This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 1

Archive 2

Archive 3

Archive 4

I have formant values of Russian vowels from Sound Pattern of Russian (1959) and would like to create an image of a vowel chart/trapezium using this information. However, there is no general formant value for these vowels. Instead, consonant vowel (pa, pʲa, va, etc) and vowel consonant (um, upʲ, up etc) sequences, as well as a few example words, are given (You can see most of the relevant appendix in the Google Preview of the book). Averaging all these together wouldn't be a good idea. Instead, we should pick an environment most representative of these vowels. Does anyone have an opinion on what might be the best choice? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 00:21, 10 January 2012 (UTC)

Dale Chock has argued in this edit summary that Bidwell (1962) is a "maverick, mistake riddled proposal" and that mention of it gives undue weight to a fringe proposal. However, there are two issues that I think complicate the matter.

1. It's my understanding, based on Stankiewicz (1962) that the analysis proposed by Bidwell (1962) is one of a number of similar analyses. While he doesn't provide full citations, Stankiewicz lists six scholars that he says make similar claims:

• Ludolf
• Olmstead
• Orenstein
• VanCampen
• Hodge
• Kuznecov

Similarly, Folejewski lists (also without full citations) of scholars who seem to implicitly adopt an analysis similar to Bidwell's:

• Trager
• Cornyn
• Karcevskij

I don't think we should remove a paragraph about something broader than Bidwell (1962) simply because of problems with Bidwell. If we had a better understanding of the place that these other scholars (as well as the ones I listed in the archives) have in the general scholarship, it might help us get a better understanding of whether mentioning this analysis really gives undue weight. I have restored the paragraph in question with a POV-section tag to draw others to this discussion.

2. Several editors have alluded (see above and in this archived discussion) to a more robust dispute amongst Russian linguists between a five-vowel analysis (prominent in Moscow schools) and six-vowel analysis (prominent in St. Petersburg) that is, whether ы represents a phoneme or an allophone. I'm not familiar enough with this dispute, but I wouldn't be surprised if the six-vowel analysis also collapses the phonemicity of hard-soft contrasts in ways similar to Bidwell and others. Again, looking into sources would help out in this regard. Either way, I don't think we have enough sourcing present in the article to back up this removal of a citation request, which is why I have restored it. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 23:23, 6 April 2012 (UTC)

AEsos's restoration is factually invalid because he restored a mass of mistakes (names misspelled, names missing, awkward English). He did not confine himself to restoring the material that he claimed was the reason for the restoration. Accordingly, in due course his broad brush action will be undone. Dale Chock ( talk) 00:11, 7 April 2012 (UTC)

Oops! My bad. I've fixed it. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 00:22, 7 April 2012 (UTC)

New reply to no. 2 above 23:23, 6 April 2012 (UTC). The commenter says "i'm not familiar enough with this dispute. Looking into sources would help." You have been editing this article for five years. You are the one who has pushed all these names in front of us -- when do you intend to "look into these sources"? . "AE" is pretending he's discussing theory. He has no understanding of the theory of any article he edits on languages or linguistics -- i've spelled this out in the talk pages of two of those other articles.

To summarize about this proposal by Bidwell ca. 1955-1960, this proposal has been a dead end for nearly half a century, nobody talks about it, so there having been eight or so other proponents 50 years ago is irrelevant. Stankiewicz's article demolished the main proponent, Bidwell. Critics at the time pointed out the inconsistency in not applying the notion to any other phonological parameter, say voicing: to claim voicing is a phoneme applied to p, t, k, f, s, ʃ. Dale Chock ( talk) 07:49, 22 April 2012 (UTC)

I don't have access to Russian-language sources. There could be English-language sources that discuss the dispute, though I haven't found any. With incomplete citations, it's hard for me to go forward in corroborating or expanding on Stankiewicz (1962). Apparently, since I have "no understanding of the theory of any article" I shouldn't be the one to do it anyway. Perhaps you could help.

I don't think we can really say what is and is not relevant until we at least try to get a broader view of the context that these sorts of claims are made. We know Bidwell specifically isn't very notable, but Lightner (1972) is. Until you or I bring any further sources that help enlighten us on the matter, I don't think it's a good idea to delete that paragraph. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 16:12, 22 April 2012 (UTC)

This might be a good starting point. The author points to three disputes (the phonemicity of [ɨ], that of palatalized velars, and that of long postalveolars), but doesn't mention dispute about the number of consonants. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 19:10, 22 April 2012 (UTC)

Per this edit, where Dale describes a table as "trivial" and "uninformative," I would have to wholeheartedly disagree. Not only does it show what kinds of clusters are possible in Russian, but it is sourced, coming from a more elaborate table in Halle (1959) that is even duplicated in Chew (2010). Remember that this project is for lay readers who may not have as intuitive an understanding about phonotactics as experts like you and me.

Also, Dale, removals of citation requests like this are inappropriate. In this case, it is a misrepresentation to say that I have requested confirmation that the phrase exists. It only takes a little bit of common sense (and also the ability to read what I wrote in the edit summary when I restored it the last time you deleted it) to know that the citation request is not for the phrase in question, which obviously exists; it is for the claim that it has a five-consonant cluster. Sources I've seen say that Russian has a maximum of four-consonant clusters, and if I felt more confident that I'd read a representative sample on Russian phonotactics I would even mark the claim with a "dubious" tag. Instead, I'm giving you (or other editors) a chance to provide sourcing that states that Russian allows five-consonant clusters. You got it from a source, didn't you? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 12:01, 25 April 2012 (UTC)

Russian has initial clusters with up to 5 consonant phonemes in a single word (in accurate pronounciation of existing obscene verb "взбзднуть" ("д" is silent) and its derivatives, a well-known example fixed in dictionaries [1]), or up to 6 if preposition is included (several theoretically possible combinations like "к взбзднувшему" or "с взбзднувшим" [2]).

The longest final clusters have up to 5 consonants among occasionally used (but real) words ("арбитрств" [1] 'of arbitratorships', "петиметрств" [2] 'of dandyships', "адъюнктств" [2] 'of adjunctships'), and up to 7 consonants in potentially constructable words ("монстрств" [1] 'of monsterships').

[1] These two examples are taken from: В. Н. Топоров, "О дистрибутивных структурах конца слова в современном русском языке" // сборник "Фонетика, фонология, грамматика", М.: "Наука", 1971 (see a footnote on page 155).

[2] These are gen. pl. forms of петиметрство and адъюнктство, both listed in "Словарь современного русского литературного языка", 17 vols., 1948-1964.

Dictionaries contain several other nouns ending in 2 consonants + -ство (i.e. gen. pl. ends in 5 consoinant letters), but all they have less than 5 consonant phonemes, as they end in -дств- or -тств- where -дс-/-тс- represent a single phoneme, affricate /ts/ (also it's why "адъюнктств" is an example of 5-consonant cluster, not 6-consonant). Actually, this is a productive word-forming model, so many examples like киборгство, оркство, бобрство, кадаврство, спектакльство, октябрьство, ноябрьство etc. are quite possible (and can be found by Google in blogs and similar places). Plus Sovietisms from fiction literature like комсоргство, профоргство, парторгство...

And the last note: potentially, final consonant clusters can be further extended if the word is followed by one of vowel-free particles (б, ж, ль). -- 68.127.102.86 ( talk) 06:07, 26 April 2012 (UTC)

While it's true that those words and phrases are written with such clusters, the issue is whether they are pronounced with them. The article already states that potential consonant clusters are reduced. This source says (p.80) that "clusters of three or more consonants are frequently simplified in pronunciation by the deletion of one of them. So a word like чувство is pronounced [ˈtɕustvə], not [ˈtɕuvstvə]. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 14:34, 26 April 2012 (UTC)

Yes, they are. At least in accurate speech. Dropping of the first "в" in "вств" is a known exception. -- 68.127.102.86 ( talk) 19:02, 26 April 2012 (UTC)

Do you have a source? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 04:16, 27 April 2012 (UTC)

Those two dictionaries I mentioned in "Soft ц" thread above. Only -вств-, -дств- and -тств- are mentioned. Initial "в" of -вств- is dropped in cases чувств-, здравств- and -лвств-, otherwise it is [f]. -- 68.127.102.86 ( talk) 11:10, 27 April 2012 (UTC)

I'm missing something. All you've backed up is that those words/phrases exist. I'm talking about the claims you've made about how they're pronounced. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 13:26, 27 April 2012 (UTC)

There are two pronunciation rules related to (something)+ств. One says that -дств- and -тств- start with affricate [ts] (see Ageenko & Zarva, p. 25, or Avanesov, p. 677), another is related to -вств- (p. 27 and 677 resp.). Ageenko & Zarva are also adding here that "ф" in -фств- is not dropped. -- 68.127.102.86 ( talk) 19:11, 27 April 2012 (UTC)

I don't have a copy of either dictionary. Do you think you could provide quotations and, if necessary, translation? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 15:17, 28 April 2012 (UTC)

I can, but why? Do we cite entire paragraphs of grammar rules to support any claim? -- 68.127.102.86 ( talk) 22:17, 28 April 2012 (UTC)

No, I'm more curious as to what generalizations the sources make about consonant clusters. I think you're saying that the sources argue for more than four consonants when suffixation occurs. Feel free to expand on the part of the consonant clusters section per the sources you've provided. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 23:48, 28 April 2012 (UTC)

Ok, here is the full list of letter combinations where consonant cluster is (or in certain subcases can be) simplified, from the two dictionaries combined together (there are tiny distinctions): с(т)н, з(д)н, с(т)л, с(т)ск, н(т)к, н(д)к, н(д)ск, н(т)ск, (в)ств, р(д)ц, р(т)ц, л(н)ц. More complex transformations (where number of sounds is less than number of consonant letters, but not just dropping one of them): set of rules of type бб/пб -> [б(:)] (the dictionaries use simplified Cyrillic-based phonetical notation instead of IPA); сж/зж/жж/жд -> [ж('):]; зш/сш ->[ш:], сч/зч/жч/сщ/зщ -> [ш':], тц/дц -> [ц:], стц -> [сц], -тся/-ться -> [ц:а], -стся/-сться/-зться -> [сца], тс/дс -> [ц], тч/дч -> [ч:]. Thus, if none pattern can be applied (and there are no ь/ъ+vowel to create an extra [j]), I may conclude that number of consonant sounds = number of consonant letters. Is it an original research or just a trivial knowledge how to read words of native language? ;-) -- 68.127.102.86 ( talk) 00:53, 29 April 2012 (UTC)

This is in response to the message from AEsos on 12:01, 25 April 2012, regardiing documentation of various lengths of consonant clusters.

I want to open with overall comments about AEsos's editing practices. He fetishizes the criterion of sourcing, believing he is free to insert anything into Wikipedia that complies with the Reliable Source policy. This is self serving, because it's not hard to find a source for something, especially something trivially true. He disregards other criteria. He usually ignores criticisms of his edits entirely. When he doesn't ignore them, his defense is usually limited to "it's sourced" -- which is always beside the point when discussing AEsos editing Wikipedia.

AEsos's approach in editing this issue over five years has been especially goofy because of the incongruity between his making a strong (i.e., remarkable) point that Russian allows four consonants word initially, and his shining the spotlight on the unremarkable shorter clusters.

The bad thing about inserting lists (tabular or otherwise) of examples of clusters of two consonants and of three consonants is that most of the world's languages have words with "clusters" (unbroken sequences) of two consonants. It is Aspergerish to document that English, Russian, or any of thousands of other languages allow sequences of two consonants. The objection that this is a point only plain to "experts" is disingenuous. AEsos also proudly claims his examples are sourced. But you are not supposed to source points of fact which are trivial or which are common knowledge to multitudes of people (there's even a discussion saying so in the Wikipedia policies).

AEsos raises a further objection: we should not assume that the Russian spelling 'k vzglʲadu' is pronounced as written. This objection is no good, because as follows. (1) We do not need to prove that a particular Russian spelling is unrealistic just AEsos, in an attitude of linguistic chauvinism, finds it hard to believe it is realistic. What we would need to prove instead is that a Russian word is NOT pronounced as spelled. Where Russian is not pronounced as spelled, THEN it is appropriate to inform the reader. Indeed, the article already discusses multiple categories of this phenomenon. (2) Furthermore, for AEsos to raise this objection only reaffirms his ignorance of even beginning Russian. By the way, on his User talk page he discloses that he hasn't tried to learn the Russian language, he only studies how it is pronounced and spelled. This after five years of editing "Russian phonology". As I was about to say: in elementary Russian, one learns that the pronunciation of the words 'k', 's', and 'v' has to be expanded to ko, so, vo before some words, as in ko mnʲe. THIS IS INDICATED IN THE NATIVE RUSSIAN SPELLING SYSTEM.

There was a telling incident in this article in 2008. AEsos reverted a fluent speaker on points of vocabulary and grammar! He insisted on a bogus gloss (gorbunʲja means 'hunchback (agentive)') based on a bogus general grammar claim (that an "agentive" meaning is indicated by a suffix -ʲja). Aesos himself originated this misinformation ( Revision as of 22:30, 25 May 2008)). When another editor corrected this on 16 Sept (the word actually means 'female hunchback'), AEsos, in act of complacency and recklessness, reverted, inventing a bogus grammar claim: "female would be gorbuna". Latin and Spanish derive feminine nouns with -a, but Russian doesn't. Only when a native speaker snapped, "Consult a dictionary!" two days later did AEsos fall into line. (Readers can see all this for themselves in the article history.) By the way, this happened at the same time that AEsos had been reverted on basic spelling (e.g., he was ignorant of the simple fact that in Russian spelling, every infinitive ends in a soft sign, except the handful than end in -i!). Aside from ignorance of Russian, let us note one point of ignorance of linguistics. As a linguistic matter, it is makes little sense to conceive of there being an agentive form of a noun denoting a personal quality, like 'tall', 'blond', 'hunchback'. Dale Chock ( talk) 07:37, 28 April 2012 (UTC)

I'm sorry if you don't like the criterion of reliable sourcing. It's kinda the thing here. See WP:RS

If you took a look at Halle (1959) or Chew (2010), you would find that Halle (1959) meant to be exhaustive in the types of clusters possible in Russian. Of the four-consonant types, he said there were only two words that even exhibited such clusters. If there were more four-consonant examples to make a typology, I would have duplicated those in a third table.

