me not to reference it elsewhere. (Taivo (talk) 03:04, 5 February 2008 (UTC))
We should include Stubb's other work though. Particularly his article on Tubar in the homage to Wick Miller is excellent and useful. IMO.

Yaqui on maps[edit]

i don't know who drew those maps, but i don't believe the yaqui were that far down into the state of sonora. if anything, they went as far north as the border with arizona today. was this drawn on someone's own recollection?

its drawn by me from the data from SIL on the current locations of Yaqui communities in Sonora.·Maunus· ·ƛ· 11:34, 22 June 2007 (UTC)

factual error in main article[edit]

The original text reads:

"Although established as a family, the subgrouping of the Uto-Aztecan language family remains controversial at present. Only eight groupings are considered unproblematic by a wide consensus of linguists: the Numic, Takic, Tübatulabal, Hopi, Pimic, Taracahitic and Aztecan branches. The higher level relations between these as well as the further subdivision of the single branches remain controversial. The Sonoran branch (including Pimic, Taracahitic and Corachol) and Shoshonean branch (including Numic and Takic,) in particular, are not accepted by some scholars."

Please note that in the list following this statement, there are only seven language groups named. —Preceding unsigned comment added by Dutch206 (talkcontribs) 00:28, 28 January 2008 (UTC)

Maps are way off[edit]

The U.S. map needs serious revision. The Numic area doesn't even reach the Great Salt Lake, but Shoshone extended into Wyoming and central Idaho and Northern Paiute extended into central Oregon. The U.S. map is a joke as it stands. (Taivo (talk) 02:48, 5 February 2008 (UTC))

I made the maps and I realize that the US map is flawed - that is because I had to impressionistically paint the areas unto the sattelite image with no state boundaries or other toponymic guideposts. I'll try to revise it.·Maunus· ·ƛ· 10:45, 5 February 2008 (UTC)
That is a better map. I'd be happy to proof another verson of the topo map if you prepare one. I'm not very skilled at the Photoshop genre of software or else I'd do it myself. (Taivo (talk) 03:27, 6 February 2008 (UTC))
I've put in a new map I made based on Campbell 1997 - I can alter almost anything if you have suggestions for improvement.·Maunus· ·ƛ· 11:50, 10 February 2008 (UTC)
Much better map. My better half is calling me to eat, so I'll make some notes later. (Taivo (talk) 11:58, 10 February 2008 (UTC))
The Tongva of the Los Angeles Basin spoke an Uto-Aztecan language but are not represented on the map.AThousandYoung (talk) 19:43, 6 May 2010 (UTC)

That part of the map is so small that all the names cannot be listed for lack of space. Thus, Cupeno, Gabrieleno, Serrano, and a number of Piman languages in northern Mexico, are all missing labels on the map. If you would like to draw a smaller scale map of just southern California and northwestern Mexico so that all the languages can be represented, then knock yourself out. (Taivo (talk) 19:55, 6 May 2010 (UTC))

Rebuilding the Classification[edit]

I've reformatted the Northern Uto-Aztecan section according to consensus Uto-Aztecan scholarship (summarized in Campbell, Mithun, and Goddard--Goddard's map that includes revisions to his Handbook article) and added references (a grammar and dictionary for each language where available). Someone else can work with Southern Uto-Aztecan because my Spanish is poor to non-existent.(Taivo (talk) 15:09, 9 February 2008 (UTC))

Good additions to the references, Maunus (Taivo (talk) 04:56, 10 February 2008 (UTC))


Greenberg's "Amerind" as well as his "mass comparison" methodology are rejected by the vast majority of historical linguists. There really isn't any debate on this at all. You have two serious historical linguists here (Maunus and myself) who agree on this. There is a small number of historical linguists who do accept this methodology and Greenberg's results, but they are a very small number and are not very influential in the field. (Taivo (talk) 04:53, 28 December 2008 (UTC))

Reference List[edit]

Actually, I like reference lists that are physically tied to the languages that they refer to. It makes it much easier to find references if one is looking for material on a single language or group of languages. (Taivo (talk) 19:34, 20 February 2009 (UTC))

TO-DO LIST[edit]

  • Add descriptions of histry of reconstruction:
Sapir (1915), Whorf (& Trager?)(1935), Voegelin, Voegelin & Hale 1962, Langacker (several), Kaufman (1981), Manaster-Ramer (Several), Dakin (Several).
  • Describe differences in phonological reconstructions.
Open or closed syllables, Final features, Lenition/consonant gradation, status of l, n, ŋ.
  • Describe basic morphological characteristics:
NI, reduplication, verb compounding, suppletion, postpositions, non-distinct arguments ala Langacker ...
  • Sound changes in different branches
  • Examples of cognate sets

·Maunus·ƛ· 21:24, 22 June 2009 (UTC)

"Campbell 1979"[edit]

Campbell 1979, which is an actual publication, was included among works cited (formerly "References"), yet it was not cited in the text in the last version prior to the bunch of edits I've made in the last 24 hours. Therefore, I've deleted it. (I retitled "References" "Bibliography"). Now, Campbell 1997 is cited many times, and it's possible that some of these citations are typos. (Of course, it's also possible that yet other inline citations were meant to be "Campbell 1979" or that any inline citation to it was inadvertently omitted.) By Googling, I find evidence that Campbell 1979 may have been cited without inline citations in many articles by one editor in particular. Let's see if someone else can match the source, Campbell 1979 to any content in this article. By the way, both Campbell 1979 and 1997 are relevant to this article. Dale Chock (talk) 22:10, 15 June 2011 (UTC)

