Talk:Weise's law/GA1

GA Review
The edit link for this section can be used to add comments to the review.''

Nominator: 18:15, 25 May 2024 (UTC)

Reviewer: UndercoverClassicist (talk · contribs) 10:38, 4 June 2024 (UTC)

Taking a look. I've done very little in the way of phonology since undergrad a long time ago, but can at least comment on clarity, prose and so forth. Initial comments below. UndercoverClassicist T·C 10:38, 4 June 2024 (UTC)


 * Hi First, thanks for taking the time to do the GA. I've taken a cursory look at your GAR and I've begun fixing some of the minor ones. I hope to get to the others tonight or tomorrow. Either way, I will ping you when I think they have been adequately addressed. ThaesOfereode (talk) 11:23, 4 June 2024 (UTC)


 * General matters


 * The article needs to be accessible, per the GA criteria, to a suitably broad audience. Here, that means people without a background in linguistics, and who can only read the Latin alphabet. There are a few things throughout that would benefit from some more attention on this front -- some examples:
 * The article frequently uses linguistic changes as chronological markers, such as it is believed that the change must have occurred prior to the centum-satem division: these should have dates supplied. If there's debate as to when these things happened, say so, perhaps in a footnote, but try to give the reader some broad sense of when we're talking about. Our own article on the "split" says that most linguists no longer treat it as an event at all, though I'd need to do some reading before I can really take a view on that.
 * It would be helpful to set out in rough terms the relationships between IE languages: a "family tree" as an illustration would be useful. In particular, it would help to clarify the distinction between a centum and a satem language, and why this is so important.
 * As far as possible, technical sections should be written in terms which non-specialists can understand: this might include glossing some key terms, adding footnotes, or writing the sentence in simpler but vaguer terms first. See for instance:
 * the law commonly affects zero-grade stems which often receive epenthetic vowels in daughter languages
 * Because the palatovelar sounds underwent assibilation in the satem languages while the plain velars did not, the merging of palatovelars with plain velars explains why these words have plain velar reflexes in words that share a common Indo-European root containing a palatovelar