It may be "Aspergerish" to exhaustively display such clusters, but that is equally true of the sources used and irrelevant to whether it should be included. If you really believe that knowledge of the types of two- and three-consonant clusters in Russian are common knowledge, then you are not a good judge of what is and is not common knowledge.

As I've said above, I question that the given phrases are pronounced as written because of the literature I've read on Russian phonology that explicitly states a four-consonant maximum and that potential clusters are reduced. I've provided two sources that say as much. It's possible that such sources are incorrect, represent one particular dialect over others, or are simply oversimplistic. But until you or others show that this is the case with other sourcing, I see the inclusion of these phrases as original research not worthy of inclusion. This is the point of a {{ citation needed}} tag and removing it is disruptive.

Teasing me on my inability to speak Russian will get you nowhere. I don't pretend to speak the language and will, of course, make mistakes from time to time. In the case above, I was most likely simply duplicating an error from the source. Perhaps you should scour your library to actually provide resources rather than scour the last 6 years for editing mistakes I've made; if you didn't have a habit of doing the latter, I'd even say it was beneath you.— Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 15:18, 28 April 2012 (UTC)

There are dozens of words with initial 4-consonant clusters. Actually, almost all combinations вз+(б,г,д)+(л,р) and вс+(п,т,к)+(л,р) are realized. After prepositions к and с, clusters expand to 5 consonants.

There are hundreds of words with final consonant+ство (unstressed), even if potentially problematic -дство, -тство and -вство are excluded. Declension generates gen. pl. -ств from unstressed nom. sg. -ство, without any other changes and without exceptions. No pronunciation rule can fit them to drop anything. Does one need to cite the entire list to prove this?

Топоров deals with pronunciation, not just with number of letters (his examples are арбитрств(о) and монстрств(о)).

For петиметрство, there is no matching simplification rule. For адъюнктство, the only rule is тс -> affricate /ts/.

For взбздн(уть), yes, "д" drops in pronunciation (I fixed the case above). Still 5 consonants remain, or 6 if after prepositions.

I suppose that your "4-consonant maximum" rule was based on a limited material, as the extensions belong to rare or marginal words. (However, I will not say that pattern "к взгляду" is too rare in modern language.) -- 68.127.102.86 ( talk) 22:17, 28 April 2012 (UTC)

The sources I've cited may be focusing on individual words... I'm not sure. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 23:48, 28 April 2012 (UTC)

(1) Wikipedia readers indeed do not have in their heads lists of consonant clusters. The point is that even readers interested in Russian phonology, the bulk of them don't care to see such lists, and they are justified in that disinterest. They would realize such lists are trivial because they are non-Aspergerish.

(2) It is invalid to compare a Wikipedia article to the professional literature. In a work meant only for specialists, or in a reference grammar, it might be reasonable to list consonant clusters, for the sake of a tedious completeness that is only of value to grammarians of, in this case, Russian, or to linguists. That is not what Wikipedia is suitable for.

(3) In this article and others, AEsos has a record of writing without having done enough research (as in the immediate example). Moreover, he sometimes makes things up (as in the immediate example), sometimes distorts quotes. This raises the question: what will turn up when we inspect his citations from his two chief sources (Halle; Jones and Ward)? Dale Chock ( talk) 05:53, 30 April 2012 (UTC)

For goodness' sake, already caught one. Where this editor was exemplifying types of clusters, e.g., consonant-consonant-liquid, CCL, where the example words include the words meaning 'squeak' and 'camel', an insertion he made five years ago, in 2007, he apparently didn't know the difference between a morpheme and a word. All these examples are morpheme internal! When you abandon that restriction, then even in Romance languages, clusters of four consonants are numerous: English construct, Spanish construir. Or, heartbreak, with /-rtbr-/. By now, even this editor knows the difference between a morpheme and a word, yet he didn't notice his 2007 oversight during this month's edit disputes. Perusal leads me to doubt that he has made more misquotes from this source (Halle 1959). I did, however, find two places where he cited the wrong page at least, if not the wrong work altogether. Dale Chock ( talk) 10:57, 30 April 2012 (UTC)

Ahh, there we go. I think we're more in harmony now. I've edited the section to be more clear about what the tables present. The Bickel & Nichols find is a good one, though the claim about five consonant clusters is still uncited ( this source only backs up that the phrase exists, which I've told you several times is not the issue). I've restored the fact tag with a reword that should clarify the claim that needs citation. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 16:29, 1 May 2012 (UTC)

This is a dishonest reply because it pretends to be on point when it isn't, and because AEsos lacks the guts to acknowledge error. He was totally in the wrong, and unintelligently so. Since 2007 he's been confusing word roots with words, reading into Halle 1959 something that wasn't there. About clusters of five, I've already explained that: the spelling is to be taken at face value. It is irrational to call for confirmation, and this has already been exhaustively explained. But if AEsos were smarter, he would have noticed an answer in Halle. Dale Chock ( talk) 05:28, 4 May 2012 (UTC)

So the point is that I made a mistake? Got it. Now let's get back on topic. If you want to talk to me about what an awful editor I am, there's a place for that and it's not here.

The other editor and I have provided enough sourcing to show that spelling in Russian can't always be taken at face value. All you have to do is find a source that backs up your claim. That shouldn't be too hard if what you claim is as obvious as you say. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 14:10, 4 May 2012 (UTC)

Can anybody provide a list of initial/final consonant clusters of English? I think it would be more interesting to demonstrate only those Russian ones that have no English counterparts. (Say, тк- or рт- would be more interesting than тр-). -- 68.127.102.86 ( talk) 04:07, 29 April 2012 (UTC)

See English phonology#Phonotactics. Emphasizing what is rare or impossible for English speakers would be enlightening, but I don't think we should eliminate mention of the kinds that English speakers could produce. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 15:28, 29 April 2012 (UTC)

Going into detail demonstrating permissible consonant clusters is a superficial, tedious, impractical exercise. Outside of a publication for specialists, I can imagine two motivations for pursuing this exercise. (1) One sincerely disagrees strongly that it is a superficial, tedious, pointless exercise. This shows one's poor academic judgement. (2) One is relegated to this exercise because one isn't sophisticated enough to report on and discuss more substantial issues. Dale Chock ( talk) 05:05, 30 April 2012 (UTC)

The English table does seem a bit excessive. I think we can limit ourselves to simple generalizations rather than an exhaustive list in the case of Russian. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 16:30, 1 May 2012 (UTC)

Regarding edits done just within the hour. AEsos has freshly restored POV content that is POV because (1) it overemphasizes morphemes over words; (2) it goes into detail illustrating points that linguists find nonnotable; (3) it reflects AEsos's prejudices.

Number (1) is due to his misunderstanding of the subject dating back to early 2008, when he inserted statements about permissible consonant sequences in Russian words citing Halle 1959. In the four ensuing years, nobody corrected AEsos that Halle meant morphemes, not words. (Of course, in Russian, as in English, many morphemes, like 'dog, tree, talk', can be whole phonological words.) AEsos was giving the misinformation during four years that sequences of four consonants were rare in Russian words because he was ignorant of the meaning of 'morpheme', which is 'word root'. In English, there are many words with four consonants in a row, like 'heartbreak' or 'construct'; note the sequences '-rtbr- and '-nstr'. In fact, in English these sequences are possible within a single phonological word due to compounding of whole words ('heart' plus 'break') or derivation (the affix morpheme 'con-' plus the lexical morpheme '-struct'). This leads us to the relevant point that the same statements are true for Russian. There are LOTS of quadruple consonant clusters in Russian due to affixation or to word compounding. On the other hand, as stated in the article, Russian has only two roots which have four consonants in a row. AEsos dwells on restrictions within morphemes, which are of less significance to linguists than restrictions within words. And he is waging an edit war to keep dwelling on morphemes. He's imposing POV. Morphemes and words usually have different restrictions as to sound structure. In Spanish, for example, words almost never can end in two consonants, and they certainly cannot in two obstruents (fricatives and stops), like in 'want' and 'list' in English. However, Spanish has at least hundreds of morphemes that end in pairs of obstruents, like 'cant-' and 'list-' in words 'canto' and 'listo'. Moreover, as the previous two Spanish words show, the pair '-st-' can freely occur in Spanish words, just not at the end of a word.

Point number (2) is about his two tables, (with example words) of permissible consonant pairs and consonant trios in Russian morphemes. To linguists, this material is trivial, therefore, AEsos is confirming his lack of expertise. Among the world's languages having sequences of two consonants is an ordinary phenomenon, totally nonnotable. What's notable is languages that prohibit two consonants in a row. Moreover, as regards categories of consonant pairs and consonant trios, it is utterly ordinary for them to include 'l, r, w, y' (to use English spelling). What's notable is consonant trios where all three consonants are stops or fricatives.

Point number (3) is about AEsos's insistence on demanding a citation for the quintuple sequence /kvzglʲ/, i.e., that this is pronounced as spelled. Contrary to what he would have us believe, Russian spelling shows Russian pronunciation, except for as noted in reference works. The reference words note MANY exceptions, many of which this article already cites. AEsos's contributions to three articles, including this one, have been discussed ad infinitum on talk pages including this one, and have been proven full of misinformation and lazy copy editing, e.g., where he misspells things and doesn't catch it for four years. But addressing this specific disagreement with him: he doesn't make the same sourcing demand for shorter consonant sequences, in Russian or English. That establishes the English centeredness of his demand. It also proves his failure to notice what all these sources tell him. He cites Cubberley, Halle, Jones & Ward copiously, plus he dabbles in specialist journals. He has inserted dozens of citations. How has he failed to notice where they say that spellings of four or five consonants to start off a word are phonetically unrealistic? The answer is that they don't say it. So for that additional reason -- i.e., the experts offer no admonitions in this regard -- he has no business demanding for this insertion to be sourced. Dale Chock ( talk) 23:55, 2 May 2012 (UTC)

I have called you to cite the claim that Russian allows for consonant clusters greater than four consonants. While you have addressed this by removing the citation request, it seems as though the anon has found a source that implies seven-consonant clusters in the syllable coda are permissible (though I'm not sure if he or the source calls these "theoretical"). There is still nothing that shows that clusters greater than four consonants are permissible in the syllable onset. In my most recent round of edits, I have found enough sourcing that I feel confident that four consonants is the maximum for the syllable onset. As such, I have replaced the citation request with a "dubious" tag. I don't have to tell you not to remove such tags before the issue is resolved here, do I? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 16:09, 3 May 2012 (UTC)

Aesos wrote, "No, I'm more curious as to what generalizations the sources make about consonant clusters." He wrote that 23:48, 28 April 2012 (UTC). Why is he still curious when he has been reading "the sources" since 2007?

The answer to the question stated by Aesos 16:09, 3 May 2012 was given by me in the very paragraph that Aesos is replying to, 23:55, 2 May 2012 (UTC) -- plus in the next to last paragraph of 07:37, 28 April 2012. Why's he still asking? Any spelling not mentioned by the grammarians of the Russian language, not subjected to "buts", is fit to go into this article. Under Wikipedia policies, Aesos is not free to obstruct it. By substituting his skepticism for the conclusions of all these sources, sources he's already read, he is committing editing violations. It is WP:SYNTH for him to overgeneralize from a report to the effect that "many clusters are frequently simplified" to proceed to fact-tag all consonant clusters he finds incredible.

Aesos wrote, "As I've said above, I question that the given phrases are pronounced as written because of the literature I've read on Russian phonology that explicitly states a four-consonant maximum and that potential clusters are reduced. I've provided two sources that say as much. . . ." 15:18, 28 April 2012 (UTC) But these statements he inserted five years ago in 2007 were a bone-headed misunderstanding. He MISINTERPRETED the literature (particularly Halle 1959), as I brought to light above, 10:57, 30 April 2012 (UTC). Nevertheless, ever since he's been reverting as persistently as ever. Dale Chock ( talk) 12:10, 6 May 2012 (UTC)

Hmm. Still no source, eh? I can wait. I figure it's probably finals time for you, so three weeks should be enough time to find a source. Good luck. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 17:39, 6 May 2012 (UTC)

Aha, the wind is loud, how can an Aesuos hear other people talking over the roaring wind? Yes, the wind, and the trains outside your window, uh huh. Your experts are satisfied with the spelling, the experts in all those books you hold right side up and turn the pages of. An Aeusoes can obstruct one or two people at a time. So we will wait and see whoever else may join in. That will take time, quite some time. Dale Chock ( talk) 21:21, 6 May 2012 (UTC)

Per WP:CHALLENGE: "Any material challenged or likely to be challenged must be attributed to a reliable published source using an inline citation"

Per WP:BURDEN: "Editors might object if you remove material without giving them time to provide references."

Per documentation at {{ citation needed}}: "Where there is some uncertainty about its accuracy, most editors are willing to wait about a month to see whether a citation can be provided."

Since most editors see a month as a fair amount of time to be given to provide attribution to disputed statements, three weeks from now (which would be more than a month from when I first first placed a citation request) seemed to me like more than enough time to provide attribution.

I must apologize, as it seems (from the above post and its edit summary) that you feel like this three-week window is unjustified. What do you feel would be fair? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 00:14, 7 May 2012 (UTC)

Two days ago, Aesos's great Wikipedia buddy advised him to take his disputes to the talk page. That advice didn't register, we see now, because he's still talking past my arguments on article substance, the content of sources or insertions. Almost everything he ever says is just about citation procedures. At the moment, he also persists in the approach of manipulativeness and aggression, bringing us chapter two of a petulant fiction that I really accept the validity of the demand for a citation. In the last week, I've had occasion to note at WP:AN/I (Archive 750) and AN/EW that Aesos hardly ever discusses my arguments about insertions and deletions. I said there I've written 4 or 8 times in the way of substantive dialog as he has. The closest he has come to engaging with my arguments on article substance is to now cite WP:CHALLENGE. WP:CHALLENGE misses a contingency, apparently in a failure of envisioning. One of his standard tactics of evasion is to falsely shift burden of proof. My apparent "claim" is something that Russianists themselves take for granted, while he is imposing his personal doubt of a Russian spelling, which is an act of original research. Cleverly, he's not inserting his OR, but rather imposing it by illogically demanding a source for someone else's insertion. I am repeating myself. Dale Chock ( talk) 13:20, 11 May 2012 (UTC)

How could I, in a post dated May 7, have been able to react to a post dated May 9?