Followup. Of course I have no objection to citing a source as further reading. Just make sure to put it under a "Further reading" heading! For it makes sense to have one bibliography section that is limited to publications that support claims in the body of the article. IF you philosophically disagree, then discuss it here. Don't just revert edits. Dale Chock (talk) 22:33, 15 June 2011 (UTC)

Hey, it occurred to me to create the "Further reading" section myself, which I've done. How come I didn't think of it the first time? Because being reverted in trigger happy fashion emotionally distracted me. Again, if somebody dislikes this solution, then certainly discuss it here; and if the reversion of my original edit can be justified by inline citations, then add the inline citations. Dale Chock (talk) 22:51, 15 June 2011 (UTC)
There was no "trigger happy" revert and when I reverted this section did not exist. Check the time stamps before you make accusations. Read WP:BRD. The status quo prevails if anyone reverts you, then you bring it here to discuss. Campbell 1979 is an important source for the history of the Uto-Aztecan question vis a vis the southern Uto-Aztecan branches. And since several of these sections were written before the ideological need to cite every sentence, we don't know what sources the original (and subsequent) authors used or didn't use in writing those sections. Campbell 1979 should remain with the other references until the entire article is referenced, and then you can see what is and is not referenced. It's rather pointless to count sources when some sections do not have their sources cited, especially when those sections are exactly those sections that Campbell 1979 is relevant to. --Taivo (talk) 01:01, 16 June 2011 (UTC)
(1) With regard to your observation that the section did not exist at the moment you reverted me, you are right, and it doesn't matter. What happened is that after I edited the article, I edited the Talk page to explain my article edit. I do not have to have the explanation ready the instant I edit the article. It was incumbent on you to wait a reasonable amount of time to see if I would follow through on my promise to make a comment at the Talk page. Indeed, my comment was short and I had it posted within a few minutes. In all my experience with Wikipedia, this is the first time I've been pounced on because my Talk page discussion wasn't immediately posted. Not only did you not wait a while, you didn't wait five minutes! That is being trigger happy.
(2) With regard to the substantial issue of the article content. Your reasoning is total contradiction: "we must keep the old citation because we don't know what passages it pertains to". Plus your action lacks an objective motivation: we don't break anything by removing a source not referenced to anything! We don't destroy evidence. You talk like you're new to Wikipedia and new to scholarship. The main purpose of footnotes is disclose what sources were consulted — evidently you disagree. To repeat myself from a previous post: if one thinks that unconsulted source X (i.e., X has not been invoked to support/justify any wording) is nevertheless an important source for topic Y, THAT is precisely the rationale for having a "Further reading" list — which is something professional encyclopedias do. "Further reading" can indicate sources which were NOT consulted, but that can be recommended as valuable for researching the topic. (Granted, "further reading" can also mean merely "reading beyond this article", so that an entry in the "further reading" may have been consulted.) For this article, when someone can use Campbell 1979 to support a specific passage in the Wikipedia article, they can add a footnote to that effect and consequently they can add it to the "Bibliography". If you still believe this is going to cause storm damage to the article, please spell out your reasoning. By the way, since I define the "bibliography" as works consulted, maybe we should retitle the bibliography "Works consulted". For now, I won't do so; I mention this by way of explanation. Dale Chock (talk) 15:41, 17 June 2011 (UTC)

High handed move by another editor[edit]

Just now, I made a minor deletion in the article, then created a new section on the Talk page to discuss it. Immediately Taivo reverted my edit with a nonexplanation explanation, "No need to delete. No new section on Talk page". You can't do that, Taivo. Besides, the new section in the Talk page is still there. I explained my reason for the deletion. I invited editors to provide inline citations to support the inclusion of the deleted material. All Taivo did was revert me without contributing any discussion. WHY is there "no need"? All this is about is a bibliography entry — why can't you explain yourself? Dale Chock (talk) 22:22, 15 June 2011 (UTC)

Why do we have a full set of bibliographic citations for further reading for each member language?[edit]

In the last two days, I was slow to realize what about the references was odd. What it is is this: the bulk of them are not sources of claims made in the article, but recommendations for further reading on particular languages. This is not a bad thing, but it is not usually how things are done at Wikipedia. If you want to study the grammar of language X mentioned in this article, what Wikipedia should do is hyperlink you to a separate article on language X, and in that article works on the grammar of language X should be cited. I grant that there is at least the possibility that there is some significant convenience in having a bibliography of grammars and linguistic researches in one place. Nevertheless, I am skeptical how much convenience it affords, and it's not how things are usually done. I am reminded of the motto that runs something like, "Wikipedia (or, 'a Wikipedia article') should not be a list of links". I am not going to scatter references in this category to the three dozen or so language articles, because of the workload. But I will create a subgroup of them in the bibliography. An alternative solution would be a table. Dale Chock (talk) 15:56, 17 June 2011 (UTC)

What is the problem? I am reminded of an old saying, "If it ain't broke, don't fix it". This article has been perfectly usable this way for a number of years (at least four or five). I disagree with your continued attempts to delete or "reorganize" references. This article does not suffer for having ample references. Unless you are a professional Uto-Aztecanist (or even a linguist), I wonder at your motivation for meddling in an article that you don't use or refer to. I have used these "extra" bibliographic references in language articles far more than I've used the text in many articles. Unless you have a valid reason for deleting or even rearranging these references, just saying, "other articles aren't like this" is not sufficient justification for trying to fix something that is not broken. --Taivo (talk) 21:31, 17 June 2011 (UTC)

Sources all need to be reread, article needs fresh start[edit]

By rewriting half of this small article today, and by adding comments to the above section, "Campbell 1979", I've made some startling discoveries. The article had two contradictions (like claiming a unified Corachol-Aztecan branch here, denying it there), and at least as many redundancies. It says very little. In my edits of the last six or so hours, I hardly deleted any points made, I just consolidated. Now I find that several statements, including ones I haven't gotten around to changing, are not faithful to the sources. This article was a mess! Basically, it has been neglected for a long time.