 * Reply to "General matters"
 * I have expanded the general accessibility of the article a bit (the particular examples are totally valid criticisms; they are exceptionally dense and needed rectification), but, now that I have fixed some particularly dense language, I think the audience is appropriate for the topic at hand. This is an extremely technical and comfortably postgraduate-level topic (its principal sources are Kloekhorst, Kortlandt, and Beekes; none of whom would be mentioned in an intro to historical linguistics course except possibly Beekes) and it is described at a level readable to anyone with a basic undergraduate linguistics background in accordance with WP:ONEDOWN (namely, that "articles on postgraduate topics can be aimed at readers with some undergraduate background"). A linguistics 101 class would learn terms relating to places of articulation, manners of articulation, and basic IPA and an intro to historical linguistics class would learn terms like assibilation and conditioned change. No one would need to know a script other than the Latin alphabet to read this article, since the Sanskrit text is functionally cosmetic. My concern is that too much explanation runs the risk of violating WP:SS. Sufficiently technical terms are linked for users with less experience and more complex terms have now been put into phrasing that I think linguistics undergraduate students would be able to fully understand after one historical linguistics class.
 * WP:ONEDOWN is a little controversial, and has been discussed at length elsewhere (from memory, at FAC talk): it was difficult to get much consensus, but generally understood that it conflicts with the GA and FA standards, particularly as far as "appropriately broad audience" is concerned. One of the major issues was that there are plenty of ways (TFA, ITN, DYK, Random Article...) that a reader might come onto any given article haphazardly, rather than consciously seeking it out via higher-level articles. Most of that doesn't really apply to GA, and there is of course a tension that you can make things "clearer" by explaining them, but making the article too verbose then makes it difficult to understand in another way. I don't think we can really give a hard-and-fast rule here, but it's worth being aware of the trade-off and evaluating whether we're on the right side of it in each situation. UndercoverClassicist T·C 06:31, 6 June 2024 (UTC)
 * I didn't realize WP:ONEDOWN was controversial in that way. Still, I'm willing to try and lower the barrier of entry here, but I want to be cognizant of not getting too verbose and ending up with a page that looks more like a summary of phonetics or historical linguistics. ThaesOfereode (talk) 00:18, 7 June 2024 (UTC)
 * Re: chronological markers – I think this is a generally good suggestion, particularly for the splitting of Anatolian languages and the latest possible event zone for the law (i.e., the splintering of Proto-Indo-Iranian), so I have cited those. With respect to the centum–satem division, as I think you've started to uncover, this is extremely problematic in terms of sourcing. For one, there are two competing theories: a theory that centers on the split of the two languages chronologically (i.e., phylogenetically) and another that centers on "wave" theory (i.e., that one or two language families may have developed satem features while others adopted them through language contact). The latter has been around since at least the 1870s, as far as I know, but Kloekhorst – being the principal source for this topic – I think clearly does not appear to back that, given that he continually categorizes languages as centum or satem based on phylogeny (i.e., a "genetic" relationship) rather than typology (i.e., a characteristic relationship), and he is a reputable source on IE linguistics. During a cursory search, I was unable to find a reputable or source that would pin the date, likely because the wave theory has fairly solid evidence and these dates are mostly developed through deduction; in other words, it appears that the jury may be out on which theory is correct and I haven't found any source supporting the phylogenetic theory that gives a date.
 * Yes, I thought this might be the case. Does any give it even roughly: as in "some time before the Bronze Age" or similar? I'm thinking of just giving people who know very little about the topic a rough sense that it was "a very long time ago" but not "unimaginably ancient". This might be anathema, but The Horse, the Wheel and Language has a diagram on p. 57 here, which gives a terminus ante quem dates of 2000 BCE at the latest. UndercoverClassicist T·C 09:07, 6 June 2024 (UTC)
 * So this is referencing some pretty controversial or deprecated theories of phylogeny. I know Italo-Celtic is (was?) receiving some renewed study, but Germano-Balto-Slavic hasn't been a common suggestion in over a hundred years (the suggestion I'm familiar with is from 1861). Greco-Armenian... I'm not sure I've ever seen that in a serious paper. Though, as I noted in the article, Kortlandt did consider Albanian to be a "transitional dialect" between Balto-Slavic and Armenian, which I had never heard before I read his paper. So it's possible that a paper supporting these conclusions is out there and I admittedly haven't kept up with the literature on that end. That said, given the other more concrete dates here and the generally unsettled state of "phylogenetic vs. wave" discourse, is it really wise to add the dates for the split? I don't think it's too much to ask the reader to guess "somewhere between 4500 BC and 3000 BC" based on context. ThaesOfereode (talk) 00:18, 7 June 2024 (UTC)
 * I think the first paragraph of "Relative chronology" is pretty good (TAQ definitely c. 3000, TPQ probably c. 4500). Could that be briefly summarised in the lead? UndercoverClassicist T·C 11:22, 17 June 2024 (UTC)


