Talk:Yeyi language

highly dubious phonology tables

A singular source is provided for all but the clicks in the phonology section, that being Seidel (2008). Transcriptions and descriptions of sounds such as a prenasalized voiceless velar ejective [ᵑkʼ] are otherwise unprecedented and particularly dubious; the PHOIBLE page for the Yeyi inventory highly conflicts with the information provided here (this is not to say that one is more correct than the other, but simply their major lack of concordance is concerning), and includes no prenasalized consonants.

Moreover, Lusekelo (2009)'s review of Seidel (2008) specifically mentions:

Seidel presents the 5-vowel system available in the language, vowel nasalization, imbrication and phonological processes, all well described and exemplified. He goes on to challenge existing literature as far as vowels are concerned. Turning to consonants, his inventory has 37 non-pre-nasalized eggressive consonants, 26 pre-nasalized ones, and Table 4 has 14 ingressive consonants i.e. clicks (p. 41). His analysis, as he explicitly states, stems from gathered data. Having read his inventory, I get the feeling that establishing the phoneme inventory for Yeyi is not an easy task as inventories of previous works somehow differ from Seidel's because for example one claims that synchronically Yeyi phonemes have some 45 consonants, leaving behind clicks (Gowlett 1997: 253, 260). Another one, Baumbach (1997: 417-421), has a total of 80 consonants, which he supplies with phonemes, alphabetic symbols as well as illustrations.

Therefore, the current inventories listed in the phonology section are staunchly WP:UNDUE; even if well detailed, they still solely rely on one author's interpretations; and not a particularly notable author either. I would argue, especially given Lusekelo (2009)'s comments, that the current tables fall under an extremely small minority interpretation lacking prominent adherents, and should not be on WP at all. ~ oklopfer (💬) 20:10, 21 October 2025 (UTC)

@Oklopfer I don’t know who cited the dreaded PHOIBLE source, but I can tell you for sure, that virtually all the the phonological information is cited straight from Seidel (2008). When I added the info here, I did not use PHOIBLE. Whoever did, needs to revert it. Fdom5997 (talk) 22:46, 21 October 2025 (UTC)

That is exactly why I find it dubious - that it is only cited straight from Seidel (2008). PHOIBLE clearly states that its inventory is from Baumbach (1997), which is directly referenced as being conflicting with Seidel in Lusekelo (2009). To use only Seidel's interpretations over the others is WP:UNDUE by definition. ~ oklopfer (💬) 22:51, 21 October 2025 (UTC)

PHOIBLE is not a good source. And a lot if the sources that the site uses are easily disputed. I would trust a written source done by an actual linguist over some site that finds whatever source is convenient for them. Fdom5997 (talk) 22:58, 21 October 2025 (UTC)

Other points aside, I wholeheartedly agree with this statement. ~ oklopfer (💬) 23:10, 21 October 2025 (UTC)

@Oklopfer and to answer your question about Seidel’s source, we should leave the info there the way it is. If there is no other source available, then that is what we use. Fdom5997 (talk) 22:49, 21 October 2025 (UTC)

@Fdom5997 Baumbach (1997) is available, at least in hard copy: https://www.namibiana.de/en/namibian-languages-reports-and-papers.html

I have to look a bit more if a digital version exists.

And it's not just PHOIBLE who cites Baumbach, Lusekelo (2009) even provides page numbers in the quote I gave above. ~ oklopfer (💬) 23:01, 21 October 2025 (UTC)

Notably, Lusekelo provides a different count of consonants than PHOIBLE, suggesting he did use the original source. ~ oklopfer (💬) 23:08, 21 October 2025 (UTC)

Look, of course there is going to be critics of each others’ sources. Just because the info has only one source does not mean it’s wrong or dubious. And the fact of the matter is, is that PHOIBLE’s sources used are much more conflicted than any individual source. Lusekelo’s source is just a review. Fdom5997 (talk) 23:09, 21 October 2025 (UTC)

@Fdom5997 Yes, Lusekelo (2009) is indeed a review, which includes citations to other sources, specifically Baumbach (1997). Please stop with WP:IDHT. I think it would be fine if the section was relabeled along the lines of "Sources have provided various descriptions of the inventory, of which the most descriptive one comes from Seidel (2008), shown below…"; again, if we only acknowledge the existence of one inventory listing when other conflicting ones do exist in reasonable literature, that is again, staunchly WP:UNDUE. We should at least note that the provided inventory only comes from this source, and that even though there are others that may conflict, they are less descriptive and less modern.