Here are two examples of things that remain to be fixed. (1) The statement that "since about 1980", some scholars have debunked the Northern/Southern classification. The problem with that is not that it's inaccurate, but that it's foolishly pointless in light of the claim by Steele 1979: 453, that the Northern/Southern theory was only launched in 1975, by Jeffrey Heath in a manuscript. (2) Somebody wrote that they were basing Northern/Southern on three sources including Mithun 1999. But Mithun 1999: 539-540 — I'm looking at it as I write this — does no such thing; it sticks with the eight branches.

I see that in my vehement comments of today, I was more right than I imagined. I mean the hullabaloo that was raised at my grouping and/or moving citations, and now it turns out that the article was sparse and garbled, and several editors hadn't read the sources, perhaps over a period of years. Neither had I, until today — but then, I only started editing this article yesterday! Dale Chock (talk) 22:52, 17 June 2011 (UTC)

As a small contribution, I've changed the infobox to throw out the useless links to Northern and Southern Uto-Aztecan (one being circular, the other red, both highly unlikely to ever be separate articles) and to reflect only the eight uncontroversial groups (they're not actually that many, so I can't see a problem with that). I knew the Northern/Southern classification was heavily problematic, but I had no idea that it's that recent and that baseless. I believe that it's always better for Wikipedia – especially when it comes to the infoboxes – to err on the side of caution, i. e., the splitters, than on the side of lumpers. --Florian Blaschke (talk) 12:20, 27 October 2011 (UTC)
  • In her review of Uto-Aztecan linguistics, Gabriela Caballero writes the following "However, despite the increasing number of descriptions of Uto-Aztecan varieties at the time, no consensus could be reached as to the division of the family into a Northern and a Southern branch, a subject that remains controversial until today: while Northern Uto-Aztecan (which includes the Numic, Tu¨batulabal, Takic and Hopi branches) has been generally recognized as a valid genetic unit (Heath 1978; Manaster Ramer 1992), there is still debate regarding the status of the southern languages (the Taracahitan, Tepiman, Tubar and Corachol- Aztecan branches) as an equivalently valid unit (Campbell and Langacker 1978; Cortina Borja and Valin˜ as Coalla 1989; Hill 2001a,b; Kaufman 1974a,b; Miller 1983a,b, 1984)." She also writes that "The question of determining the internal classification of the family has recently been approached through novel methodologies: Cortina Borja et al. (2002), building on Cortina Borja and Valin˜ as Coalla (1989), make a statistical and multivariate analysis of data from 19 Uto-Aztecan languages in order to determine their relationship in low-level and higher sub-groupings. Through comparison of both lexical and phonological data, they established similarity matrices between the languages in their sample and found evidence for a significant difference between Northern- and Southern-Uto-Aztecan languages, as well as the status of particular branches within both of these larger groupings (through measures of individuality and similarity of particular languages in particular sub-branches)." She also points to Haugen 2008 for a detailed review of the SOuth-North division debate.User:Maunus ·ʍaunus·snunɐw· 17:59, 29 May 2014 (UTC)

Taracahitic vs. Taracahitan[edit]

Mostly, the article uses the term 'Taracahitic', which seems to be used by the majority of scholars (including Campbell), but there are two references to 'Taracahitan', an alternative term for the subfamily that I've seen occasionally, e.g. in works by Wick Miller and Jane Hill. The article should be consistent in what term it uses. (I've made note of the alternate term in the separate page on Taracahitic; I'm not sure if alternate names need to be listed anywhere on this page.) Unfortunately, in the two places where 'Taracahitan' is used, it is in context of citing another source, and the original source might have used 'Taracahitan', so I'm reluctant to change it. (I don't have easy access to those sources.) What should be done? I see two options: a) change it to 'Taracahitic' to match the rest of the article or b) acknowledge that the original source used 'Taracahitan' (if indeed it did) and give an explanation in parentheses, such as '(i.e. Taracahitic)'. AlbertBickford (talk) 19:13, 12 November 2012 (UTC)

An example of Jane Hill's use of the term 'Taracahitan' is here. AlbertBickford (talk) 19:16, 12 November 2012 (UTC)

Historical phonology[edit]

Would it not be normal for an article about a proto-language to have a discussion of historical phonology? I came here to confirm / look into the rumour I have heard that initial *p becomes zero in Nahuatl nouns but is preserved in verbs, but I found nothing.Tibetologist (talk) 14:11, 30 November 2012 (UTC)


I've made this table meant to substitute the section that lists the UA languages in order. It still needs to incorporate the links to the languages, the demographic information and the sources. I want to link the sources in {{harvcoltxt|}} format. I also want to choose different background colors to show the divisions more clearly. I will be working on this slowly but steadily, and all help will be appreciated - including tweaking the subgroupings of the table, and to the table formatting. Just try not to screw up the table (not easy).