Anyway, it's clear that you believe that the claim in question is obvious, a position akin to that taken at WP:BLUE. Neither that essay nor the one at WP:NOTBLUE provide me any further blanket policy statements that I can use to convince you otherwise. It is up to me, you are saying, to make the case that doubt should be given to the claim in question. Correct? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 00:58, 12 May 2012 (UTC)

People usually edit Wikipedia outside of a Wikipedia account because they don't expect to participate at length in any one article, or because they want to be naughty or disruptive. Since neither of these describes your participation, I wonder why you haven't created an account at Wikipedia after a couple of dozen or so edits over nearly one month. Although a few of your remarks are constructive, it seems overall like a waste of time to reply to, or try to build on, contributions and comments by someone who doesn't even commit to being an accountable member of the community. I'm curious what you think. Dale Chock ( talk) 05:13, 30 April 2012 (UTC)

It is inappropriate to use an article talk page to discuss off-topic issues. If you would like to discuss something with the anon, I recommend you leave a message on their talk page. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 16:56, 1 May 2012 (UTC)

Until recent weeks, the article devoted many citations to a single point: which syllable gets the stress in the infinitive of the word for 'to rust'? These footnotes documented that the speakers are split on whether the first syllable or the second.

It turns out this bloated insertion arose from a lingering emotional state that started in 2008. Editor AEsos provided a list of examples to illustrate a category of morphophonological alternation: hard and soft consonants. One example involved this infinitive. At some point, AEsos reverted a native speaker on the pronunciation of this word: specifically, AEsos insisted the word is not pronounced as the native speaker claimed. What a scandalous act! Well, it happens that this word has alternative pronunciations among Russians. So it would also be wrong to claim there is but one correct pronunciation. One day (18 Sep 2008), the native speaker fixed the bad edit. Anyway, ever since, AEsos has been accumulating dictionary citations in order to prove that there are alternative stress patterns. Just one source would do. Well, also in order to prove that one of the pronunciations has been gaining favor with the decades--a point which is equally useless in this particular article.

Originally, I modified this discussion only by reducing the number of dictionaries cited. But actually, it is a poor idea to dwell on sporadic unpredictable or irregular words. Since this particular phenomenon is trivial and also is not invoked to prove any other point in the article, and because of finding out the motivation for all the citations, I have deleted it. Dale Chock ( talk) 11:28, 30 April 2012 (UTC)

Ah yes, I remember that dispute. As you can see here, the other editor and I discussed the matter cordially and came to a quick agreement on what to do.

The problem with deleting the example entirely is that the list is introduced as "the different types of alternations" not "some" types. In other words, the list is meant to be exhaustive. In this case, the word pair shows an alternation between an adjective and a corresponding infinitive. Perhaps you or the anon can come up with a different example. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 16:52, 1 May 2012 (UTC)

Tons of them. The most similar ones to ржавый/ржаветь are крова́вый/крова́веть and пра́вый/праве́ть (two examples are fully parallel to those two variants of stress in ржаветь). -- 68.127.102.86 ( talk) 08:39, 3 May 2012 (UTC)

Great! I've added the first pair, though I couldn't find what крова́веть means. Also, I've added a new information in the consonant cluster section but am in need of some examples (marked by the {{ example needed}} template. Do you think you could help fill those out? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 18:18, 3 May 2012 (UTC)

I just realized that I had originally included more examples than are there presently. Since Dale has thinned these out a bit, and since we don't have to repeat Lightner's exhaustiveness, I've changed how the list is introduced to reflect that there are other types of morphological changes that lead to hard/soft alternations. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 01:29, 6 May 2012 (UTC)

I have just restored glosses inserted by at 13:12 30 April 2012 and 10:52 1 May 2012 by Special:Contributions/68.127.102.86. The deletions were done by User:Aeusoes1 at 15:25 1 May with the edit summary, "glosses don't need to be exhaustive". We're talking about a mere nine instances where "68.127.102.86" used 2 to 4 words instead of one word. This is a disgrace, for an ensconced editor to hurry (it was with only a day's delay) to throw water on concise, helpful additions by a new editor. The new editor at least insinuates that he/she is fluent in spoken Russian, which would be a big enhancement to the production of this article.

I consider these deletions an act of hypocrisy on the part of User:Aeusoes1 because since 2008 he has indulged in excesses in exemplification and in sourcing.

(1) He cited 62 instances in a row (in the form of a table) of the alternation between 'e' and 'o'. (2) He cited 17 instances in a row of hard-soft paired consonant alternations. (3) As documented in a section above during this week, he compiled an excessive number of citations of dictionaries over a period of years just to prove that one particular word can be pronounced two ways, and his motivation for this miniproject was he had told a native speaker that speaker's own pronunciation was wrong.

That was as of 27 March 2012. I myself have since thinned these excesses. Dale Chock ( talk) 17:57, 3 May 2012 (UTC)

What, exactly, is your point? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 18:00, 3 May 2012 (UTC)

On 6 May 2012, User talk:aeusoes1 left a long response about these glosses on my user talk page. He didn't explain why not here. From perusing both this page and that response, it doesn't seem like he repeated the parts of that response on this page. That is really something. More than once, he has suggested that a remark I have placed on an article talk page belongs elsewhere. Whether he's right or wrong about that, here I have him NOT placing on an article talk page a comment that cannot be construed any other way than that it is about the insertions in the article! You wonder whether it was a slipup, that he meant to post it here. Anyway, I won't be reading that comment on my talk page. I will read those remarks if they're posted here. Dale Chock ( talk) 05:08, 6 May 2012 (UTC)

Hmm, you seem to have a mistaken understanding of some of our talk page guidelines. You might want to take a look at WP:TPO, which says that you are free to delete comments on your talk page but such removals are taken as proof that you have read what was written.

I had posted in your talk page because the discussion was about issues that go a bit beyond the subject matter of this page. It also seemed, from your missing response to my question (dated 18:00, 3 May 2012), that this thread was a dead end; you've certainly abandoned enough threads midstream to have me think this was another instance. If you feel more comfortable discussing the matter here, you have my permission to move the comments into this section. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 17:01, 6 May 2012 (UTC)

"На месте глухих согласных перед звонкими (кроме [в]) произносятся соответствующие звонкие".

Avanesov's pronunciation guide in "Орфоэпический словарь русского языка" (Borunova, Vorontsova, Yes'kova, 1983), p. 670.

Apologies are expected. -- 68.127.102.86 ( talk) 08:12, 4 May 2012 (UTC)

But hermitlike unregistered user, you failed to notice this passage in our article, under Phonological processes:

"/v/ and /vʲ/ are unusual in that they seem transparent to voicing assimilation; in the syllable onset, both voiced and voiceless consonants may appear before /v(ʲ)/ . . . When /v(ʲ)/ precedes and follows obstruents, the voicing of the cluster is governed by that of the final segment (per the rule above) so that voiceless obstruents that precede /v(ʲ)/ are voiced if /v(ʲ)/ is followed by a voiced obstruent (e.g. к вдове [ɡ vdɐˈvʲɛ] 'to the widow') . . . ."

attributed to Lightner 1972. This fact is reinforced by Halle (1959:64):

"Note that {v} and {v,} play no independent role. Everything transpires as if {v} and {v,} had been absent; e.g., мог войти [mokvajt,'i] ... мог вернуть, [mokv,ern'ut,] ..., but мог вздохнуть, [mogvzdaxn'ut,]".

That is, in the last case г = [g]; or to put it another way: мог вздохнуть is not pronounced мок вздохнуть. Obviously, the great Avanesov you cite left out an important exception.

From your earliest contributions on this page, you seem to be someone who knows Russian, but doesn't know grammar or linguistics. Here we see that you have trouble recognizing this language you apparently know so well when it is filtered through indirect representations. But please continue to help us with the occasional translation. Dale Chock ( talk) 04:00, 6 May 2012 (UTC)

As it usually happens, the truth is in between. Both of us were not fully right (my personal experience supports [k] взгляду, [k] вдове). And what I did just found (many thanks to Google books' database!) in an academic source (М. В. Панов, "Русская фонетика", М., 1967, p. 87):

...наблюдение показывает, что перед группами [вб], [вд], [вз] и т. д. у одних говорящих по-русски происходит «озвончение» глухих, у других его не происходит (т. е. у одних невозможны сочетания «глухой согласный + [в] или [в'] + звонкий согласный», а у других возможны). Это — случай, когда «норма... состоит в отсутствии нормы» (Л. В. Щерба). Объяснение очевидно: эти сочетания редкостны, ведь говорится обычно не к вдове, под вздутием, от взвода, а ко вдове, подо вздутием, ото взвода. На стыке же полнозначных слов (идет вдова) вообще законы озвончения менее строги, чем в середине слова...
(My note: the "evident explanation" is not correct, as the Google books' statistics is strictly opposite. Especially in what relates to от(о) взвода: more than 4000 usages of от взвода vs only three (just three, not three thousands) ото взвода, and two of these three are from Panov himself and from a citation of his book.)

So, the observed pronunciation is dual. However, two pronunciation dictionaries make no special exception for this case, thus we have to implicitly suppose that the nearest general rule must be applied (I've cited it above), i.e. the prescribed pronunciation is rather [kvz...] than [gvz...]. -- 68.127.102.86 ( talk) 06:26, 6 May 2012 (UTC)

Broken English ("nearest"), impossible to make sense of this. As for this Panov, what you think he thinks would be a big exception in 70 years of research on Russian phonology, therefore he is not the default choice. I cited Halle 1959, who is classic in the field (which is not to say that everything he says is guaranteed correct). Here's an unpublished paper by Padgett, ca. 2001: "But like a sonorant [сонорный, i.e., /m, n, r, l/], [the phoneme /v/] does not trigger voicing assimilation . . . . The basic facts of Russian voicing assimilation have been well described (Avanesov 1956, Jakobson 1956, Halle 1959, among many others). The best known treatments of them in the recent generative tradition are Hayes (1984) and Kiparsky (1985)." http://roa.rutgers.edu/files/528-0702/528-0702-PADGETT-0-0.PDF

The first three were leading authorities in older Russian phonology. The latter two are leading authorities in general phonology from 1970 to at least 2000. Avanesov, Jakobson (Moscow, 1896), and Kiparsky are native speakers (Kiparsky was born and raised abroad, but his father was a Russian Slavicist). Halle and Kiparsky were Jakobson's students. It is ridiculous to suppose all these Russian speakers and linguists missed something M.V. Panov noticed. The paper by Padgett, who is a specialist in Russian phonology, examines some phenomena which are controversial among the specialists. But the phenomenon we are discussing is not controversial. Even if you have understood Panov's opinion perfectly, the most you can insist on is to report his contrarian opinion. Dale Chock ( talk) 13:58, 6 May 2012 (UTC)

Friends provided me with a wonderful link: there is a possibility to search through multimedia samples in the "National Corpus of the Russian language" ( http://ruscorpora.ru/search-murco.html) -- just type, say, "к вз*" (without quotes) in the first input field, and there are 4 related short fragments of Soviet/Russian movies ready (with 7 occurrences of "к вз..."). All cases contain [kvz], not [gvz]. Also, youtube.com has about 10 different performings of the popular song "Огромное небо" (from 1960s to recent years), with the fragment "И вздрогнул от взрыва берёзовый лес" (~1:50 from the beginning); all performers sing [tvz], not [dvz]. -- 68.127.102.86 ( talk) 18:45, 3 June 2012 (UTC)

To the unregistered new editor, 68.127.102.86, who has been inserting a whole lot of footnotes in the last few hours. Your format is out of compliance with Wikipedia style. Therefore, these edits are going to create a lot of extra work for fellow editors to rectify them. A first stop for learning footnote and bibliography style is WP:Cite. See also WP:Cite book, WP:Cite journal, WP:Cite article, etc.

Beyond formatting of dates, pages and author names, the tone of your wording in footnotes is less encyclopedic than it should be. Phrases like "see" are usually considered too redundant in Wikipedia style. In general, your style is an old fashioned wordier style, which is out of place here. It is even more outdated in having the year late in the citation. The correct sequence in the bibliography is author, year, title.

It is preferable not to put bibliographic citations in footnotes. Instead, just author, year, and page (or chapter or section).

I note that you do not reply to my justifications for edits and reverts stated in edit summaries or on the Talk page. You do not match my reasons with your own, including in your edit summaries. This abruptness makes you edits more liable to being reverted. For example, you insist on talking about "the names of the letters", but the name of Ы is not the bare vowel, in contrast to the case with И. This makes your claim seem to be in disregard of the facts. Another example is your insistence that the Russian education ministry teaches six vowels down to the present, but as you persistently add sources, these additions persistently do not include sources that confirm present day practice. Dale Chock ( talk) 13:10, 4 May 2012 (UTC)

I cannot understand your position. I make a statement. You remove it with an extremely long comment not related to the matter. I restore it with the reference and citation. You say that this is bad/obsolete/isolated reference. I add more references. You say that they are superfluous. Is it a constructive behaviour?

Do you not trust the dictionaries where letter Ы has name [ы]?

Do you think that 2007 is too far from the present day practice?

And about reference style. Why to make two-stage referencing (text -> footnote -> bibliography item) if a book is used only once? Moreover, if N sources support the same claim, why to produce N footnotes + N bibliography items, instead of just one footnote? And where to place citations from sources in your system? -- 68.127.102.86 ( talk) 13:50, 4 May 2012 (UTC)

If there's a problem with the anon's citations, we can bring it to the talk page, but snapping at a new user for not having a sophisticated understanding of citation conventions is a bit bitey.

And, per the name of ы, I can vouch from personal experience that [ɨ] is indeed a possible name for the letter, although it sounded more like [ɯi]. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 14:22, 4 May 2012 (UTC)

About the 2007 reference. You added it a about an hour after the 1951 and 1970 references. As I remember, in the meantime I started editing, while multitasking. I did not discover your later insertion of the 2007 work.

When I was taught Russian, the letter name we were taught was yerɨ. Even such a long time ago, yes maybe there were two names in use in Russia; but I learned Russian in America, and my teacher was neither a native speaker nor was he a non-Russian Soviet (although he had visited Russia twice). There are statements you could insert at Wikipedia that every native speaker would agree with. But you have to consider that we readers are hardly any of us native Russians. We usually have to judge the reliability of the content based on citations, not based on personal knowledge, and we do not have Russian acquaintances we can run out and ask. If this article had the level of participation that there were many Russian speakers jumping in whenever a wrong statement were inserted, we might be able to use fewer footnotes--we'd let those readers catch the mistakes.