To your comment:

Just because the info has only one source does not mean it’s wrong or dubious.

Per WP:SCHOLARSHIP, it very well can. For reference, Seidel (2008) has 6 cites, Lusekelo (2009) has 16 cites. Seidel (2008) is the least cited of them, strongly making it the minority opinion. WP:UNDUE. ~ oklopfer (💬) 23:19, 21 October 2025 (UTC)

It is hard to get a count of Baumbach (1997)'s citations, as it is included with the collection it was published with, but Gowlett (1997), also with 16 cites, provides yet another phonology table (p. 260), and the alphabet summary published by SIL suggests yet a fourth. ~ oklopfer (💬) 00:35, 22 October 2025 (UTC)

To be honest, I do agree with this solution of yours:

I think it would be fine if the section was relabeled along the lines of "Sources have provided various descriptions of the inventory, of which the most descriptive one comes from Seidel (2008), shown below…"... We should at least note that the provided inventory only comes from this source, and that even though there are others that may conflict, they are less descriptive and less modern.

As we can state how this is the most descriptive and thorough source out there despite any other conflicting ones. Definitely better than labeling the section with a "disputed" note. Fdom5997 (talk) 05:43, 22 October 2025 (UTC)

I'm happy with a change like that, so long as it recognizes there is some divergence from previous literature. My purpose for keeping the disputed note in place was solely for there to be discussion held about this on the talk page and clarity to be added to the page afterward, not as a badge of shame.

As for why the number of cites would matter, it is one of many measures of a source's reliability and notability/reputability, as suggested in WP:SCHOLARSHIP ('Citation counts' bullet). ~ oklopfer (💬) 06:34, 22 October 2025 (UTC)

Well nonetheless, I find that pretty weird how Siedel’s source is surely the most descriptive, but yet cited least. Really shouldn’t be this way. Fdom5997 (talk) 18:17, 22 October 2025 (UTC)

Just because it is the most descriptive does not necessarily mean it is the most reputable. Anyone (in theory) has the ability to write and publish on the topic; their reliability is not solely determined by the quantity of what they've written, but also by how their publishing has been received by the broader academic community.

Even if it is highly descriptive and appears on the surface to be of high quality, but is still poorly cited by other sources, then it is likely that the broader community has doubts of its credibility and/or reproducibility; particularly if each source that has documented the phonology conflicts with the previous, and especially if it is proclaiming the attestation of sounds that have never before been documented in any other scholarly setting, and likewise do not appear to be documented in any since its publishing; therefore, they simply are not received with undue prominence or weight, which is to be expected.

It looks like Seidel has a few other papers surrounding Yeyi's classification (2005, 2007, 2015) which receive a handful more citations, though only (2008) actually mentions its phonology. 2005 has 10 cites, 2007 appears to have just 2 cites, and 2015 has 21 cites; so it is evident that Seidel's classification of the phonology is not taken as seriously as some of his other publishings.

There is a very good possibility that the descriptions given by Seidel (2008) are an overgeneralization of the inventory from a select handful of speakers, and do not accurately represent the true inventory of the language broadly; there is also a solid possibility that Seidel's lines drawn for bracketing phone(me)s are rather arbitrary, and another author analyzing the same content may find a completely different inventory (again, this seems especially likely given that many listed are otherwise unheard of sounds).

I am not necessarily claiming WP:FRINGE applies to this situation, but it feels to me like a highly relevant essay regarding all of this, and gives a lot of good points to think about. Finally, I do want to note that it appears the vast majority of attempts to transcribe Yeyi with an orthography of any kind, phonetic or otherwise, have been done for missionary purposes and Bible translation; a la SIL Global#Criticism, take of that what you will. ~ oklopfer (💬) 20:34, 22 October 2025 (UTC)

Yes but are you saying that Seidel (2008) is not that reputable? I am very certain that it is. Why should there be such doubts of it if it was so thoroughly done? And why would any other previous sources published before Seidel’s source be any better?