Genealogical classification of Uto-Aztecan languages
Family Groups Languages Where spoken and approximate number of speakers Works
Uto-Aztecan languages Northern Uto-Aztecan Numic Western Numic Paviotso, Bannock, Northern Paiute 700 speakers in California, Oregon, Idaho and Nevada
Mono About 40 speakers in California
Central Numic
Shoshoni, Goshiute 1000 fluent speakers and 1000 learners in Wyoming, Utah, Nevada, Idaho
Comanche 100 speakers in Oklahoma
Timbisha, Panamint 20 speakers in California and Nevada
Southern Numic Colorado River dialect chain: Ute, Southern Paiute, Chemehuevi 920 speakers of all dialects, in Colorado, Nevada, California, Utah, Arizona
Kawaiisu 5 speakers in California
Takic Serran Serrano, Kitanemuk(†) No native speakers currently, but learners of Serrano in Southern California
Cupan Cahuilla, Cupeño 35 speakers of Cahuilla, no native speakers of Cupeño
Luiseño-Juaneño 5 speakers in Southern California
Tongva (Gabrielino-Fernandeño)(†) (extinct since ca. 1900) Sta. Catalina Island, Los Angeles, Southern California.
Hopi Hopi 6,800 speakers in northeastern Arizona Hopi Dictionary Project (1998)
Tübatulabal Tübatulabal 5 speakers in Kern County, California Voegelin (1935)Voegelin (1958)
Southern Uto-Aztecan Tepiman
Pimic O'odham (Pima-Papago) 14,000 speakers in southern Arizona, US and northern Sonora, Mexico
Pima Bajo (O'ob No'ok) 650 speakers in Chihuahua and Sonora, Mexico
Tepehuan Northern Tepehuan 6,200 speakers in Chihuahua, Mexico
Southern Tepehuan 10,600 speakers in Southeastern Durango
Tepecano(†) Extinct since 1972, spoken in Northern Jalisco
Taracahitic Tarahumaran
Tarahumara (several varieties) 45,500 speakers of all varieties, all spoken in Chihuahua
Upriver Guarijio, Downriver Guarijio 2,840 speakers in Chihuahua and Sonora
Tubar(†) Spoken in Sinaloa and Sonora
Yaqui 11,800 in Sonora and Arizona
Mayo 33,000 in Sinaloa and Sonora
Opatan Opata(†) Exctinvt since approx. 1930. Spoken in Sonora.
Corachol Cora 13,600 speakers in northern Nayarit
Huichol 17,800 speakers in Nayarit and Jalisco
Aztecan Pochutec(†) extinct since 1970s, spoken on the coast of Oaxaca Template:Havcoltxt
Core Nahuan Pipil 20-40 speakers in El Salvador Campbell (1985)
Nahuatl 1,5 million speakers in Central Mexico
You're missing Comanche in Central Numic. And "Southern Paiute" is part of the Colorado River dialect chain. Oh, you probably meant "Comanche" instead of "Southern Paiute". I'll make that change. --Taivo (talk) 02:18, 31 May 2014 (UTC)
Thanks!User:Maunus ·ʍaunus·snunɐw· 02:24, 31 May 2014 (UTC)
Linked the languages. Easy if you manually bracket the names to be linked and then use the 'Treat search string as a regular expression' option in the search & replace tool at the upper right of the edit window:
            \[\[([^\]\|]+)\]\] → [[$1 language|$1]]
kwami (talk) 18:13, 1 June 2014 (UTC)
"Easy" is, of course, relative. My degree's not in computational linguistics :p --Taivo (talk) 20:07, 1 June 2014 (UTC)
Mine neither. This is just what's used for WP search coding, and a bit of the basics can make repetitive editing a lot easier. You can look it up in the AWB instructions, which is all I did. It's just hard to read because you have to opt out of symbols (like brackets) that are used as commands with a back slash, so each searched-for bracket needed to be written as \[ or \], but creating it isn't hard. The first part looks for [[(string which does not contain | or ])]] and replaces it with [[(that string) language|(that string)]]. Anything inside parentheses is interpreted as a string, and $1 means string number 1. So \[\[...\]\] searches for s.t. inside double brackets, the (...) records the string, and [^...]+ means a string of one or more (+) characters which are not (^) whatever is "...", in this case not a bracket \] or a pipe \|. So [[Pipil]] is converted to [[Pipil language|Pipil]]. I've probably used simple variants of that line a hundred times, whenever I want to link lists of languages. — kwami (talk) 06:09, 2 June 2014 (UTC)
Its not that useful though since many of the links created in that way go through redirects.User:Maunus ·ʍaunus·snunɐw· 17:05, 3 June 2014 (UTC)

Refs for individual languages[edit]

I have now included refs for all languages, I dont think it is necessary to include all of the sources below, but if some of them seem indispensable feel free to include those as well.


These refs all need to be formatted into the harvard template so that they can be cited in the table. This will take a while, unless someone has a smart way of doing it automatically. @Kwamikagami: perhaps?