About footnotes, the system you use was formerly standard, and even today it is common in some fields of academia and in books for a mass audience. However, it is not suited to Wikipedia. The most important points of our system is that there should be a list of works consulted (most Wikipedians call it "bibliography") and that this list be complete. It's bad for it to be incomplete (which is the result of listing some sources only in footnotes). And it's not at all acceptable that all sources be in footnotes only because that's not an alphabetized list and it is not a list were each entry is uniformly laid out on the page for easy scanning with the eye. As to why not be economical and collect all sources that were used only once and that bear on the same point: this is a small saving which gets undone in editing Wikipedia. It gets undone because (1) Wikipedia is a forum where the article is constantly being revised (changed). When the sources are not all displayed separately, they, and the footnotes, are harder to count. (2) The editing is done in modern fashion, i.e., in some markup language. The plain text is interspersed with tags, data field labels, etc. When you are editing marked up text, it becomes hard to make sure you saw everything.

Another issue is that a source is often cited more than once throughout the article. You mention the case where N sources bear on a single point. But some of those N sources may be cited again for different claims. With Wikipedia, we enjoy a different editing economy, especially when you use a markup technique other than <ref></ref>: in the list of footnotes, each source is named just once, but each occurrence in the text is indicated with a linked superscript letter. Dale Chock ( talk) 05:01, 6 May 2012 (UTC)

Name of Ы. Last time "еры" was used in school in: С. Г. Бархударов и С. Е. Крючков, Учебник русского языка, ч. 1. Фонетика и морфология. Для 5-го и 6-го классов средней школы, изд. 7, М., 1960 (page 4). The next edition (изд. 8, переработанное, М., 1961) has "ы" (page 20). But the name "ы" itself is known since mid-19th c. (in "sound method" of education proposed by Konstantin Ushinsky). -- 68.127.102.86 ( talk) 08:21, 6 May 2012 (UTC)

Bundling of footnotes. I discovered that Wikipedia enables this while retaining separate hyperlinks. I restored your bundles. Dale Chock ( talk) 10:35, 6 May 2012 (UTC)

I propose do not mention Ostapenko there. She lists some examples (maybe most frequently used ones) and adds "etc." I collected the full set from common dictionaries. With the exception of two occasionalisms, the words are quite common; I believe, any big enough dictionary contains them (or at least similar words with the same prefixes+roots).

The two occasionalisms were found by Google, namely:

• встлеть, two independent occurences in form "встлеет": http://www.stihi.ru/2011/05/08/375 and http://litprom.ru/thread41528.html (the second text can be found under several addresses). Made from "затлеть" where prefix за- was replaced with a high-style one, воз-/вос-/вз-/вс-, cf. закричать/вскричать, загореться/возгореться, запретить/воспретить etc.
• вздлить, one occurence http://litfest.ru/publ/v_zashhitu_nasekomykh/218-1-0-6466 (the last verse; it is written with non-standard spelling, but the particular word is orthographically correct and meaningful). The word was made from "продлить" with a similar replacement (про- -> воз-/вос-/вз-/вс-, cf. прославить/восславить, прогреметь/взгреметь). The difference is not only in style register (as for за- vs воз-), but in nuances of meaning (воз- refers more to the beginning of the process, and the process itself is rather one-moment action than a continuous one).

I'm not sure whether these links are important enough to be placed in the article. -- 68.127.102.86 ( talk) 15:22, 4 May 2012 (UTC)

Ah, I missed the "etc." I've moved the citation up and removed the citation tags.

The claim in her article is "It is interesting that all the clusters that contain four consonants begin with the vz-/fs- (regressive voice assimilation) combination." while this article currently says "The source of many of these clusters are lexical words that begin with the prefix вз-/вс-." Technically, she doesn't say the combination is a prefix, though I don't know if that would be common knowledge to a native speaker or not. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 16:50, 4 May 2012 (UTC)

Twice this week Aesos has been reverted with the allegation by me that a passage is a misquotation of the source. In arrogance, this other editor has blithely restored the disputed passages without explaining how his insertion does not constitute misinformation.

The source, Cubberley p. 82, is unfortunately ambiguous: one cannot be sure of his intended meaning. But logically that means you cannot assume the strong interpretation, as Aesos does. The problem passage in the source is:

"Here we have remnants of older simplifications which occurred as the language shifted from open- to closed-syllable structure. Currently, here as elsewhere, the language is undergoing spelling influence which resists simplification. Some of the groups and words involved are:

(1) by dissimilation." [Here, Cubberley lists some spelled consonant clusters that retain the reduced pronuncation, others that are being "restored" by some speakers.]

"(2) by deletion." [Here, Cubberley lists some words spelled '-vstv-' that retain the simplified pronunciation, others that are being "restored". But, then he lists seven clusters of three or more consonants, which are all dentals or mostly dentals, that underwent simplification, and for none of these seven does Cubberley report any counterexamples.]

Aesos's confusion is to have wrongly extended the scope of the early part of the quotation. Aesos created "counterexamples" that are not clearly affirmed by Cubberley, let alone stated by Cubberley, and it's possible that Cubberley would reject them. Logically, you can't cite a source in support of an insertion that the source does not clearly support. Once again, this other editor misunderstands what he has read--and once again, he doesn't bother to refute arguments against his insertions. In this case, he needs to find an unambiguous source. Dale Chock ( talk) 06:12, 9 May 2012 (UTC)

You are right to challenge the information, as it is either drawing from another source or synthesizes personal experience with Cubberley's "the language is undergoing spelling influence which resists simplification."

However, it's only fair to give editors a reasonable amount of time to back up their claims. It's also fair practice to retain disputed text in the article with citation requests. This is why I restored the text in question with a fact tag.

Importantly, the editor who added this was the anonymous user who has been active in this talk page, not myself. This is not the first time that you have mistakenly attributed edits to me and I ask that you be careful about this. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 13:55, 9 May 2012 (UTC)

Just to be clear, we're talking about the sentence "In certain cases, dropped consonants are restored in more modern pronunciations under the influence of written form." Right? Your most recent edit also removed other content and I just want to make sure that you're not referring to something else. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 22:33, 9 May 2012 (UTC)

Something you inserted here at 13:55 9 May 2012, ". . . give editors a reasonable amount of time to back up their claims" is misapplied to the case. He did give a source and the source does not support his insertion. Now, I have to hand it to you, your indulgent scenarios to account for this are plausible. However, I think the stronger explanation than either of those is that he is relying only on Cubberley p. 82, in which case his insertion is a misquotation of the source. Being so is strong evidence that the claim is all in his mind. Perhaps you have covertly gone a step further and adopted a "give the benefit of the doubt" standard. That would be inappropriate. You seem to generally favor "one size fits all" prescriptions; that is, exceptionless rules. Another consideration is to judge an editor's particular insertion not in isolation, but comparing it to their previous participation. This editor is intellectually erratic and his reasoning is sometimes facile. Moving on, I apologize for misattributing his insertion to you. Dale Chock ( talk) 11:39, 11 May 2012 (UTC)

Those are fair points regarding the edit in question. I happen to interpret the anon's edits a bit differently. Considering that his other edits made use of two other sources, I supposed that he may have drawn from them instead while maintaining the same examples. It's not a big deal either way, since the anon can always reintroduce the edits with proper attribution. I believe we are in agreement on this point. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 00:47, 12 May 2012 (UTC)

In response to the concluding remark at 13:55 9 May 2012, claiming I have mistakenly attributed edits to Auesos more than once. Prove it. I anticipate it's possible, but we can't take your report at face value. I don't remember it happening multiple times. I do remember the incident when you vehemently denied the existence of a certain statement in an article -- i.e., claiming not merely than you didn't write it, but that nobody did, since it wasn't there -- and then you zipped to the opposite stance of informing me, with a smirk, that yes you did insert it. Dale Chock ( talk) 14:44, 11 May 2012 (UTC)

As I mentioned in your talk page last week, you attributed the excessive citations of a particular Russian word to me when I had not made those edits. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 00:47, 12 May 2012 (UTC)

I've reverted prior to the edit war. Deleting citation requests is not acceptable.

Go ahead and restore the changes that you both agree on. As for the rest, please settle it here. — kwami ( talk) 23:12, 9 May 2012 (UTC)

"Deleting citation requests is not acceptable" is a precipitate, uninformed opinion. See multiple contributions under the above section, Consonant clusters. Dale Chock ( talk) 10:55, 11 May 2012 (UTC)

There is currently a dispute in the article, one that is spread throughout this talk page. This dispute has modified over time as new information is uncovered and edits to the article have accumulated. It is not my intention to document the evolution of the dispute but, rather, focus on the current state of editors' positions. It is my hope that, by centralizing the discussion in one thread, involved editors are more likely to respond to each other and that outside editors considering involvement will not feel the burden of having to read all the conversation that has occurred in the last two months. I will first present the content under dispute, then summarize the most up-to-date arguments for its inclusion, and, finally, provide my own response.

Part 1: The content. The dispute in question is whether Russian phonotactics allows for the pronunciation of clusters with more than four consonants in the syllable onset. Since late April, the article has claimed that it does, though with various changes in wording. Its current form in the article is:

Thus, prepositions (especially the three that consist of just a single consonant: к, с, and в) contribute to phonological words with up to five consonant clusters in the syllable onset. Examples are к взгляду [gvzglʲadu] 'to [the] gaze', к встрече, 'to/for [a, the] meeting', as in 'ready for a meeting'.

Part 2: Arguments for. The quotations below, which are more illustrative than exhaustive, are from Dale Chock, who originally included the claim under dispute and is its primary proponent in this talk page. Since these arguments are located in this talk page, I have not provided links to them. My focus is on just the arguments for the claim in question; other methods of rhetorical persuasion, such as appeals to authority or ad hominem attacks, will not be addressed.

• Point 1: Accuracy in orthographic representations
  • "We do not need to prove that a particular Russian spelling is unrealistic...What we would need to prove instead is that a Russian word is NOT pronounced as spelled. Where Russian is not pronounced as spelled, THEN it is appropriate to inform the reader."

Russian has a very transparent orthography. While there is a morphemic basis to Russian spelling, it more-or-less offers a fairly accurate correspondence to pronunciation. Thus, in relation to consonant clusters, the argument goes that consonant clusters are assumed pronounced as written until shown otherwise.

• Point 2: Exceptions to accurate representations are noted in dictionaries/grammars
  • " Any spelling not mentioned by the grammarians of the Russian language, not subjected to "buts", is fit to go into this article"

Cubberley (2002), cited in the article, puts forth a list of "older simplifications which occurred as the language shifted from open- to closed-syllable structure" (p. 82). In addition, an anonymous editor has access to two Russian-language sources (what look like a linguistic grammar and an article) that lay out well-known exceptions to accuracy in orthographic representations. Whether or not these sources intend these listed exceptions to be exhaustive, the burden falls on the editor seeking to show further exceptions.

• Point 3: Affixation and phonological words
  • "There are LOTS of quadruple consonant clusters in Russian due to affixation or to word compounding"

Certain affixes in Russian may add to an already-existing consonant cluster. This is also true of prepositions that have no inherent vowel. This means that a word's dictionary citation form with a three- or four-consonant onset may be further clustered with a preposition or prefix attached to it.

• Point 4: Orthographic representation of epenthesis
  • "in elementary Russian, one learns that the pronunciation of the words 'k', 's', and 'v' has to be expanded to ko, so, vo before some words, as in ko mnʲe."

This is an extension of the transparency of Russian orthography. As Cubberley (2002) indicates, there are a number of general situations where difficult clusters are broken up with epenthetic vowels. These epenthetic vowels are indicated orthographically. A rudimentary example is the three prepositions that consist of a single word unless they precede an already difficult consonant cluster, in which case an /o/ is epenthesized.

Part 3: Ƶ§œš¹'s responses and rebuttals. The justifications for my position, expanded on below, boil down to this: Russian has well-known patterns of mismatches between orthography and pronunciation, often due to consonant cluster reduction. This, in combination with explicit claims in scholarship of a four-consonant maximum in the syllable onset, prompts me to believe that four consonants are the maximum in the syllable onset, with potentially larger clusters being reduced through deletion and epenthesis (processes that also occur with smaller clusters). I believe it is original research to assume that these claims of a four-consonant maximum are limited to lexical (rather than phonological) words, and I don't believe that the orthographically represented process of epenthesizing ‹о› with prefixes and prepositions can account for Russian speakers' general strategies in dealing with complex onsets.

Points 1 and 2: Inaccuracies in orthographic representations. I do not dispute that Russian has a transparent orthography (Point 1). I also do not dispute the general position of Dale's that the burden is on editors wishing to argue that a given orthographic representation is inaccurate in Russian (point 2). While I have made efforts to fulfill this burden, it seems that it has not yet been convincing. As such, I have taken the effort to search resources available to me to further back up my case that the claim in question (that Russian allows more than four consonants in the syllable onset) requires attribution. Here are my justifications:

• Patterns of mismatch between orthography and pronunciation.

As presented in the article, there are certain patterns where consonant clusters are simplified from their written form. Importantly, the patterns that Cubberley (2002) lists for "word-specific clusters" are not exhaustive (it is unclear whether he meant them to be), as an anonymous user has contributed two more examples of such patterns. Furthermore, Halle (1959) lists an example of deletions in his "P" (phonological) rules that closely parallels these "historical" rules. This rule, which says (p. 69) that dental stops are dropped in the position between a dental continuant and |*n| (e.g. |ˈle*stʲnij| = [ˈlʲɛsnɨj] 'flattering'), suggests that the more complex (that is, pre-deletion) forms are present in speakers' underlying mental grammars and deleted in the process of production, rather than being "historical." More importantly, though, these show that Russian spelling is not always accurate.

• More general claim of deletions

Cubberley (2002) states that "Clusters of three or more consonants are frequently simplified in pronunciation by deletion of one of them" (p. 80). He gives an example of a general pattern of deletion with geminate consonants that are simplified in clusters and then moves on to a section on the historical "word-specific clusters" that I allude to above. This is the totality of his coverage of deletion and I suspect that he did not mean to be exhaustive in his coverage of such deletions. Cubberley does not explicitly state that four consonants are the maximum for onset clusters.