 * Re: clarifying the distinction between a centum and a satem language, and why it is important – I think I summed this up with The law affects the palatovelar consonants of the Proto-Indo-European language: *ḱ *ǵ *ǵʰ. In the satem languages, as well as in Albanian and Armenian, these sounds became fricatives rather than remaining as stops. This effectively sums up the distinction, I think.
 * It does and it doesn't: I do think would be clearer if we gave the IPA for these as well. I'm not sure there's many readers who know what a fricative is but don't know what a satem language is. UndercoverClassicist T·C 06:31, 6 June 2024 (UTC)
 * I'm not sure IPA will make it clearer (is IPA commonly used outside linguistics? I have linguistics books that play fast and loose with IPA!) and even if it did, I don't think there's full agreement on what the palatovelar series actually was. I think it's generally assumed to be palatal stops, but they easily could have been palatalized velars. Holopainen (2021) discusses the relative rarity of palatal stops becoming plain velar stops and that there are some arguments that palatovelar stops in general may have been a satem development, unique to those languages. In other words, "palatovelar" seems like more of a placeholder than a specific place of articulation, so I'm reluctant to put it in IPA. Moreover, if palatovelars are unique to the satem languages, I haven't read anything re: actual pronunciation value, since Balto-Slavic and Indo-Iranian linguists use different notations that don't account for a post-PIE, Proto-Balto-Slavo-Indo-Iranian language. Holopainen cites Lipp (2009), but I haven't read it to determine if it even provides a pronunciation for said family. Even then, it might breach WP:FRINGE since a two-way velar distinction (plain vs. labialized) is extremely controversial. ThaesOfereode (talk) 16:25, 6 June 2024 (UTC)
 * I think I might be a few steps behind you: I just wanted a plain-language definition of "fricatives" ("sounds like // and //", or similar) UndercoverClassicist T·C 16:31, 6 June 2024 (UTC)
 * Damn. I really wrote all that over a misunderstanding. Yeah, I can definitely do that. My bad. ThaesOfereode (talk) 16:34, 6 June 2024 (UTC) — Symbol confirmed.svg Done


 * Smaller nit-picks


 * centum–satem divide and similar use an endash, not a hyphen.
 * Symbol confirmed.svg Done.
 * Text in non-English languages should be in language templates, and italicised if in the Latin alphabet.
 * Symbol confirmed.svg I think I got all of them.
 * Quotations (Despite having "been largely forgotten by the scholarly world" need their author and a bit of context given in the body text.
 * Symbol confirmed.svg Done.
 * Glosses (մերձ merj, "near, close to" use single quotes, not double.
 * Symbol confirmed.svg Done.
 * The navbox for IE sound laws gives (I think) thirteen different sound laws: why do we single out the boukólos rule for the "See also"?
 * The boukólos rule is singled out because it is nearly a direct correlation of Weise's law in the centum languages, dealing with a different set of dorsal consonants; that is, Weise's law deals with a change affecting the palatovelar sounds which have a drastic effect in the satem languages and no effect in the centum, while the boukólos rule affects the labiovelar sounds which have a drastic effect in the centum languages and no effect in the satem.
 * Any chance of explaining that briefly in the article? UndercoverClassicist T·C 06:31, 6 June 2024 (UTC)
 * I don't mind doing so, but I think it's off-topic and probably violates WP:NOTSEEALSO, namely: "As a general rule, the 'See also' section should not repeat links that appear in the article's body." ThaesOfereode (talk) 00:18, 7 June 2024 (UTC)
 * I expected the explanation to be in the See also section. The short description there doesn't mention that it's a counterpart in the centum langauges: I think your explanation here of why it's relevant is much better than the one we already have. UndercoverClassicist T·C 06:49, 7 June 2024 (UTC)
 * I see. I think what I've settled on is satisfactory. Symbol_confirmed.svg ThaesOfereode (talk) 01:50, 8 June 2024 (UTC)
 * I agree. <b style="color:#7F007F">UndercoverClassicist</b> T·C 07:21, 8 June 2024 (UTC)
 * Great! Unless there's anything else that needs tidying beforehand, I'm looking forward to the source review. I've transitioned the page from r citing to harvnb and sfn formats for ease of browsing (and to make it easier to read for the reader). Let me know what you think. ThaesOfereode (talk) 14:12, 8 June 2024 (UTC)