  • Jean O. Charney. 1993. A Grammar of Comanche. Studies in the Anthropology of North American Indians. Lincoln, Nebraska: University of Nebraska Press.
  • Jon P. Dayley. 1989. Tümpisa (Panamint) Shoshone Grammar. University of California Publications in Linguistics Volume 115. Berkeley, California: University of California Press.
  • Jon P. Dayley. 1989. Tümpisa (Panamint) Shoshone Dictionary. University of California Publications in Linguistics Volume 116. Berkeley, California: University of California Press.
  • John E. McLaughlin. 2006. Timbisha (Panamint). Languages of the World/Materials 453. Muenchen: LINCOM Europa.
  • Richley H. Crapo. 1976. Big Smokey Valley Shoshoni. Desert Research Institute Publications in the Social Sciences 10. Reno: University of Nevada Press.
  • Beverly Crum & Jon Dayley. 1993. Western Shoshoni Grammar. Boise State University Occasional Papers and Monographs in Cultural Anthropology and Linguistics Volume No. 1. Boise, Idaho: Department of Anthropology, Boise State University.
  • Wick R. Miller. 1972. Newe Natekwinappeh: Shoshoni Stories and Dictionary. University of Utah Anthropological Papers 94. Salt Lake City: University of Utah Press.
  • Wick R. Miller. 1996. "Sketch of Shoshone, a Uto-Aztecan Language," Handbook of North American Indians, Volume 17, Languages. Ed. *Ives Goddard. Washington, D.C.: Smithsonian Institution. Pages 693–720.|group=A}}
  • Drusilla Gould & Christopher Loether. 2002. An Introduction to the Shoshoni Language: Dammen Daigwape. Salt Lake City, Utah: The University of Utah Press.
  • D.B. Shimkin. 1949. "Shoshone, I: Linguistic Sketch and Text," International Journal of American Linguistics 15:175–188.
  • D. B. Shimkin. 1949. "Shoshone II: Morpheme List," International Journal of American Linguistics 15.203–212.
  • Malinda Tidzump. 1970. Shoshone Thesaurus. Grand Forks, North Dakota.
  • Laird, Carobeth. 1976. The Chemehuevis. Malki Museum Press, Banning, California.
  • Edward Sapir. 1931. Southern Paiute Dictionary. Reprinted in 1992 in: The Collected Works of Edward Sapir, X, Southern Paiute and Ute Linguistics and Ethnography. Ed. William Bright. Berlin: Mouton deGruyter.
  • Pamela A. Bunte. 1979. "Problems in Southern Paiute Syntax and Semantics," Indiana University Ph.D. dissertation.|
  • Jean O. Charney. 1996. A Dictionary of the Southern Ute Language. Ignacio, Colorado: Ute Press.
  • Rosalie Bethel, Paul V. Kroskrity, Christopher Loether, & Gregory A. Reinhardt. 1993. A Dictionary of Western Mono. 2nd edition
  • Evan J. Norris. 1986. "A Grammar Sketch and Comparative Study of Eastern Mono," University of California, San Diego Ph.D. dissertation.)
  • Anonymous. 1987. Yerington Paiute Grammar. Anchorage, Alaska: Bilingual Education Services.
  • Arie Poldevaart. 1987. Paiute–English English–Paiute Dictionary. Yerington, Nevada: Yerington Paiute Tribe.*Allen Snapp, John Anderson, & Joy Anderson. 1982. "Northern Paiute," Studies in Uto-Aztecan Grammar, Volume 3, Uto-Aztecan Grammatical Sketches. Ed. Ronald W. Langacker. Summer Institute of Linguistics Publications in Linguistics Publication Number 57, Volume III. Dallas, Texas: Summer Institute of Linguistics and The University of Texas at Arlington. Pages 1–92.
  • Timothy John Thornes. 2003. "A Northern Paiute Grammar with Texts," University of Oregon Ph.D. dissertation.*Sven Liljeblad. 1966–1967. "Northern Paiute Lessons," manuscript.
  • Sven Liljeblad. 1950. "Bannack I: Phonemes," International Journal of American Linguistics 16:126–131|group=A}})
  • Alice J. Anderton. 1988. The Language of the Kitanemuks of California, University of California, Los Angeles, PhD dissertation.
  • Hansjakob Seiler and Kojiro Hioki. 1979. Cahuilla Dictionary. Banning, California: Malki Museum Press
  • Jane H. Hill & Rosinda Nolasquez. 1973. Mulu'wetam, the First People: Cupeno Oral History and Language. Banning, California: Malki Museum Press.
  • William Bright. 1968. A Luiseno Dictionary. University of California Publications in Linguistics 51. Berkeley: University of California Press.
  • Villiana Hyde. 1971. An Introduction to the Luiseño Language. Banning, California: Malki Museum Press.
  • Eric Bryant Elliott. 1999. "Dictionary of Rincon Luiseno," University of California, San Diego PhD dissertation.|group=A}}
  • Dean Saxton, Lucile Saxton, & Susie Enos. 1998. Dictionary: Tohono O'Odham/Pima to English, English to Tohono O'Odham/Pima. 2nd edition. Tucson, Arizona: University of Arizona Press. (Upper Piman)
  • Roberto Escalante H. & Zarina Estrada Fernandez. 1993. Textos y gramatica del pima bajo. Sonora: Departamento de Letra y Linguistica, Universidad de Mexico.
  • Donald H. Burgess. 1984. "Western Tarahumara," Studies in Uto-Aztecan grammar 4: Southern Uto-Aztecan grammatical sketches. Ed. Ronald W. Langacker. Summer Institute of Linguistics Publications in Linguistics 56. Dallas: Summer Institute of Linguistics and the University of Texas at Arlington. Pages 1–149..
  • Andrés Lionnet. 1978. Elementos de la lengua cahita (yaqui–mayo). México: Universidad Nacional Autónoma de México.|group=A}} (Yaqui
  • John M. Dedrick & Eugene H. Casad, ed. 1999. Yaqui Language Structures. Tucson: University of Arizona Press.
  • David L. Shaul. 1999. Yoeme–English English–Yoeme Standard Dictionary. New York: Hippocrene Books.
  • Howard Collard & E. Collard. 1962. Vocabulario Mayo. Vocabularios Indígenas No. 6. México: ILV.
  • Jeff Burnham. 1984. Una gramática de la Lengua Mayo. Hermosillo, Sonora: Universidad de Sonora.
  • Natal Lombardo. 1702. Arte de la Lengua Teguima vulgarmente llamada Opata. Mexico: Miguel de Ribera.
  • Andrés Lionnet. 1986. El eudeve, un idioma extinto de Sonora (Study based on materials of J. Johnson, Loaysa, Bartlett, and Smith). Mexico: Instituto de Investigaciones Antropologicas, Universidad Nacional Autonoma de Mexico.
  • David L. Shaul. 1991. "Eudeve morphosyntax: an overview," International Journal of American Linguistics 57:70–107.
  • Franz Boas. 1917. "El dialecto mexicano de Pochutla, Oaxaca," International Journal of American Linguistics 1:9–44.|group=A}}
  • Yolanda Lastra de Suárez. 1986. Las áreas dialectales del náhuatl moderno. Mexico: Instituto de Investigaciones Antropológicas, Universidad Nacional Autónoma de México.
  • Lyle Campbell. 1985. The Pipil language of El Salvador. Berlin: Mouton. Mouton grammar library; 1.
  • Fray Alonso de Molina. 1992 [1555]. Vocabulario en Lengua Castellana y Mexicana y Mexicana y Castellana. Mexico, D.F.: Porrúa.
  • Horacio Carochi. 1983 [1645]. Arte de la lengua mexicana: con la declaración de los adverbios della. Mexico, D.F.: Porrúa.