• In explicit generalizations about Russian, I have never seen anything beyond claims of four consonants in the syllable onset.
  • This presentation slideshow (based on Grigorenko 2006, " If John were Ivan, would he fail in reading") says:"frequent multiple consonant clusters and syllables as complex as CCCCVC"
  • Ostapenko (2005) [cited in article]: "The possible onset in Russian can be even more complex, as it tolerates up to four consonants at the beginning of the syllable.” (I expand on this in my next point below)
  • Proctor (2009) chapter 6: "Russian allows longer clusters than most languages – up to four consonants in both onsets... and codas" (p. 126) (also, a table in chapter 1 describes a list of four-consonant onsets as "maximal onset clusters in Russian” (p. 2).
  • Russian for Dummies: "Combinations of two, three, and even four consonants are quite common."
• In less explicit generalizations, four consonants seem to normally be the maximum indicated for onsets
  • Trapman (2007) divides up Russian onset clusters by section and stops at 4 consonant sequences (p. 29). (Incidentally, Trapman stops with 3 consonants in the section on coda clusters. What a bag of worms!)
  • Chew (2010): Discussion of complex onsets and codas (beginning on p. 76); according to my limited Google Books preview (I can't see all of pages 83-99), Chew ends coverage of onsets with four consonants, saying "Finally, we need to consider four-consonant onsets..." (p. 86). There doesn't seem to be any mention of five- or six-consonant clusters.

The lone exception I have found, if I can call it that, is Kochetov (1999), who says, in passing, "The patterns attested in three- (46), four-, and five-consonant clusters are governed by the same principle and the presence of palatalized consonants in them is even more restricted" (p. 201)

As can be seen, there is (IMHO) sufficient justification to cast doubt on the claim of more than four consonants in the syllable onset in Russian. What I have found could very well be misrepresentative or even flat-out wrong, but sourcing should be used to indicate this.

Point 3: Affixation and phonological words. To put it bluntly, Dale's edits show a reliance on source synthesis to back up the claim that Russian allows more than four consonants in the syllable onset, a claim that none of the sources cited backs up. WP:SYNTH states:

Do not combine material from multiple sources to reach or imply a conclusion not explicitly stated by any of the sources. If one reliable source says A, and another reliable source says B, do not join A and B together to imply a conclusion C that is not mentioned by either of the sources.

Here is how the claims present in the article are an OR synthesis of the sources presented:

• Subclaim A: Russian has fewer phonotactic restrictions on consonants (Davidson & Roon 2008) [verified]
• Subclaim B1: Prepositions in Russian act like clitics (Rubach 2000) [verified]
• Sublcaim B2: Prepositions combine with a following word into a phonological word (Halle 1959; Bickel & Nichols 2007) [verified]
• Conclusion C: Phonological words with prepositions can have syllable onsets with five consonants. [not present in sources used for A or B)

This shows that, despite the sourcing present, one can't extrapolate the claim in question from the sources given.

In addition, Dale has rephrased the wording of a claim using a source (Ostapenko 2005) that contradicts his claim and cited it to back up claims it does not make. Again, the source in question states: "The possible onset in Russian can be even more complex, as it tolerates up to four consonants at the beginning of the syllable.”

While this statement was used in this article to back up a claim about a four-consonant maximum in the syllable onset, Dale has reworded it to: "Some maintain that there is in fact a systematic four-consonant limitation in the syllable onset of lexical words." This adds extra claims that are not backed up in Ostapenko. Here is the breakdown:

• Statement about claims in scholarship regarding four-consonant limitation [not present]
• Four-consonant limitation in onset [verified]
• Parsing of lexical words and phonological words [not present]

The first bullet point is something that I don't think Dale intended. It seems that he meant to hedge usage of Ostapenko's strong claim about onset limits with weasel wording designed to cast doubt into how representative this view is. Altering the attribution to explicitly state that Ostapenko is meant to be an example of this first bullet point would still not fix this problem; as I have shown above, her view is much more representative than "some maintain" would imply. It would also be original research, as an article's use as an example of meta-claims would be insufficient verification of such meta-claims.

The third bullet point involves Dale assuming in his reading of Ostapenko the very thing he wishes to prove (that there is a difference between onset limits of lexical words and those of phonological words). This false attribution might be more understandable had Ostapenko's examples been composed solely of lexical words. But she used no such lexical examples to reinforce this claim. Even if she had, Dale's wording still introduces novel information that she does not claim and that non-experts—even native speakers of Russian—would not see as obvious.

Point 4: Orthographic representation of epenthesis. Rubach (2000) says (p. 53) that the epenthesized vowel of single-consonant prepositions and prefixes occur when the following onset is a consonant cluster beginning with the same consonant as the preposition (barring voicing distinctions). This backs up Cubberley's (2002) more general statement about geminates in consonant clusters. Importantly, Cubberley (p. 83) states that there are two "lexically specific clusters" that this epenthesis is extended to: мн- ('me') and вс- ('all'). I believe Dale pointed out this process of vowel epenthesis to show that Russian has a method (indicated in the orthography) of dealing with difficult or awkward consonant clusters. However, the environment where these prepositions and prefixes occur with an epenthetic vowel is a very specific one, meant primarily to deal with geminate consonants in clusters and not as a general method of dealing with otherwise difficult consonant clusters.

Like I have already said, it's possible that the sources I have access to provide a skewed presentation of Russian phonotactics, but I have shown that there is reason to doubt the claim of more than four consonants in the syllable onset. This is why I have marked the claim in the article. Editors are welcome to contribute to the discussion, though I ask that you please keep them in the Discussion subsection immediately below. Thank you — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 15:35, 12 May 2012 (UTC)

Discussion

• Without even reading Part 3 (TLDR), the way forward seems obvious to me: WP:OR requires that we supply a source for reasonably contested claims. If Dale Chock wants to claim that Russian allows C⁵ clusters, then he is required to reference that claim with a RS that states that Russian allows C⁵ clusters. If he knows it's true because he's a native speaker, strict adherence to the rules would still require an independent source; if he's inferring it from grammars or dictionaries, then that would be an obvious case of OR violation. Such a claim may be deleted at any time as unsourced/OR. Certainly repeated deletion of {{ cn}} tags would be reason for a block. — kwami ( talk) 21:10, 12 May 2012 (UTC)

• I comment without specific knowledge of Russian or the literature on Russian phonology. But if (as appears to be the case from Ƶ§œš¹'s presentation above) grammars state that up to four consonants is the rule, that should be reflected here. If any sources suggest that five-consonant-clusters exist, those sources should also be cited, being careful not to give them undue weight relative to the total literature. If most of the existing literature is wrong, that's unfortunate for linguistics, philology, and Russian language studies as fields, but not really something to be remedied by Wikipedia. (Compare Timothy Messer-Kruse#Wikipedia Controversy.) Cnilep ( talk) 08:09, 13 May 2012 (UTC)

• I cannot devote adequate attention to this discussion this week. A few points raised in this section seem urgent. There is this obsequious opinion that "Russian spelling is very transparent". That this silly belief could be adopted by someone who has spent the last five years shaping this article into a tedious pronunciation manual of standard Russian just reinforces some of my previous assessments. For now, mainly I need to dissociate myself from this opinion. Changing subject, the citation in connection with this thread of a master's thesis which is devoted to comparing second language acquisition across three European languages is absurdly bad judgement. What's really bad about this is she just looked up and cited -- who? Two of the sources you quote ALONGSIDE her! Obviously you think quantity equals quality. Why you would imagine, already having Chew and Cubberley and some others here (along with Halle, etc. in the article generally), that you would have to pile on with citing some student who's only doing what you do . . . .

• Turning to Kwamikagami. I can only hope that if things come to my being blocked, it will be implemented by an administrator who hasn't spectacularly indulged in bias in this dispute. Oh, I don't want any biased administrator to involve themselves, but especially not a spectacularly biased one -- that's you; I'm not going to link to the outbursts for now, does anybody miss that? And let's discuss the idea of your weighing in on appropriate citation of sources. Although I have seen you coach and cajole others (Aesos himself somewhere, I didn't make note of which article) about giving sources, I have hardly ever seen you give a source or even make a content edit. That's an understatement. I remember sometime last year, you seemed in a hurry to delete an insertion about Turkish having some particular allophone. You and the inserter couldn't find any source, but who but me myself should have happened along. In 15 minutes of Web searching I served up to Kwamikagami a book at Google Books. The discussion is probably in his user talk page archives, but perhaps in the Turkish language talk page. Dale Chock ( talk) 08:47, 15 May 2012 (UTC)

So you don't think Russian spelling is transparent now? Or have I misinterpreted your comments? You know what transparent means in regards to writing systems, right?

As far as the quantity-quality distinction, it doesn't really hold up when there aren't any contradictory sources. But maybe you can show otherwise. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 12:57, 15 May 2012 (UTC)

Liar, you're putting words in my mouth. It's possible YOU think it's transparent. It seems just as likely that you pretend to think so out of a perceived rhetorical advantage. You seem to be attracted by the argument form, "even though I agree that Russian spelling is 'transparent', I nevertheless disagree with Dale". You seem to think that assuming that Russian spelling is "transparent" is a good launch toward your destination. Dale Chock ( talk) 06:19, 4 June 2012 (UTC)

If accusing me of bad faith instead of addressing the substance of my posts is all you care to do, then I'm not interested in this conversation. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 14:37, 4 June 2012 (UTC)

As of a few days ago, you are demonstrating your good faith by stalking my edits on articles you have no interest in and indulging in the pettiness of sarcastically imputing a false camaraderie. You only undertook this conversation (what an agony for you) after twice failing to get me discplined. I'm about to address the "substance" of your posts. No one else has been holding their breath. Dale Chock ( talk) 22:28, 4 June 2012 (UTC)

The low level of interest in this dispute from third parties over the last three weeks is noteworthy. More broadly, in the two months preceding there had been no substantive comment by third parties on disputes over this article. Even more broadly, in the long run there has been no sustained participation in this article by any editor besides AEsos.

I have already argued at length on this and probably every other issue of dispute. By contrast, my opponent did not strive to deal with my arguments until after twice failing to get me disciplined, and he has settled on this one issue, one out of half a dozen. Only one third party, Cnilep, has weighed in on the merits of the current question. Unfortunately, he too confined himself to stating a bald opinion, "X is so". He did not give reasons, and he especially did not refute reasons I gave. Since he didn't explain himelf, his contribution is unconstructive and trivial. I will argue my case as Aesos requested, then comment on some of my opponent's latest points. Given the observations I just made, it would be a waste of effort to reply point by point to AEsos's belabored presentation, half of which is redundancies. This is already too long, sorry. But enough time spent on it.

By the nature of human language, the building up of phonological words from individual phonological units is mostly compositional. The pronunciation of 'cat' is the composition of /k/, /æ/, /t/ (ignoring the phonetic transition phases between these segments, and in this case the little bit of assimilation between vowel and consonants). Although the phonetic realization of base segments is impacted by several factors (in some languages, impacted dramatically), the pronunciation of a word is by default obtained by the composition of its syllables, or of its single segments. Explaining the exceptions is a major portion of the phonology. This article already goes into great detail about the exceptions for Russian.

AEsos has still not responded to a point I have raised in previous months. By his accusation of "original research" he alleges that I am being venturesome and he is not. But I have countered that his position is also venturesome, at least as O.R. as mine might be. It's Russian spelling versus him -- some random, dilettante Wikipedian. That's not a genuine conflict matchup that I have to provide a citation for. For he is surreptitiously rejecting what Russian spelling and the phonological science on Russian language all tell us. The science tells us explicitly that the words 'k, s, v' form clusters with the following word. Halle (1959) tells us the spelling 'k vdovʲe' is pronounced 'gvdovʲe'. The spelling system gives us 'k vzglʲadu'. The spelling along with the reported exceptions are the sources. Against them, we have the phonetic sensibilities of AEsos, some random person on Wikipedia, a person who has learned the Intl. Phonetic Alphabet and whose only learning in Russian is the Russian alphabet and what he gleans from reading phonological literature.

By AEsos's and Cnilep's logic, we cannot say that the pronunciation of 'diagonalize' is 'diagonal' + '-ize', unless some "reliable source" says so. After all, it is in principle possible that of all the combinations of 'attested word' + '-ize', the language irregularly disallows some of the expected combinations by extra, arbitrary rules (or constraints, if one prefers). Of course, no scholar or organization is ever going to objectively verify the pronunciation of every one of millions of possible words of English, including 'diagonalize', 'diagonalization'. The policy WP:OR is obviously intended primarily for discoveries of new knowledge. The extension of the O.R. policy to the received spelling of a language is a real stretch. Although there are uncertainties about the whys of Russian phonotactics (combining of sounds), I don't think there are any about the whats.

I have previously pointed out that AEsos refuses to offer a positive reason why we should not take Russian spelling as a source. He declares no basis for raising a doubt, which is a different matter from demanding a source. His and Kwamikagami's stance is that no good reason is needed to demand a citation, just that you think up the objection.

Now to the second part of this reply: to address the notable points in AEsos's latest argument.

AEsos's citations, among them Ostapenko, all have a fatal flaw: they do not take notice of the uncontroversial phenomenon just mentioned, that most prepositions, including 'k, s, v', fuse with the following word, thus augmenting it from the front. Therefore, these citations are of no weight for AEsos's objections.

"Point 4: Orthographic representation of epenthesis. . . . . However, the environment where these prepositions and prefixes occur with an epenthetic vowel is a very specific one, meant primarily to deal with geminate consonants in clusters and not as a general method of dealing with otherwise difficult consonant clusters". This passage reminds me of arguments the same person has made on other talk pages where it seemed that he wasn't so much making a statement that could be deemed true or false as he just was unable to formulate his thought; so that, perhaps, what he wrote was something other than what he meant to say. Nevertheless, I am pretty sure I discern what AEsos is trying to say. It's what I've discerned all along, and what you can't get him to spit out: in his opinion it just has to be the case that pronunciations he personally objects to don't exist in Russian. There must exist additional "methods of dealing with" consonant clusters -- even though in five years of wading through authoritative sources, AEsos hasn't come up with any that bear on the consonant clusters he objects to. You see, he, some random English speaker, finds certain Russian consonant cluster combinations too "difficult". He, some random English speaker, just cannot believe that in Russian certain long consonant clusters are pronounced. In spite of him, the 'kvd' in 'kvdovʲe' ('to [a/the] widow') undergoes no vowel epenthesis and no reduction (a fact reported on a page he's read going back five years; it's cited on this talk page). Aside from all that, there is nothing "meant" by anybody, because he is referring to what the language does, not to spelling conventions (which are a deliberate creation by a limited number of individuals). Actually, I have to take back one thing I said: AEsos is not pushing original research, but "original nonresearch".