 * Image review


 * None present: under the GA criteria, it is preferred for articles to be illustrated if possible. As above, I think an IE family tree (perhaps with some dates) and perhaps separately an chart of centum vs. satem languages would be extremely useful additions. Are there any images of Weise himself or his publications you could use? If feeling particularly creative, a visualisation of the sound change in action would be excellent.
 * I looked for an image of Weise before and not even Google turned anything up. I found his epitaph in Eisenberg that I think should suffice. There is a scan of the title page of his Syntax der Altenburger Mundart ('Syntax of the Altenburg Dialect', 1900), but I figured the epitaph was better. Happy to use either though, to be honest. With respect to images, I have added triptych of the assibilation process – from palatal stop, to postalveolar affricate, to alveolar sibilant. Another user has done a wonderful job illustrating it and I think it fits the bill. Similarly, in chronology, to illustrate the geographic divide, I have added a map which distinguishes (uncontroversial) PIE daughter families by centum and satem categorization.
 * Nice images. Could we add the IPA to the caption for (e.g.) "palatal stop"? On another note, we don't seem to italicise centum or satem: is that intentional? Sources seem to be divided but the MoS would generally advise italics for non-English words. <b style="color:#7F007F">UndercoverClassicist</b> T·C 06:31, 6 June 2024 (UTC)
 * Thanks! I've added the IPA. As for italics for centum–satem: when I wrote this, I was still getting used to the MOS and probably just matched the convention used on the actual centum–satem page, which doesn't italicize it. I prefer without italicization because I feel like that's probably more common in the literature (though I don't know for sure if that's true or not), it's likely to get confused with other language text which is italicized throughout the text, and it's a term that's been technicalized; if it still retained some semantic relationship to 'hundred', there would be no question (such as using fait accompli or amigo), but given its usage in English, it seems unnecessary. That said, I'll defer to you; many authors prefer to italicize it, including Kloekhorst and Kortlandt. ThaesOfereode (talk) 15:30, 6 June 2024 (UTC)
 * The general rule, especially at GA, is that almost anything goes as long as it's consistent: I think that should be our principle here. If you think it's best deitalicised, go ahead -- just make sure it's always deitalicised in the same context. <b style="color:#7F007F">UndercoverClassicist</b> T·C 16:26, 6 June 2024 (UTC)
 * Agreed. It looks like all instances of both terms are fully unitalicized throughout the page, so I will leave them as is. Symbol confirmed.svg ThaesOfereode (talk) 16:29, 6 June 2024 (UTC)

New images:


 * File:Eisenberg Wohnhaus Oskar Weise Tafel.jpg: Germany does have freedom of panorama, so the CC licence is fine.
 * File:Voiceless alveolar sibilant affricate.gif and its two cousins are under CC, which is fine.
 * File:Centum Satem 2000BC.png is appropriately licensed, but needs a source for its data. Given that it's not saying very much that's controversial, it should be possible to find a source which supports it and add that in.
 * Symbol confirmed.svg Done. ThaesOfereode (talk) 21:39, 9 June 2024 (UTC)


 * Source review

To follow, once the above is sorted.


 * 5a: I can see the citations to Weise and Meillet but am not seeing, particularly the latter part, explicitly stated in the text.
 * It looks like it should cite the following two pages as well, since in §7 unambiguously confirms that it occurs and credits Weise, but most of the following sections attack elements of its proposition, most notably in that §10 disputes its phylogeny, §11 shows an expanded scope in Balto-Slavic (kind of an addendum to §10 in my eyes), and §13 cites Meillet in that there seems to be the need for some other kind of rule to explain issues with depalatalization before *l. I think that summarizing Kortlandt's paper as arguing that the scope is small fits the general argument of what he was attempting to get at; I've fixed the sfn to reflect the larger scope of the argument. ThaesOfereode (talk) 21:39, 9 June 2024 (UTC)


 * 5b: I am not seeing, but this may be because it's stated in some other terms that I'm not picking up as equivalent. This is, however, also cited to note 11a, so not a verifiability problem: the citation just needs to be (re)moved if it doesn't also support this bit.
 * That was a typo; it has been corrected to "morphemic boundary". I have added a multiref cite bundle to clarify which information each citation confirms since Kortlandt only discusses the clear derivational relationship (which he refers to as a "model for [...] restoration") ThaesOfereode (talk) 21:39, 9 June 2024 (UTC)


 * Notes 11a and 11b seem to check out.
 * Notes 19 and 20 check out. It would be preferable to (also) have a third party discussing Beekes' criticism to establish that it is notable enough to be included under WP:DUEWEIGHT, but I don't think that's strictly essential, at least at GA.
 * I'll defer to you on this. I think Beekes' reputation as a definitive source in PIE linguistics and the relative lack of discourse on Weise's law in general is enough to warrant his inclusion here. ThaesOfereode (talk) 21:39, 9 June 2024 (UTC)

Could you take a look at these, point out anything I've missed and make any necessary fixes?