To do[edit]

We still need to include 1. overview of reconstructions of phonology and differences between them. 2. overview of reconstruction of grammar. 3. subsections on the 8 major language groups showing the main phonological and grammatical changes from PUA to the individual branches. 4. Expand the section on homeland to include the arguments and evidence used to argue for southern vs. northern homelands respectively. 5. expand the section on classification to give a better overview of the history of classification and the arguments and counter arguments used in subgrouping. 5. A section on macrofamily hypotheses including Aztec-Tanoan and Amerind.User:Maunus ·ʍaunus·snunɐw· 15:18, 6 June 2014 (UTC)

Glottolog classification[edit]

I notice that Glottolog has a completely different classification, unlike any I've seen before, with reference to Jane H. Hill and Victor Golla (both 2011), hence apparently taken or synthesised from these sources:

Omomil and Jova are extinct and unclassified within Uto-Aztecan, so their placement is less of a problem (apart from the acceptance of the Northern/Southern branches, which we would ignore for a consensus classification), but Takic and Taracahitic are broken up in this scheme into Cupan and Serrano–Gabrieliño (or rather, Tübatulabal is integrated into Takic at the same level of the two primary branches) or Cáhita, Opata-Eudeve, Tarahumaran and Tubar respectively.

So, which groups are really fully accepted by all scholars?

(I must admit I'm a bit wary about Jane H. Hill as a ref because her homeland proposal is not widely accepted in the field either, but I'm not an expert and her reasons may appear sound and her classification better received in the mainstream.) --Florian Blaschke (talk) 16:54, 17 January 2015 (UTC)

  • Her subgrouping has been gaining some traction lately, I also saw Haugen use it synthesised with Merrill's. I think at this point SUA is accepted as a unit at least. And it seems likely that Corachol and Aztecan can be grouped together as well. I think we should perhaps let it simmer a while and see how it is taken up.User:Maunus ·ʍaunus·snunɐw· 20:40, 17 January 2015 (UTC)
At the special Uto-Aztecan session at the LSA in Portland this month, there was general consensus (based a lot on Shaul 2014) amongst those who talked about it that the subfamilies are Numic, Takic, Tepiman, Tarahumaran, Opatan, Cahitan, Corachol, and Nahuan. (This, of course, excludes the single-language members). Takic is gradually being seen as not comprising a valid subgroup because there are no shared innovations--just shared borrowings. Takic will probably eventually be broken up into Cupan and Serran. There are then four single languages outside subfamilies: Tubatulabal, Hopi, Jova, and Tubar. Northern and Southern UA were not accepted or even used as labels in most of the papers. One paper specifically about Tubar pretty well demonstrated that it's alone. The key to the new UA consensus is that Taracahitic is no more and Takic is probably a fiction, although it is on life support as opposed to the dead Taracahitic. (That's my report from the "front lines". My own paper was on diffusion in the northern languages [Hopi, Tubatulabal, and Numic].) BTW, Hill's southern origin (and classificaiton) was generally ignored by the Uto-Aztecanists present at the meeting, so I'm really not sure where you are getting the "gaining traction" statement. --Taivo (talk) 21:07, 17 January 2015 (UTC)
Oh I didnt mean the southern origin hypothesis is gaining traction, I agree that one is dead in the water, but her recent 2011 classification based on shared innovations is pretty much the one you are describing above (with the disollution of Takic). I am surprised that noone took into account Merril and Hills data supporting SUA as a valid grouping. I was referring to this conference paper by Haugen which combines Hill and Merrills classifications.User:Maunus ·ʍaunus·snunɐw· 21:13, 17 January 2015 (UTC)

Haugen's 2014 classification based on Hill 2011 and Merrill 2013 looks like this:

I. Northern Uto-Aztecan

(Manaster Ramer 1992)

A. Numic
B. Californian
1. Serran
2. Gab-Cupan
3. Tübatulabal
C. Hopi
II. Southern Uto-Aztecan(Merrill 2013)
D. Tepiman
E. Cahitan
F. Opata-Eudeve
G. Tarahumara-Guarijío
H. Tubar
I. Corachol-Aztecan

Haugen showed off his lexicostatistics graphs at LSA, but when asked about the Tubatulabal-Hopi "connection" he didn't link them either together or to "Takic" and he didn't mention "Californian" at all. As far as the chart here in Wikipedia goes, there is solid consensus for breaking Taracahitan up. The consensus for breaking Takic up is almost there, but not quite as firm as for breaking down Taracahitan. There's no real consensus for "Californian" at this time. And it seems that scholars either like "Northern Uto-Aztecan" or they like "Southern Uto-Aztecan", but not both. The most cautious approach would be to remove both nodes until some sort of consensus develops. --Taivo (talk) 02:45, 18 January 2015 (UTC)