I will address again AEsos's citation of a master's thesis by Trapman, falsely representing it as a source. For her table of possible consonant clusters, all Trapman did was cite two sources that he cited. Get that? AEsos is writing an article for Wikipedia and cites some sources. Some student somewhere writes a research paper and cites the same sources. AEsos cites the student citing! Trapman's mention was a pro forma act to be expected from any college student writing a term paper. On top of that, her thesis is not about Russian, but about speakers of Spanish and Russian learning Dutch (although she postpones plainly announcing this until the second page of her introduction). Scrupulous? Good faith?

AEsos complains that I distort the quotations from Ostapenko. No, he does. In this and other articles, AEsos has specialized in taking quotes out of context: reading into them more authority and consensus than actually exists in the field; making them more categorical than the author intended. For example, as just mentioned, Ostapenko is one of those who've overlooked what happens when you stick the words 'k, s, v' in front of her examples. That's not a hypothetical. What I did with his Ostapenko citation is provide context you would never get from AEsos. (Incidentally, two days after inserting Ostapenko as a source, AEsos referred to her as 'Ostabenko'. [3] Since the 'p' and 'b' keys are far apart on a keyboard, this was not a typo, but a speech sound confusion. Are intervocalic 'p' and 'b' merging in his English? That's at least the third time in this article he has spelled the names of sources he himself inserted multiple ways; the others were 'Schenker' and 'Glovinskaja'. You know, non-American names.)

In conclusion, this article is going to improve only if additional savvy editors join in and stick with it. Dale Chock ( talk) 23:29, 4 June 2012 (UTC)

So when I provide actual sources that directly contradict your claim, and you provide zero supporting it then I am doing original research? And when you synthesize sources to form an OR conclusion it's okay because it was to "provide context you would never get from Aesos"? I suppose that's true, considering how this "context" is one of your own invention. I'm sorry, but that's not how Wikipedia works

I don't really care to respond to your mischaracterizations of my talk page behavior here. You either honestly don't understand our policies or are so bent on saving face that you would rather pretend to not understand what is asked of you than move on to other issues. Either way, it's not worth it. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 01:06, 5 June 2012 (UTC)

In Dale's most recent edit], he introduced some problematic changes. The most relevant to the discussion is the restoration of the claim that has failed verification:

• Original: "...constitutes a phonological word that acts like a single grammatical word. For example, the phrase с друзья́ми ('with friends') is pronounced [zdrʊˈzʲamʲɪ]"
• Dale's edit: "...they produce phonological words with onset clusters of five, e.g., [kfstrʲetɕe], 'to [a/the] meeting, as in [gɐtof kfstrʲetɕe] 'ready for a meeting'."

This constitutes a blatant case of WP:IDIDNTHEARTHAT. Other edits that I have undone:

• Dale has deleted tables that he believes to be uncencyclopedic in their depth of coverage. Further discussion is warranted before these removals have agreement from other editors; so far, two editors (myself an an anon) have reverted such deletions, showing a clear disagreement on the matter. As such, I have restored the tables.
• Dale has removed the sentence "However, the four-consonant limitation persists in the syllable onset" This statement is backed up by Ostapenko (2005) and Proctor (2006)
• Dale has removed the sentence " All word-initial four-consonant clusters begin with [vz] or [fs], followed by a stop (or, in the case of [x], a fricative), and a liquid"
• Without consulting the source material, Dale has changed "theoretically up to seven consonants” to "up to at least seven consonants” which doesn't even make sense.

— Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 04:06, 7 June 2012 (UTC)

Having been away from the article for a few weeks, it came to my attention only today that on 20 May, Aesos made the following amendment, thereby misidentifying case forms:

BEFORE: Because prepositions in Russian act like clitics, ^[1] the syntactic phrase of a preposition and a following word constitutes a phonological word that acts like a single grammatical word; e.g., к вдове [ɡvdɐˈvʲɛ] 'to the widow', от вдовы [ɐdvdɐˈvɨ] 'from the widow'. ^{[2] [3]} Thus, prepositions (especially the three that consist of just a single consonant: к, с, and в) contribute to phonological words with up to five consonant clusters in the syllable onset. Examples are к взгляду [gvzglʲadu] 'to [the] gaze', к встрече, 'to/for [a, the] meeting', as in 'ready for a meeting'. In the syllable coda, suffixes that contain no vowels may increase the final consonant cluster of a syllable (e.g. Ноябрьск 'city of Noyabrsk' |no'jabrʲ|+|sk| > [nɐ'jabrʲsk]), theoretically up to seven consonants: монстрств ['monstrstf] 'of monsterships'. ^[4]

AFTER: Because prepositions in Russian act like clitics, ^[1] the syntactic phrase of a preposition (most notably, the three that consist of just a single consonant: к, с, and в) and a following word constitutes a phonological word that acts like a single grammatical word. ^[3]

For example, the phrase с друзья́х('with friends') is pronounced [zdrʊˈzʲax].

In the syllable coda, suffixes that contain no vowels may increase the final consonant cluster of a syllable (e.g. Ноябрьск 'city of Noyabrsk' |no'jabrʲ|+|sk| > [nɐ'jabrʲsk]), theoretically up to seven consonants: монстрств ['monstrstf] 'of monsterships'. ^[4]

Against his usual practice, Aesos did not source this boneheaded pairing of a Russian expression and an English gloss (which confuses a gen/loc plural with an instrumental plural). His phonetic transcription wasn't even correct. Where capital 'C' means a consonant, Cья doesn't read [Cʲa], it reads [Cʲja]. The next day, a native speaker corrected the translation error by changing the Russian form to fit the English meaning; but without correcting the error in phonetic transcription.

What a brilliant reinforcement that Aesos only knows isolated sentences that he collects for citations purposes, lacking any true familiarity with his subject. Dale Chock ( talk) 21:31, 6 June 2012 (UTC)

re: "was always soft few decades ago" - "few decades ago" must be replaced with absolute time reference. (I hope wikipedia or its content will survive a few more decades :-) - Altenmann >t 09:40, 7 June 2012 (UTC)

The recent edits by the anonymous editor added some good information, though the tables didn't work very well, in my opinion. I think the problem is that there are so many words in them that they don't do a good job of quickly summarizing information. Perhaps, instead of replacing the prose, we can add simplified tables to augment it. For example:

Phoneme	Position
	V(C)	VCʲ	CV(C)	CVCʲ	CʲV(C)	CʲVCʲ
/a/	[a]					[æ]
/o/	[o]				[ɵ̞]	[ɵ̞]
/u/	[u]					[ʉ]
/e/	[ɛ̝]	[e]	[ɛ̠]	[e̠]	[ɛ̝]	[e]
/i/	[i], [ɨ]		[ɨ]		[i]	[i̝]

If we do want that last column with some comments, they should be short.

A couple of other things

• What is the source for the pronunciation of ржаной and жасмин?
• With words like этап, with an unreduced /e/, are we sure that the pronunciation is [e] and not [ɛ]?
• I had added palatalization to d in вздлить and removed it from r in Ноябрьск, based on Halle (1959), who says that only soft [dʲ] appears before soft [lʲ] and that only hard [r] appears before hard [s]. Is Halle wrong?

Let me know if I've accidentally removed citations or information in the process of removing the tables and touching up the article. I believe I kept it all in there, but I may have missed something. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 15:40, 18 June 2012 (UTC)

re: " only soft [dʲ] appears before soft [lʲ] " -- this is quite wrong. длина подлец для all have hard 'd' . In fact, I have difficulties to think of the opposite. - Altenmann >t 05:31, 19 June 2012 (UTC)

So is that just an older style of pronunciation? Halle (1959) is over fifty years old, so I can see the language changing since then. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 13:21, 19 June 2012 (UTC)

I have no idea what kind of confusion here, but d is hard in these (and was hard 50 year ago, as far as I remember my speech in my teen years :-).- Altenmann >t 01:33, 24 June 2012 (UTC)

Also, what is the point in this edit of mentioning the word взбзднуть? If the point is to provide an exception to the four-consonant limitation, we may need something more explicit, namely something that shows that the word is pronounced as it is spelled. As you can see in above discussions, this was an issue that has gotten some contentious attention. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 13:21, 19 June 2012 (UTC)

the point is that it is just a curiosity. as for pronunciation, let us wait for 10 more years; this word have just recently beeen entered some dictionaries, because taboo vocabulary was, I'd say, underresearched in the former soviet union, so officially there was no such word.:-) The only thing I can say that russsian phonology perfectly permits non-reduced pronunciation here. However this word makes me notice that the section on consonant clusters misses an important issue of modified articulation of consonants in some clusters, such as дн in взбзднуть or 'днo', and дл in 'dlina', also tl, tn. I.e. t sounds differently in tl and tk: d/t are more nasal before l and more aspirational before k. And there is much more to that, but I am not in a position to detail this, unfortunately. There are much more phonological issue to cover. For example, two massive phonological shifts: one with the advent of total literacy, another, with the advent of total 'radiofication' in the soviet union (DYK that in many remote parts of the contry 'radiofication' was done well before electrification, I guess because Stalin's words brought more light than "Lenin's lightbulb" (lampochka Ilyicha)). For example the word vodka was stably pronounced as 'votka' (and even occasionally written so in 18th century), but under of the influence of the 20th century written word the letter 'd' became slightly vocalized. - Altenmann >t 02:01, 24 June 2012 (UTC)

I think it would be better if we contextualized things. Russian phonology is not simply a collection of "curiosities." Without contextualizing it, it doesn't make much sense and even contradicts other claims made in the article. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 02:29, 24 June 2012 (UTC)

that's why I didnt put all the above in the article, even if with references. i am aware this must be done in a systematic way. btw please dont abuse my word 'curiosity'. the correct word wold be 'exception'. the english seem to love the [ of their language] and wikipedize them abundantly. - Altenmann >t 03:04, 24 June 2012 (UTC)

Nice job on the modified table, Incnis Mrsi. It looks even better than the one I proposed above. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 22:59, 21 July 2012 (UTC)

Lamentations caused by Ƶ§œš¹

• Yes, tables were too heavy (what about structured itemized lists instead?), but why you generally restored less precise and less complete old versions of "prose"? And of references? (Some examples: -ся is generally soft, not hard (as it was 100 years ago). Средняя школа is high school, not middle scholl, despite the literal meaning of words. /a/ has a special reduction not only in declension of лошадь, but also in derivations like лошадиный...) Formatting and templates do not replace validity and completeness of data! I still can't muster up the appropriate level of enthusiasm to fix all this, sorry...
• Source for ржаной and жасмин: one of the two pronunciation guides from the two dictionaries (I can't answer more precisely as they are not with me now). There were exact references in my version of the text!
• этап: a closed sound. (More closed than stressed "э" in это.)
• Halle is wrong. Assimilation in [d(ʲ)lʲ] is too archaic (and has been already archaic when his book was published!). Orthographical <ь> after paired consonants always reflects palatalization. Yes, /rʲ/+/s/ in almost all cases hardens to [rs] (январь:январский, царь:царский:царство, зверь:зверский:зверство, Тверь:тверской, море:морской etc. -- note that spelling has no <ь>!), but there are at least four exceptions: сентябрь/октябрь/ноябрь/декабрь + -ск(ий). The main Soviet holiday (the Seventh of November) was День Великой Октяб[rʲs]кой Социалистической Революции, not *Октяб[rs]кой! Less common similar cases: Вепрьский лес, Дебрьский монастырь. I guess /rʲ/+/s/ gives [rs] after vowels and [rʲs] after consonants, but it's difficult to find extra examples after consonants.

P.S. Why to refer to a German reprint of Avanesov instead of the original book? It has no new information (even as a republisher's preface), and is not better available now than the original, even in Germany. Moreover, the reference "Avanesov 1975" is somewhat confusing -- a reader may think that it is an improved edition (they do exist, 4th edition of 1968, 6th edition of 1984, but I have none). The reprint can be mentioned only as a possible back-up access method to the original edition.

Why you add parentheses in cases like кто [kto] ('who')? They are superfluous; academical sources use just single quotes. And they create problems if there are another parentheses inside these quotes.

Why to place и/ы topic above all other vowel-related text? It's not more than just a partial question, and therefore must be discussed after more general things: each vowel has its front/back variants, and и/ы is just the case with the most developed and visible distinctions. -- 68.125.55.244 ( talk) 10:55, 12 July 2012 (UTC)

Can you show me what you mean by structured itemized lists? I'm having trouble seeing such lists as quickly conveying information.

The hard pronunciation of -ся is a claim introduced by another editor, though it is technically uncited. Perhaps a fact tag is in order for the time being. I've fixed some of the other things you mentioned.

Do you have a source for what you're saying about /d(ʲ)/ + /lʲ/ and {{IPA|/rʲ/{{+/s/? It's not so much that I doubt you, just that I'd like to be thorough in our citations.

I'm not sure about Avanesov. It may have just been an issue of access for the user who added it. Perhaps they weren't aware it was a reprint.

Parenthetical glosses are the norm across Wikipedia. In my opinion, they do less to upset the structure of the prose. Any problems of nesting parentheses are easily dealt with.

I can see the ы/и issue moved down to the top of the front vowels section, though it's not much of a move. Did you have another place in mind? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 12:20, 12 July 2012 (UTC)

Russian is my native language, and some examples looks really funny for me in this article.