Furthermore, can we stick to the solution that you made earlier, that I thought was good?

Like stating how this is the most descriptive and thorough source out there despite any other conflicting ones? Fdom5997 (talk) 21:56, 22 October 2025 (UTC)

What gives you the certainty of its reputability? WP:UNDUE explicitly states: "Keep in mind that, in determining proper weight, we consider a viewpoint's prevalence in reliable sources, not its prevalence among Wikipedia editors or the general public." And you are literally responding to my massive explanation as to why there would be doubts even if it was 'thoroughly done', because that doesn't really stand for anything on its own.

My point at the end of my comment was referring to the fact that there are potentially no reputable sources documenting Yeyi's phonology at all; I am not saying any others published before Seidel's are better, but that Seidel's description is incredibly unorthodox and remains the minority opinion. In my personal opinion, I don't think any consonant tables should be listed, as there is no concordance of the phonological inventory among any two studies regarding Yeyi. While there are many consonants that are in common in the different inventories, they are not in corroboration of each other, simply in parallel; we could attempt to synthesize the different sources into a 'commonly agreed upon' table, but that would be blatant violation of WP:PRIMARY policy.

That being said, having no tables at all is not helpful to a reader either, who then has to go and dig up the various studies to compare and make a determination of their own. If all of the sources conflict with each other, then while it is preferable not to give undue weight to any of them, it is more helpful to use the most descriptive one if a table must be shown. That does not automatically make it reputable, though, as I explained above; that simply makes it our best option for usage here.

So yes, I continue to support my previous suggestion, as long as we are contextualizing for the reader the weight of the source being used, and recognizing it is one of many. It would additionally be helpful to include citations to the conflicting ones in the note being added (i.e. "previous sources, such as Baumbach (1997) and Gowlett (1997),"). ~ oklopfer (💬) 23:38, 22 October 2025 (UTC)

Ok so for that matter, can we just get rid of the "dubious" template, and replace it with that statement?

Also, I am sorry, but you thinking that there should not be any phonology tables, is just incredibly ridiculous. I strongly believe that we must keep the phonology tables, regardless if there are conflicting sources. I firmly believe that Seidel (2008) is at least a decently reputable source to use. Maybe not "the most" reputable, but at least acceptable. It is not "unorthodoxed" and it is surely the best to find out there. Seidel is a licensed anthropologist and linguist with a PhD and many other publications in these same fields. Until there are actually better sources published on Yeyi's phonology, we need to stop being so pedantic. Fdom5997 (talk) 00:14, 23 October 2025 (UTC)

WP:IDHT again? Per my previous comment:

What gives you the certainty of its reputability? WP:UNDUE explicitly states: "Keep in mind that, in determining proper weight, we consider a viewpoint's prevalence in reliable sources, not its prevalence among Wikipedia editors or the general public."

And yes, it is definitionally unorthodox, when it is introducing sounds documented nowhere else, before or after. Nearly all of the sounds listed in the current "Prenasal consonants" table cannot be found in any credible literature outside of Seidel (2008). Stop accusing me of being pedantic by bringing up clear violation of WP:COPO, that is incredibly bad faith. ~ oklopfer (💬) 00:31, 23 October 2025 (UTC)

Oh please. I just answered your question. As if being done by an actual linguist withing the field of linguistics is "not" a reputable source. No, you are being incredibly pedantic, and obsessively citing rules verbatim and acting as if this source that is used, is pure misinformation or done unprofessionally.

So what if there are no prenasal sounds in any of the other sources? Have you even looked at the other sources yourself? If not, then who are you to determine any of this?

If you can't find any other sources that are better than this one, then we need to at least come to the conclusion that Seidel's source is still the best source out there for this material, as of this moment.