I think it is fine to maintain them as they are in the article with a note that it is controversial whether to consider them genetic or geographical. User:Maunus ·ʍaunus·snunɐw· 03:09, 18 January 2015 (UTC)
So that means only the following subgroups are uncontroversial genetic units:
  1. Numic
  2. Cupan
  3. Serran
  4. Hopi (isolate)
  5. Tübatulabal (isolate)
  6. Cahitan
  7. Opatan
  8. Tarahumaran
  9. Tepiman
  10. Tubar (isolate)
  11. Corachol
  12. Nahuan
with Jova and Omomil belonging to the poorly attested unclassified languages thought to be Uto-Aztecan.
Further subgrouping of the listed units is not settled.
Have I got that right?
If so, I believe the infobox and the "six subgroups are accepted as valid by all experts" part in the text should be corrected, and perhaps also the table in the section "Present scheme". --Florian Blaschke (talk) 15:20, 18 January 2015 (UTC)
Yes, Florian. That is the most uncontroversial listing right now. --Taivo (talk) 16:02, 18 January 2015 (UTC)
I am a little worried that we dont have published sources for this shift in classification yet, and that apparently the most recent publications have not been integrated into what seems to be the emergent consensus. That puts us in a bad situation regarding WP:V.User:Maunus ·ʍaunus·snunɐw· 16:15, 18 January 2015 (UTC)
Actually, Shaul (2014) has Florian's breakdown. While Shaul still lists Takic, the text following his list clearly points out that Takic probably doesn't exist other than as a borrowing zone rather than a genetic one. Since Merrill and Hill are in "conflict" over the existence of Northern UA and Southern UA, then leaving them out is verifiable in the sense that neither grouping enjoys a consensus of the linguistic community. WP:V is satisfied with a WP:RS. --Taivo (talk) 21:06, 18 January 2015 (UTC)
OK, I will get my hands on Shaul's book - I have been meaning to. User:Maunus ·ʍaunus·snunɐw· 01:00, 19 January 2015 (UTC)
It's got some quirks to it (such as inexplicably calling Western Numic "Northern Numic"), but overall it's a good thought-provoking piece of work. --Taivo (talk) 21:44, 20 January 2015 (UTC)
What I really liked is that it contains a good and balanced summary of the debates and evidence about UA prehistory, such a book has been missing for awhile.User:Maunus ·ʍaunus·snunɐw· 21:49, 20 January 2015 (UTC)
Thank you, Maunus, for the update, and for working on the article in general.
I note that if we take the new scheme on face value (i. e., Southern and Northern Uto-Aztecan are not valid branches – but areal groupings at best –, and all the listed units are primary branches), then (per centre-of-gravity-type considerations) it is even less plausible that the origin of Uto-Aztecan is in Mesoamerica (as per Hill). Instead, the branches cluster even much more decidedly in and around the traditional homeland region in western Aridoamerica, in the Sonoran (and western Chihuahuan) Desert. In fact, the centre of gravity is even clearer here than in most big families. --Florian Blaschke (talk) 14:00, 22 June 2015 (UTC)
Yep. that is true. I think Shaul makes some really excellent arguments on the homeland in his recent book.·maunus · snunɐɯ· 14:36, 22 June 2015 (UTC)
Actually, Shaul places the homeland farther north so that at least a part of it is in the Central Valley of California and extends somewhat into the southwestern Great Basin (he's not entirely clear on how far south or east it goes). --Taivo (talk) 18:04, 22 June 2015 (UTC)
Yes, what I meant is that he makes some good rebuttals to the southern placement. And some good arguments for the Tepiman corridor being the avenue of spread of SUA into Mexico.·maunus · snunɐɯ· 19:08, 22 June 2015 (UTC)
Between him and Merrill, I think that the Southern origin hypothesis has been thoroughly refuted. --Taivo (talk) 19:12, 22 June 2015 (UTC)
Pretty much. Hill would have to pull something exceptional out her sleeve.·maunus · snunɐɯ· 19:17, 22 June 2015 (UTC)
Taivo: That would be basically SoCal, perhaps the southern tip of Nevada (Las Vegas), a large part of Arizona, and some part of northern Sonora, right? So if SUA is a valid branch after all, a more northern homeland like this would be completely expected. If not, I'd have expected it to reach further south and perhaps not quite as far north. (Where might one place the Yuman homeland? Perhaps in Baja? Yuman languages are or were spoken right in the middle of the described area, so one would expect them to be intrusive.) Anyway, the general picture is clear enough. --Florian Blaschke (talk) 20:05, 22 June 2015 (UTC)
Shaul is vague about the southern boundary of his PUA homeland, but I suspect it doesn't reach to the Sonoran deserts, and probably ends in the Mojave desert. The Yuman homeland would be in the lower Colorado River/upper Baja area. Since Cochimi and Yuman are linked, it wouldn't be any more northerly than perhaps the middle of the California/Arizona border. If Shaul is correct in placing PUA north of that, the previous overlap between Proto-Yuman and Proto-Uto-Aztecan is eliminated. --Taivo (talk) 21:43, 22 June 2015 (UTC)
But isnt the time depth of Yuman a couple of thousand years shallower?·maunus · snunɐɯ· 21:50, 22 June 2015 (UTC)
Probably, but it depends on whether you factor in Cochimi or not. I'm not a Yumanist so I'm not readily conversant with the subtleties of how time and Cochimi factor into the equation. --Taivo (talk) 22:31, 22 June 2015 (UTC)
In any case I wouldn't be too worried about partially overlapping homelands - since that is indeed practically possible (especially for nomadic peoples), and since its not exactly an absolute science any way. You are of course right that Cochimi might suggest a homeland further south and closer to the coast. I have been looking a bit at Yuman lately and it would be interesting to explore the possibility of contact between the proto-languages - that would of course be the last piece of evidence needed to bury the southern hypothesis.·maunus · snunɐɯ· 07:41, 23 June 2015 (UTC)
────────────────────────────────────────────────────────────────────────────────────────────────────Excellent points all around. I agree that the time-depth of Yuman (3000 years?) appears to be considerably shallower than that of Uto-Aztecan (5000 years?), but Cochimí is the major uncertainty factor in both timing and placing of Proto-Yuman. I would not necessarily expect Proto-Yuman and Proto-UA in the strict sense to have been in contact, but perhaps Proto-Yuman (or perhaps Proto-Core-Yuman, if such existed? Or Proto-Pai, or Pre-Proto-Pai, etc.) with early UA and protolanguages of individual UA subbranches, or intermediate stages; one may compare how Proto-Uralic was not necessarily in direct contact with Proto-Indo-European, but – and this actually seems more likely, as Proto-Uralic appears to be younger – (Pre-)Proto-Indo-Iranian and an early form of northwestern Indo-European (which was not necessarily the direct ancestor of any attested Indo-European language). (The range of Indo-European most likely overlapped with Proto-Uralic as well, in view of the numerous early loans.) Any evidence of early contact like this with Uto-Aztecan (whether of Yuman, Chumashan, Yukian, Maiduan or whatever else) would be fascinating to know about. --Florian Blaschke (talk) 21:16, 23 June 2015 (UTC)