1) Some examples are correct, but not common (I've never met such words during my life, but you can really find them in dictionaries, it is very-very special words, or correct but not used forms): гёзы, кяриз, хянга, хэппенинг, сердчишко

2) Some examples use Cyrillic, but it is not Russian, like e.g. кок-сагыз

3) Some examples are totally not correct. These words are wrong: вздлить, встлеть. Photon82 ( talk) 09:53, 10 July 2012 (UTC)

Thanks for your input. The first class is okay to keep, since what's being talked about is kind of a rare phenomenon anyway. The other two were added by an anonymous user, and I'd assumed they were a native speaker. Can you come up with better examples? It looks like the three are designed to show a hard г before ы, and initial clusters of вздл and встл. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 12:29, 10 July 2012 (UTC)

1. Yes, all these examples were taken from dictionaries. But almost all are not so special: гёзы are known via "Till Eulenspiegel" (Charles-Theodore-Henri De Coster), кяриз -- via classical Central Asia diaries of V. A. Obruchev or via any modern book about Afghan war, хэппенинг -- via numerous Soviet critical descriptions of Western art. Finally, сердчишко is absolutely normal Russian word, used by many popular Russian writers, both classical (Dostoevsky, A. N. Ostrovsky, N. Leskov, I. Lazhechnikov, M. Pogodin, V. Nemirovich-Danchenko, V. Krestovsky...) and new (A. Bushkov, M. Semyonova, S. Vol'nov...). Only хянга was new for me.

2. The word кок-сагыз is Russian. This technical plant was very popular in 1940s-1950s as a domestic source of natural rubber.

3. These two are correct, but not common. Scroll the page up, they are discussed in the middle of the page (section "About source request for initial 4-consonant clusters"). When initially inserted in the article, both were marked as "occasional neologisms", but someone later has erased the information. I'm almost sure that there are no better examples for встл-/вздл-. -- 68.125.55.244 ( talk) 07:51, 12 July 2012 (UTC)

Here is stated that in East Slavic languages this sound doesn't appear even as positional variant. There are mentions (I don't know if they are not invented by W-pedians) that some speakers pronounce it before /k/ and /g/, but I think this must be sourced or removed from all Wikipedia (especially from non-linguistic articles as Leningrad). For me, (St. Petersburg native, 26) I pronounce функция somehow like /ˈfunt͡sɪə/. Ignatus ( talk) 21:24, 15 December 2013 (UTC)

Jones (1969) states that Щ is long or short "voiceless palato-alveolar fricative" that is in standard IPA transcription it's /ʃ(ː)/. Ч is т + щ (short) hence it should be /tʃ/.-- Lüboslóv Yęzýkin ( talk) 08:10, 3 January 2014 (UTC)

I recently added some tables of character names in Russian and romanization with accent marks to The Brothers Karamazov and Crime and Punishment. This should be somewhat helpful to readers, but it would be much more helpful if reduction of unstressed o and e were shown, because English does not consistently reduce o and e to a and i the way standard Russian does. That is, it would be most helpful to transcribe Alekséi Fyódorovich Karamázov as Alikséi Fyódaravich Karamázav and Katerína Ivánovna Verkhóvtseva as Katirína Ivánavna Virkhóvtsyva.

Does anyone know if there's an accepted romanization of Russian, perhaps one used in teaching, that shows vowel reduction in this way (none is listed in that article, at least), or would transcribing vowel reduction in the character names be WP:OR? — Eru· tuon 00:23, 5 April 2015 (UTC)

I've never heard of that. There's not actually a hard and fast OR rule when it comes to in-house conventions for transcription (that's why we can get away with our virtually unique diaphonemic transcription system for English), though you'd have an uphill battle in getting your way on romanizing Russian that way, I think. The best place to bring it up is at Wikipedia talk:Romanization of Russian. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 02:46, 5 April 2015 (UTC)

Considering that not all Russian dialects have vowel reduction, I'm not sure that this is really such a great idea. Why not simply stick to the traditional, essentially diaphonemic, scientific transliteration, from which the pronunciation can be easily derived for every dialect? -- Florian Blaschke ( talk) 02:16, 4 September 2016 (UTC)

The problem with the traditional transliteration is that it's just as opaque as the standard spelling. There is very little representation of phonological processes, so you have to learn a lot of rules to figure out how to convert from spelling to pronunciation, even if the stress is marked. Having phonological processes marked makes it much easier. (I created a set of examples in my userspace).

Is standard spelling diaphonemic? I really don't know. There may be unpredictable splits and mergers, as there are between British and American English. Russian dialects doesn't give detailed enough information. And I don't have any books... — Eru· tuon 04:21, 4 September 2016 (UTC)

AFAICS, Latin-based transcriptions aren't supposed to show phonological processes, nor pronunciation, directly, and especially in this case they can't even really do that readily. Unless you introduce so many diacritics that you might as well use IPA, and that's exactly what I'd recommend to use to show pronunciation.

Ideally, of course, you'd have sound files, for those who can't figure out the IPA, don't have enough phonetic training to reproduce the native Russian pronunciation – though most readers will likely be contented with a rough approximation, because they don't care about seemingly irrelevant nuances –, and have no idea about Russian phonology and phonetics in the first place (if you have, the traditional Latin-based transliteration is of course enough). Many US readers in particular, I fear, won't find transcriptions like on your user subpage very helpful (hardly more helpful than IPA, which we keep getting complaints about) and may not be contented with anything short of Anglicised respelling. What's the audience you're aiming at? -- Florian Blaschke ( talk) 22:38, 4 September 2016 (UTC)

One don't need to read all IPA letters for a particular language. Maybe, we should put more links to IPA for Russian page, where are examples from English word for every sound used in Russian. Tacit Murky ( talk) 11:33, 5 September 2016 (UTC)

@ Florian Blaschke: Well, I came up with the idea when a friend of mine wondered how the names in The Brothers Karamazov were pronounced. I used a similar spelling system (though I didn't distinguish alveolopalatals and retroflexes, or palatalized consonants and consonants followed by /j/), and she found it very helpful. Since she did, I assume at least some other people would, though I suppose it depends how linguistically aware they are. — Eru· tuon 20:29, 9 September 2016 (UTC)

Hmm. What's her educational background and origin? Americans are notorious for having trouble with "Continental" vowel values, which is probably because they are used to English-based transcriptions rather than IPA-based ones. -- Florian Blaschke ( talk) 21:20, 10 September 2016 (UTC)

She's American, like me, and has studied Latin and Ancient Greek. That would make her much more familiar with "Continental" vowels than most Americans. Though, I'm curious if more Americans are familiar with "Continental" vowels than before because of the growing importance of Latin American Spanish... — Eru· tuon 21:25, 12 September 2016 (UTC)

Could you please add a table that shows how the respective consonantal - but possibly also vocalic - phonemes are represented in writing? Especially those that are sometimes represented by combinations of letters. — Preceding unsigned comment added by 93.206.129.80 ( talk) 03:50, 9 September 2015 (UTC)

Is what you're looking for at Russian alphabet? — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 06:38, 9 September 2015 (UTC)

I think these words are pronounced as [kəəpʲɪˈrat͡sɨjə] and [səəbrɐˈʐatʲ], unlike what the article says starting at "Adjacent to a hiatus, when the same sound occurs on both sides of the hiatus..." -- Anatoli ( talk) 07:13, 14 October 2015 (UTC)

Also:

1. нецелесообра́зный [nʲɪt͡sɨlʲɪsəɐbˈraznɨj] and "авиаотря́д" [ɐvʲɪəɐtˈrʲæt] - only the second "о" (1st word) and the only "о" (2nd word) is pre-tonal ([ɐ]), the vowel before it, is further away from the word stress ([ə])
2. ааро́новец [ɐɐˈronəvʲɪt͡s] and ааро́новщина [ɐɐˈronəfɕːɪnə] - the first "а" is word initial, so the first vowel should be [ɐ].

Basically, the pronunciation of two non-iotated vowels together should work the same way as if they had a consonant between them. -- Anatoli ( talk) 23:41, 15 October 2015 (UTC)

Does the soundfile given in the article (I've included it here for convenience) sound wrong to you? To me, it sounds like the person is saying [sɐɐ-] or [sɐː-] rather than [səə]. — Eru· tuon 23:57, 15 October 2015 (UTC)

The claims in question are sourced, though the sources in question are pretty old. If you can find a newer source that backs up what you're saying, feel free to provide it. Otherwise, I we should not make a changes based on original research impressions. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 06:22, 16 October 2015 (UTC)

«Vowels may become voiceless between two voiceless consonants». OK, but none of those 3 examples actually demonstrate it. Moreover, there is a clear distinction between опыт and опт. I can't find dialect-agnostic examples for such devoicing, unless it is a form of relaxed pronounce.

About «Phonological descriptions of /j/ may also classify it as a consonant even in the coda»: after 1970-s modern understanding of /Й/ is as a consonant; before it was considered to be semivowel. So, no diphthongs then.

(More later.) Tacit Murky ( talk) 04:33, 15 May 2016 (UTC)

There is a special case for initial /и/ (including conjunction «и» = and). Since (grammatically) this soft vowel is the only one (of 5 vowel letters), that isn't yotting in the word-starting position, in order to preserve hardness of preceding consonant, it can be changed into /ы/. It's obligatory after always-hard consonants (ш-ж ц) and (apparently) alveolar ones (т-д с-з л н р). Other consonants may be optionally softened:

• Дом и сад - [dom ɨ sat] or [domʲ ɪ sat];
• Сад и дом - [sad ɨ dom].

With that case, we should update vowel allophony table. Tacit Murky ( talk) 23:47, 25 July 2016 (UTC)

I don't understand how what you've described goes against the vowel allophony table as it is. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 02:47, 26 July 2016 (UTC)

I suggest that «V, CV» cell for /ɨ/ line should be updated to «C V, CV», because there is no «ы» word in Russian (unless name of the letter itself is meant), but there are a lot of cases with «…@ и…», where @ is a hard consonant. Moreover, this — «When a preceding consonant is hard, /i/ is retracted to [ɨ].» — is true only for above-mentioned set of hard consonants. Others may be softened. You can listen to colloquial phrases with «… и …» on Forvo. Tacit Murky ( talk) 14:11, 26 July 2016 (UTC)

Sorry. I think you're mistaking phonology and orthography. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 20:14, 26 July 2016 (UTC)

No, because orthographically И is changed into Ы only if it's a first root vowel, preceded by prefix with its last consonant being hard: имя (name) - безымянный (nameless, anonymous). Root «им» is spelled as «ым» explicitly because of без- prefix. But that's not the case with cross-word morphophonology: «Сад и дом» is spelled as /сад ы дом/» (I've corrected above example). Neither «V» or «CV» is not the case here («C V» is), unless this notation is supposed to be irrelevant to word-boundaries (which is probably not so, otherwise other combinations of V's and C's wouldn't be so diverse). But even with that in mind, «V» is not supposed to be there, because there is no Ы word. Tacit Murky ( talk) 21:38, 26 July 2016 (UTC)

You're a little confusing because you're talking about phonology and morphophonology but using orthographic representation. But I see what you're saying. [ɨ] doesn't normally appear as a standalone vowel unless it follows a word boundary and the preceding word ends in a hard consonant. That does seem like an important distinction to make. What do we make of those rare instances of word-initial [ɨ] that aren't in this same context? — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 01:28, 27 July 2016 (UTC)

Well, absolute word-initial [ɨ] (= explicit «ы») is possible in case of: letter name, onomatopoeia (Ых… Ы-ы-ы!), some non-Slavic toponims (village names in Russia's Yakutia — Ытык-Кюёль, Ыныкчанский), and maybe something else. Theoretically, that's enough to call it a rare case of non-allophonic [ɨ]. Otherwise it can be something like «человек без имени» with initial И pronounced as [ɨ]. But there is a problem: some hard consonants are softable by a following /i/ (if crossing a word boundary, but not inside the word). Largely: /k-g x/ («итог игры» - gʲɪ); optionally: /b-p v-f m/. Not sure if there are any sources for that… Tacit Murky ( talk) 03:05, 27 July 2016 (UTC)

There are. But that variation is more about morphophonology than phonology. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 14:13, 27 July 2016 (UTC)

Right, but still we should mention it somewhere, isn't it? Tacit Murky ( talk) 17:00, 27 July 2016 (UTC)

Yes, and then there is a case of pseudovowels forming pseudosyllables. This is happening more often than in English because of long consonant clusters. Usually obstruents (/к-г т-д б-п/) act as a semi-syllable bound, forming additional unstressed syllable with sonorant pseudovowel (usually /р л/, less often /н м/): контрпример - кон·тр-при-мер, корабль - ко-ра·бль, ри·тм, рок-н-ролл. Eng. example: battle ['bæ·tl]. Here is a survey with more examples from Russian poetry (where number of syllables have to be preserved to keep poetic metre): джен·тль-мен (verse by Луговский), ок-тя·брь (3 syllables in Pasternak's piece, but 2 in Pushkin's «Осень»), rhyme ру·бль - у-быль (by Саша Чёрный). There are even interjections like кс-кс-кс (to attract a cat, equivalent of «here-kitty-kitty»), where [s] is forming a real syllable (lacking actual vowel); Eng. EQ: «psst!». IMO that's important enough to mention. Tacit Murky ( talk) 17:00 — updated on 18:10, 27 July 2016 (UTC)

The German article mentions the occurrence of automatic phonetic glides between vowels and consonants (and vice versa), similar to the situation in Irish, and this is consistent with my experience – especially a stressed [o] can almost sound like a diphthong [u͡ɔ] (more like [ᵘɔ]). Do RS make no mention of this phenomenon? -- Florian Blaschke ( talk) 02:28, 4 September 2016 (UTC)

A remark on this phenomenon is buried in the Supplementary notes in the Phonological processes section. — Eru· tuon 03:45, 4 September 2016 (UTC)

Thanks, I missed that. -- Florian Blaschke ( talk) 22:26, 4 September 2016 (UTC)

So, vowel table states that ё for all «CʲV» syllables produce [ɵ]. I'd counter that: Лёля and тёмин may produce [ɵ] (non-iotted counterpart for [o]) in a fast and/or relaxed pronunciation, but that's a «CʲVCʲ» case; meanwhile, Пётр and алё clearly have [o]. So it's like ю case. (Removed about «JV» below…) Tacit Murky ( talk) 23:34, 9 September 2016 (UTC)

I hear a centralized [ɵ] in this recording of Tchaikovsky's name... Maybe o between two iotated consonants is even more front than central? — Eru· tuon 23:48, 9 September 2016 (UTC)

@ Tacit Murky: Please provide a source for that. Also, I'm not sure what ⟨-йя⟩ has to do with the [ɵ] allophone.