Also, just to add to your point on the citation count WP:SCHOLARSHIP also claims at the end that:

The number of citations may be misleading if an author cites themselves often. Fdom5997 (talk) 00:42, 23 October 2025 (UTC)

I have indeed looked at the other sources myself, I hope you have too. You did not answer my question, you gave his basic credentials, not any evidence of the specific source being reputable. Your argument from authority is again, bad faith. I am citing the rules verbatim because you are suggesting violation of them, verbatim. Also, per my previous comment:

That being said, having no tables at all is not helpful to a reader either, who then has to go and dig up the various studies to compare and make a determination of their own. If all of the sources conflict with each other, then while it is preferable not to give undue weight to any of them, it is more helpful to use the most descriptive one if a table must be shown. That does not automatically make it reputable, though, as I explained above; that simply makes it our best option for usage here.

I recognize your stance in wanting the tables maintained, and I am trying to work to compromise. But we must be truthful about its actual academic standing and weight, per Wikipedia's defining policies. If this is going to keep going in circles and you are going to continue addressing my arguments in bad faith, I am happy to bring it up on WP:RSN or WP:NORN.

I am confused how your point of authors citing themselves being potentially misleading works in your favor, as that would then imply that the mere 6 that Seidel (2008) currently has could well be even less, further diminishing its reputability. We also can simply check exactly which sources the articles are referenced in, though, as those are listed in all of the links I gave above. So I am also not really sure how that is relevant. ~ oklopfer (💬) 00:56, 23 October 2025 (UTC)

Alright, well at least you do agree and are open to keeping the tables and all the phonological info. And I am also open to adding more info from any of the other sources you mentioned. As I have looked at Gowlett, but did not look at Baumbach's source, if you have it available, I would like to possibly see it myself. Fdom5997 (talk) 01:14, 23 October 2025 (UTC)

I have not yet found a digital copy of Baumbach (1997), though a purchasable PDF did exist up until last year, when the founder of the publishing group passed away; so I presume it is still out there. I may also see what libraries have physical copies. So while I cannot comment on what exactly it contains, Lusekelo (2009) and PHOIBLE's documentation both directly cite Baumbach (1997); even ignoring PHOIBLE, Lusekelo (2009) is enough to confirm for us that Baumbach (1997) is a source which contains phonological tables conflicting with Seidel (2008), and that really is all that needs to be noted of Baumbach.

As for the other source I provided earlier, SIL's alphabet summary, would also be likely be considered both a notable and conflicting one. ~ oklopfer (💬) 01:30, 23 October 2025 (UTC)

Ok, but are you still open to keeping the tables and phonological info? And do you still agree that Seidel is still probably the best and most descriptive source out of the two? Fdom5997 (talk) 02:34, 23 October 2025 (UTC)

Yes, I agree with both of those statements. Even if it is not necessarily the most prominent or widely accepted, it is by far the most descriptive. But it ought to be noted that it is still an outlier, not the standard, and that these others do exist. ~ oklopfer (💬) 02:47, 23 October 2025 (UTC)

Well then great, and surely agreed. But let's not kid ourselves and think that only one of them are the "right" source.

Also, after reading the SIL page, and then rereading Seidel's source, both of them do agree that there are prenasalized sounds (however, as to how many, SIL's may be dubious). But then after reading the PHOIBLE page that cited Baumbach, the only real missing consonants were the labialized ones as to which Seidel states that "nearly all of the phonemes above except a few have been found to form a Cʷ cluster". So they do exist, however; it seems as if their phonemic status is uncertain so they are not quite listed as phonemic symbols. Fdom5997 (talk) 03:10, 23 October 2025 (UTC)

Why should the amount of cites matter? Seidel (2008) is still the most descriptive source regarding the phonology of the language. And even more than Gowlett's source too. Fdom5997 (talk) 05:57, 22 October 2025 (UTC)

Also, please see Template:Dubious and Template:Disputed section, as it seems you don't understand what they're for. I understand there is a provided source, that's exactly why I am using them. Please do not remove the tag until we come to a consensus/resolution. Until then, it is by definition disputed. ~ oklopfer (💬) 23:26, 21 October 2025 (UTC)