 * Response to volley 1

I think I've fixed up several issues that you've brought up and have at least begun discussions for everything else. I'm looking forward to your comments and getting this page to GA. I hope you find what I've changed (and argued) satisfactory. Let me know what you think.


 * Response to volley 2

I've responded substantially again. Let me know your thoughts. ThaesOfereode (talk) 00:18, 7 June 2024 (UTC)


 * Response to volley 3

I think I got everything, though this page is getting longer and harder to read as time goes on so if I missed anything just let me know and I'll come back and fix it. ThaesOfereode (talk) 21:39, 9 June 2024 (UTC)


 * Looks good; as we had some wrinkles with the sourcing, I'll come back and do another sample. <b style="color:#7F007F">UndercoverClassicist</b> T·C 06:35, 10 June 2024 (UTC)
 * Sounds good. Any update on this since it's been a few days? ThaesOfereode (talk) 11:56, 14 June 2024 (UTC)
 * It has: I'm afraid I probably won't be able to get to these until next week, but I'll make sure that I do. <b style="color:#7F007F">UndercoverClassicist</b> T·C 12:18, 14 June 2024 (UTC)
 * No worries; Lord knows life has to happen first. Thanks for letting me know. ThaesOfereode (talk) 16:41, 14 June 2024 (UTC)
 * OK, a few more:
 * Note 3a: in passing, is a long and not quite grammatical sentence (noting doesn't fit with the structure established by the early part). I'm not seeing support for, though I can wear  as a slightly loose summary of what K. has to say about the exceptions and problems he raises.
 * Kloekhorst is also full-named and linked twice in a row.
 * Note 15 checks out for both sources.
 * Note 7: I'm going to be very picky here, but is not supported by the source (it would need to actually say "this is the first time I have presented this idea" or similar).
 * <b style="color:#7F007F">UndercoverClassicist</b> T·C 06:37, 17 June 2024 (UTC)
 * Responding:
 * I admit it's a little abstracted, but Kortlandt and Beekes demonstrate (from the paper's 2011 date) over 30 years of support (over 45 years from today). I can loop in another source (i.e., moving up cite 4 to that particular sentence as well) if you want me to use Meillet to indicate that the law has been supported since the 1890s.
 * Fixed
 * Done
 * Let me know what you think. ThaesOfereode (talk) 12:10, 17 June 2024 (UTC)
 * On the first one: taking two sources that support a claim, noting that they're 30 years (or however long) apart, and turning that into "30 years of support" is unfortunately WP:SYNTH. There's nothing wrong, though, with saying "in the 1890s, Meillet supported it, as did so-and-so in the 1990s and such-and-such in the 2010s". However, to turn that into a wider claim that the source has "a long history of support", we need a source that actually says as much. <b style="color:#7F007F">UndercoverClassicist</b> T·C 12:48, 17 June 2024 (UTC)
 * Alright, fair enough. I'd like to eventually bring this to FA, so it's good we're catching these now. It should be fixed now in accordance with your recommendations. You'll see the Kloekhorst 2011 source notes it's found "sporadically" in the literature. ThaesOfereode (talk) 00:18, 18 June 2024 (UTC)
 * I think we're there now, as far as GA is concerned. Nice work on the article and thank you for your patience and responsiveness with the review. <b style="color:#7F007F">UndercoverClassicist</b> T·C 06:28, 18 June 2024 (UTC)
 * I think we're there now, as far as GA is concerned. Nice work on the article and thank you for your patience and responsiveness with the review. <b style="color:#7F007F">UndercoverClassicist</b> T·C 06:28, 18 June 2024 (UTC)