No discussion above about reducing Serran and Cupan, so I restored Luiseno and Tongva. If they don't belong there, they should be listed as branches of Uto-Aztecan. — kwami (talk) 23:29, 20 January 2015 (UTC)

(New) Cupan does, indeed, include Luiseno (at least). Tongva is up in the air about whether it is an independent branch of UA or part of Cupan. Best for now to leave it in Cupan. --Taivo (talk) 01:11, 21 January 2015 (UTC)

If Shaul places PUA north of the Mojave Desert and thus in the Great Basin along the California–Nevada border, that is a very different region from the homeland portrayed as traditionally advanced in the article; much farther to the northwest. Taivo, could you add Shaul's view of the homeland to the article? --Florian Blaschke (talk) 23:06, 28 December 2015 (UTC)

@TaivoLinguist: @Maunus: I forgot to ping you two. --Florian Blaschke (talk) 14:06, 29 December 2015 (UTC)
I think Shaul's view falls well within the range of 'traditional' locations of the UA homeland which go as far north as Death Valley, but sure we could mention it.·maunus · snunɐɯ· 18:39, 29 December 2015 (UTC)
Actually not exactly, Maunus. Shaul's proposal places PUA in the Central Valley of California (see page 188 of his book). While Shaul's view abuts the 'standard view', it's farther north. Fowler's "standard view" places PUA squarely within the Mohave/Sonoran desert regime, not as far north as the Central Valley complex. Nichols' homeland proposal (from which Shaul derives a core of his "California contact" evidence), places the homeland in the NW Great Basin, spreading into the entire Central Valley. --Taivo (talk) 18:53, 29 December 2015 (UTC)
Hmmm, I would have said that almost the entire SW region would make up the space where UA homelands have been "traditionally" located, although each location is of course much narrower. Where did Kroeber and Sapir and other early proposals locate the homeland? My lacking knowledge of California geograph may be at fault, I would have thought that the "central valley" also included the Inland Empire, the area around Palm Springs and South to what is today the Salton Sea.·maunus · snunɐɯ· 21:51, 30 December 2015 (UTC)
The Central Valley ends at the mountains just south of Bakersfield--I don't know if you've ever travelled in that part of the world, but there is a definite mountainous divide between the Central Valley and the northernmost reach of the Mohave desert complex south of the mountains in the Antelope Valley (Palmdale-Barstow). There is also a major climate boundary between them--the Central Valley is not part of the Mohave Desert as it is well-watered from both westward draining rivers from the Sierras and winter fog banks. The Central Valley has only been proposed by two researchers--Mike Nichols in the late '70s and David Shaul in 2012. At some point I'll write up a couple of sentences for the article, but it's my vacation and I'm not really in the mood for serious writing until after the new year. Florian, you'll just have to wait. --Taivo (talk) 00:53, 31 December 2015 (UTC)
No problem, it's not urgent. Thanks for the explanations, anyway. Happy New Year, everyone! --Florian Blaschke (talk) 02:27, 31 December 2015 (UTC)
Aha! The Central Valley! See, that's worth knowing – I couldn't figure that out on my own. (That would also indeed remove the possibility of – at least – immediate early contact with Yuman. I understand the issue now much better.) Since at the time of contact the Central Valley was inhabited by speakers of Yokutsan and other Penutian languages, Shaul clearly agrees with other researchers who consider the Penutian languages of California later arrivals from the north and east. It also makes more sense to me to put the homeland of a later huge and influential language family in a fertile area instead of a godforsaken wasteland, I have to say. It's just that purely on the basis of the centre-of-gravity consideration mentioned above I wouldn't have come up with the idea of placing the homeland this far to the north, but I understand Shaul's argument is based on ancient language contact from what you are writing. Sometimes it wouldn't hurt to be a bit more explicit instead of letting me try to puzzle the picture together on my own. :-) --Florian Blaschke (talk) 21:34, 30 December 2015 (UTC)
By the way, does anybody of you happen to know for certain if frequent commenter on Language Hat and Language Log, marie-lucie, is Marie-Lucie Tarpent? (Even though it seems obvious ...) I started to wonder because a statement by marie-lucie is used as a source on Penutian languages. Not that I have any reason to doubt its accuracy. It would be nice to know the full citation, though, not just "(Jane Hill 2002)" ... --Florian Blaschke (talk) 21:42, 30 December 2015 (UTC)
I would think that the wiki user MarieLucie is also Tarpent. You can ask her, hard to see why she would deny it if she is.·maunus · snunɐɯ· 21:51, 30 December 2015 (UTC)
Good idea. The wiki user is inactive but I'll just mail her. --Florian Blaschke (talk) 02:27, 31 December 2015 (UTC)