@ Erutuon: According to German Wikipedia, Russian central rounded vowels can in some cases be phonetically front, but there's no citation for that. Mr KEBAB ( talk) 03:28, 10 September 2016 (UTC)

Unfortunately, all I got is Forvo's examples, but there is a lot of them. Пётр (many cases) may sound like [ɵ] by some speakers, however, we should compare that to true «CʲVCʲ» cases: Лёля, Колёсико, Кёнигсберг (originally Königsberg with a true [ɵ]). So, yes, it does sound more fronted than in «CʲV(C)» position, closer to German-ish [ɵ]. (Case of «JV» should be replaced in this new section below.) Tacit Murky ( talk) 13:23, 10 September 2016 (UTC)

All of those sound more or less central to me, so [ɵ] is a perfectly appropriate transcription. Also, remember that we describe Russian vowels as they are pronounced by educated Moscow/St. Petersburg speakers. Such an accent may or may not coincide with regional standards of pronunciation, such as Ukrainian Russian or Belarussian Russian ( here two speakers are from Ukraine). Mr KEBAB ( talk) 13:35, 10 September 2016 (UTC)

Yes, they all are somewhat central; but then how would you mark up traditional German/Scandinavian ö/ø? It must be even more fronted, and Russian «CʲVCʲ» are seemed to be closer to those, than «just central» [ɵ]. Tacit Murky ( talk) 13:42, 10 September 2016 (UTC)

Let's not confuse IPA and orthography. I assume you mean the close-mid front rounded vowel [ø]. Two things:

- Please provide a source that confirms its existence in Russian (German Wikipedia is not one).

- We don't and don't have to go for the fully narrow transcription. There are levels of narrowness to the phonetic transcription. [ɵ] is narrow enough for our purposes. I think you're looking for a problem where there simply isn't one. Sorry, I've just realized that the discussion is about a table where a fully narrow IPA is called for. My bad. In that case, just look for a source and we can put it there. Mr KEBAB ( talk) 13:58, 10 September 2016 (UTC)

Actually, «CʲёCʲ» much better correspond with Close-mid front protruded vowel (which doesn't sound so close to ü [ʉ]) — listen to Swedish «öl» there, which is pretty close to Лёля above. Yes, I'd better find some source… Tacit Murky ( talk) 15:42, 10 September 2016 (UTC)

As of now, I can only find this work with frequency of occurrence analysis for various vowel phonemes. Russian WP-article links there and has a very different looking table. Tacit Murky ( talk) 20:21, 10 September 2016 (UTC)

Yes, please.

Thanks for the link. You mean allophones? Russian has only 5 (or 6, depends on how you count them) phonemes, and none of them are */ɵ/ or */ø/. I can see that the front rounded vowels are labelled with "1". Does that mean "occurred once in our study"? If so, I don't think it's worth including, they have to be more common than that.

Well yes, the table does look different, but there's not a single mention of front rounded vowels there. Can we stay on topic, please? Mr KEBAB ( talk) 23:57, 10 September 2016 (UTC)

Sorry, yes, I mean allophones. Unfortunately, Степанова et al. don't show occurrences of |ё|, but table 9 shows 7 occurrences of [ɵ] for «ой» combination. (Same table also show almost total absence of [j] for «ий» and «ый» cases, but not for «ой», as described by Бондарко in «Фонетика современного русского языка»; may be worth noting in consonants section here.) Front rounded vowels are absent in the Russian WP-table, but there are central round ones, probably because „Close-mid front protruded“ is unrecognized by russists even on allophonic level. Page for [ө] shows it for «тётя» (that is proper «CʲVCʲ») as [ˈtʲɵtʲə]; Russian Wiktionary spells more precisely: [ˈtʲɵ̞tʲə]. However, when listening to audio example of [ө] for IPA, it's clear to me that it sounds closer to [ʉ], and vowel in тётя is much closer to [ø̫] as for Swedish öl (audio on English page for ø). But my initial point was that sound of |ё| for «CʲV(C)» is much closer to [o], and no near [ø̫]. However, is still shows as [ө] like in мёд. So dispute is between [ɵ̞]/[ø̫] and „general“ [ө]. Tacit Murky ( talk) 16:58, 11 September 2016 (UTC)

If you want to stay in the realm of subjectively analysing audio recordings then I must say 'let's stop here'. Find sources first. Mr KEBAB ( talk) 23:50, 11 September 2016 (UTC)

A quick index of vowel pronunciation

Phoneme	Position	Letter (typically)	Stressed	Reduced
/a/	V, CV	а	[ ä, [ ɑ	[ ə, [ ɐ
	CʲV(C)	я	[ ä	[ ɪ
	CʲVCʲ		[ æ
/o/	V, CV	о	o̞ ~ ɔ	[ ə, [ ɐ
	CʲV	ё*	[ ɵ	[ ɪ
/e/	CʲV(C)	e	[ ɛ̝
	VC	э	[ ɛ
	CVC	э, e		[ ɨ̞
	CVCʲ		[ e
/u/	V, CV	у	[ u	[ ʊ
	CʲV(C)	ю
	CʲVCʲ		[ ʉ	[ ʊ̈
/i/†	V, CʲV	и	[ i	[ ɪ
/ɨ/†	V, CV	ы, и	[ ɨ	[ ɨ̞
* Reduced ⟨ё⟩ is written as ⟨е⟩. † Distinction is disputed.

Coreydragon replaced line breaks in the table of Russian vowel allophones with table cells. Aeusoes1 reverted, saying that taking out line breaks implied there was variation that is not there. I'm puzzled; Coreydragon's version (which I pasted here) looks okay to me. Could you explain, Aeusoes1? — Eru· tuon 06:27, 10 September 2016 (UTC)

Before I edited, and now again that it has been reverted, the table has, within the same cell, two entries separated by a line break, and further over in the same row, another cell with the same arrangement (this happens twice; the second time there are three such cells). This looks ambiguous, viz. are the entries along the same line to be taken as corresponding to the entries further over? or are the cells to be taken as a whole (iow, variable pronunciation)? In the case of the former, the rows must be divided, as I have done, like the other such cases in the table. In the case of the latter, which is, to my understanding, what Aeusoes1 is claiming as the case, then there should be no line breaks, and a comma (or tilde) separating the two variant pronunciations, and furthermore the environments {CʲV(C), CʲVCʲ}, if they are to be taken together, should be listed simply as CʲV(C), non? That the two are distinguished in this way would seem to imply that my interpretation is correct. More importantly, the text of the article says, with regard to these, of /a/ 'Between soft consonants, /a/ becomes [æ]', and of /u/ 'As with the other back vowels, /u/ is centralized to [ʉ] between soft consonants'. These statements support my version of the table. If I am wrong, then the prose needs to be stricken and the table should be changed as I have described (viz. to remove the misleading line breaks, replacing them with commas between phonetic realisations and simplifying the environments). (The current version of the table is reproduced, below my own, for the convenience of the reader)

A quick index of vowel pronunciation

Phoneme	Position	Letter (typically)	Stressed	Reduced
/a/	V, CV	а	[ ä, [ ɑ	[ ə, [ ɐ
	CʲV(C) CʲVCʲ	я	[ ä [ æ	[ ɪ
/o/	V, CV	о	o̞ ~ ɔ	[ ə, [ ɐ
	CʲV	ё*	[ ɵ	[ ɪ
/e/	CʲV(C)	э, e	[ ɛ̝
	VC		[ ɛ
	CVC			[ ɨ̞
	CVCʲ		[ e
/u/	V, CV	у	[ u	[ ʊ
	CʲV(C) CʲVCʲ	ю	[ u [ ʉ	[ ʊ [ ʊ̈
/i/†	V, CʲV	и	[ i	[ ɪ
/ɨ/†	V, CV	ы, и	[ ɨ	[ ɨ̞
* Reduced ⟨ё⟩ is written as ⟨е⟩. † Distinction is disputed.

I agree that Coreydragon's version looks less ambiguous. CʲV(C) and CʲVCʲ are different cases. Tacit Murky ( talk) 13:31, 10 September 2016 (UTC)

My reading of the changes were that, for example, /a/ appearing between or after a soft consonant could be either [ä] or [æ]. This free variation is not something I've seen in phonetic literature on Russian. I understand why Coreydragon was confused, since the "letter" column interrupts what would otherwise be adjacent cells. Perhaps we could move the "letter column to the left or right. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 01:45, 11 September 2016 (UTC)

I don't understand. Corey's table looks to me like it's saying that /a/ (spelled я) in the environment CʲV(C) is [ä] and /a/ in the environment CʲVCʲ is [æ]: no free variation. It would imply free variation if [ä] and [æ] were in the same cell, as they are in the version you reverted to! — Eru· tuon 02:50, 11 September 2016 (UTC)

Oh [expletive deleted]. My bad, guys. I opened up two new tabs to compare the two different versions and got them mixed up. I'll revert my revert. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 06:12, 11 September 2016 (UTC)

As long as it all worked out in the end! No worries Coreydragon ( talk) 07:02, 11 September 2016 (UTC)

♦Here's more relevant place for my above-mentioned proposal: in all cases «C» can be any consonant, except Й [j]. Explicit «JV» may produce additional effects (like -йя = [jːə]) apart from implicit iotting like «V» after «V», ь, ъ, «-» and word-initially. Therefore, it may be better to differentiate these cases, when dealing with iotted soft vowels. Tacit Murky ( talk) 13:31, 10 September 2016 (UTC)

♦IMO, it's incorrect to put «э, е» in the same cell. We have 4 cases here:

1. CʲV(C): impossible for э; correct for е as [ɛ̝];
2. VC: correct for э as [e]/[ɛ] (not sure…); impossible for е, as initial letter Е must be iotted, so it should fall for above case, unless we distinguish between them and put explicit «JV(C)» case;
3. CVC: correct for э as [e]/[ɛ]; rare for е, as it would soften preceding consonant, except cases noted below;
4. CVCʲ: same.

Note: always-soft й, ч and щ are never followed by э, even in loans. In recent (not fully rusified) words е after consonant is often pronounced as э, just like after always-hard ш, ж, ц: тест (loan), железо (native). But there are no cases where э is pronounced as е (with or without iotting). Hence, we should put a particular letter for at least 2 of 4 rows. Tacit Murky ( talk) 16:45, 10 September 2016 (UTC)

Like this? (See the table at the top of this section, where I made the change.) [edit:] Oops, there we go. — Eru· tuon 19:54, 11 September 2016 (UTC)

Aha, that's it. And then there is a notion on iotting: which of these cases is for «ель» = [jɛ̝lʲ] ? Assuming «Position» means «…for/of a phoneme» (not a letter) — should be «CʲV(C)» for «е», where Cʲ=[j]. But in plain text iotting is almost always implicit (there's no Й). AFAIK, some linguists deliberately distinguish [j] from rest of consonants, by showing cases like «JV» for «е/ё/я/ю». So, a subtle difference can be seen between allophones of /u/ for Ю́ля and пилю́ля — first isn't exactly [ʉ] (despite [j] being soft consonant). Tacit Murky ( talk) 22:01, 11 September 2016 (UTC)

@ Tacit Murky: /e/ in the CVC environment can only be [ ɛ, not [ e (which can only appear before and between soft consonants).

We don't need to treat /j/ as separate from other consonants. Once again, you're mistaking orthography (⟨ё, я, ю⟩, which represent the sequences of phonemes /jo, ja, ju/) for phonetics.

So, a subtle difference can be seen between allophones of /u/ for Ю́ля and пилю́ля — first isn't exactly [ ʉ (despite /j/ being soft consonant). With all due respect, this talk page is not your personal blog (see above). The sources say they're the same, so we treat them as if they were so. If you want to change that, look for sources, or else this talk page will be flooded with your OR interpretation of Russian sounds, even though reliable sources say something else. That's totally counterproductive. Mr KEBAB ( talk) 00:00, 12 September 2016 (UTC)

I'm not mistaking orthography, just dropping intermediate phonemic presentation. I assume you don't oppose changing table rows for /e/ as described? I also agree to exchange «Letter» and «Position» columns. Tacit Murky ( talk) 01:31, 12 September 2016 (UTC)

I don't. There's no need to exchange those collumns, and it'd make the table somewhat harder to read. Mr KEBAB ( talk) 02:55, 12 September 2016 (UTC)

	Phoneme	Letter (typically)	Position	Stressed	Reduced
"Letter" and "Position" switched	/a/	а	V, CV	[ ä, [ ɑ	[ ə, [ ɐ
		я	CʲV(C)	[ ä	[ ɪ
			CʲVCʲ	[ æ
	Letter (typically)	Phoneme	Position	Stressed	Reduced
"Letter" moved all the way to the left	а	/a/	V, CV	[ ä, [ ɑ	[ ə, [ ɐ
	я		CʲV(C)	[ ä	[ ɪ
			CʲVCʲ	[ æ

To me, the result looks easier to read. See what it looks like on the right. I also tried moving the letters all the way to the left, but that definitely makes it harder to read (though it is more logical: letter → phoneme → environment → allophones). — Eru· tuon 07:19, 12 September 2016 (UTC)

Woooah, talking about confusion:

- Aeusoes1 proposes moving the 'letter' column all the way to the left

- Tacit Murky makes a different proposal (swapping 'position' for 'letter')

- I somehow manage to misread TM's proposal (thinking 'position' actually read 'phoneme')

- You bring back Aeusoes1 proposal and talk about it.

OK, so I have nothing against TM's proposal, but I'd rather keep the 'phoneme' column as the leftmost one. Mr KEBAB ( talk) 07:56, 12 September 2016 (UTC)

Huh, I didn't notice Aeusoes1 talking about moving "letter" to the left. Well, we have three people (you, me, and Tacit Murky) who like the first version ("position", "letter" → "letter", "position") switched, so it can probably be added to the article. — Eru· tuon 08:57, 12 September 2016 (UTC)

Oh yes, your second proposal was a (accidental) mix of TM's and Aeusoes1's proposals. My bad. Mr KEBAB ( talk) 09:11, 12 September 2016 (UTC)

I'm fine with TM's proposal. My suggestion of moving the letter column to the left was because it seems to interrupt the thought process of "here's the phoneme" "here are the positions we are talking about" and "here are the pronunciations." But that was to bypass confusion that I think was introduced more strongly with edits that have since been reversed. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 15:09, 12 September 2016 (UTC)

Or should it start with «here is the letter I want to pronounce right» ? However, article section is about phonemes… And yes, «Letters» as leftmost column makes harder to read the rows. Tacit Murky ( talk) 15:45, 12 September 2016 (UTC)