Talk:Lithuanian phonology

Consonants
As far as I can see, most of this article is based on an obsolete work, thus needs a complete update. For instance, Lithuanian has no post-alveolar consonants, thus the opposition "/t͡ʂ, d͡ʐ, ʂ, ʐ/ are hard counterparts of /t͡ʃ, d͡ʒ, ʃ, ʒ/" is false. "/ɫ, lʲ/ are laminal denti-alveolar [ɫ̪, l̪ʲ]" is also not correct: [ɫ̪] is dental, while [lʲ] is alveolar. This should be reflected in the table of consonants.--Ąžuolas (talk) 20:08, 8 August 2015 (UTC)
 * It shouldn't, as the opposite is stated by the sources. Peter238 (talk) 20:34, 8 August 2015 (UTC)
 * There is only one obscure source written in 1964, directly contradicting the Grammar of Modern Lithuanian, published in 1997 by the institute of the Lithuanian language and later republished, the quoted linguist and phonetician Aleksas Girdenis, as well as the main academic source of Lithuanian phonology, published in 1995 by Antanas Pakerys and later republished.--Ąžuolas (talk) 21:13, 8 August 2015 (UTC)
 * Directly contradicting? So what do those books say? Peter238 (talk) 21:23, 8 August 2015 (UTC)
 * In the book of A. Pakerys "Phonology of modern standard Lithuanian", page 93, the author classifies [ʃ, ʒ] as: sibilants (pučiamieji); laminal (liežuvio priešakiniai); alveolar (alveoliniai); voiceless/voiced (duslieji/skardieji); hard (kietasis); for [ʃʲ, ʒʲ], the only difference is palatalization. All of these works use the traditional Lithuanian transcription: [š], [š’], [ž] and [ž’]. Nevertheless, the Grammar, published by the institute of the Lithuanian language, even has a translation table between IPA and the traditional transcription (pages 17–19). It shows that the corresponding signs are [ʃ], [ʒ], [ʃʲ] and [ʒʲ].––Ąžuolas (talk) 21:54, 8 August 2015 (UTC)
 * Done. The article was actually correct all along. Peter238 (talk) 22:21, 8 August 2015 (UTC)
 * can you give me the original titles of the books you talked about? Peter238 (talk) 11:36, 10 August 2015 (UTC)
 * Both major sources have already been already quoted, except for the Grammar: "Dabartinės lietuvių kalbos gramatika. Vilnius, 1997 m."
 * I have corrected the table so that it is fully compliant with the major modern Lithuanian sources of phonology.
 * Several claims in the current version have been marked as dubious, since modern sources state otherwise: /nʲ/ is palatalized laminal alveolar [n̻ʲ] – it is not, it’s dental, as is every palatalized dental in Lithuanian (Pakerys, page 90). Hard /t͡ʃ, d͡ʒ, ʃ, ʒ/ are velarized [t͡ʃˠ, d͡ʒˠ, ʃˠ, ʒˠ] and slightly labialized [t͡ʃʷ, d͡ʒʷ, ʃʷ, ʒʷ] – no velarization or labialization is mentioned in any phonology works, and the reason is that this would entail a vowel change as in Slavonic languages, which doesn’t happen in Lithuanian.--Ąžuolas (talk) 18:57, 10 August 2015 (UTC)

Thanks for the title. The source that says that are velarized and labialized is modern, as it's from 1997. We'd need a source that says that they are not velarized or labialized. The statement that velarization/labialization would trigger a vowel change is just WP:OR on your part, we don't know that. Peter238 (talk) 19:29, 10 August 2015 (UTC)
 * And re-adding the dental diacritics to the text is hardly an improvement. Only the sibilants contrast dental-alveolar POA, and the dentals aren't even purely dental - the sibilants are laminal post-dental (dentalized laminal alveolar), whereas the rest is laminal denti-alveolar (the tip of the tongue touches the back of the upper teeth, whereas the blade touches the gum behind the upper teeth). Peter238 (talk) 19:33, 10 August 2015 (UTC)
 * I’m holding the Lithuanian version of that source. It does say that any Lithuanian consonant is labialized before any back vowel, not just [ʃ ʒ t͡ʃ d͡ʒ] (page 29, §25(d)). But if we go as deep as that, we should describe every alophone instance, not just those four.––Ąžuolas (talk) 19:57, 10 August 2015 (UTC)
 * Yes, they don’t contrast inside the language, but what’s the harm? You have already removed this indication from Help:IPA_for_Lithuanian, stating that one should look for information in this article, and yet here we can’t put the dental sign even here. This indication is required to make a difference with other languages, especially English.––Ąžuolas (talk) 19:57, 10 August 2015 (UTC)
 * I'll take a look at the English version once again and will get back to you shortly. All I know is that on page 36, the book says exactly that: Lithuanian hard alveolar sibilants are velarized and somewhat labialized, there's nothing about back vowels.
 * There's no harm really, it's just that the diacritic is redundant, especially since we've already stated their actual place of articulation. As I already said, $⟨⟩$ in official IPA may be dental, denti-alveolar (well, not in the case of $⟨⟩$), alveolar or postalveolar, depending on the language/accent. They are not alveolar by definition. Peter238 (talk) 20:34, 10 August 2015 (UTC)
 * As for the the alveolar sibilants, please indicate the paragraph, since pages might differ from the translation.
 * Considering the variety the IPA characters stand for, what is the point for anyone to choose the most umbiguous version of one character when we have the means to do otherwise? It brings an unnecessary umbiguity to the article, creates a mess for people who study and compare different languages and, most importantly, diminishes the utility of IPA notation, since, after all, the pronunciation of any Lithuanian word can be easily derived from the regular spelling itself.--Ąžuolas (talk) 21:06, 10 August 2015 (UTC)

The relevant paragraph reads as follows: "The hard (non-palatalized) consonants (especially ) are characterized not only by the absence of palatalization, but also by velarisation, i.e. by raising of the back part of the tongue towards the soft palate (velum). Besides, the hard are slightly labialized." I can't find anything on the page 29, and I'm not sure what "§25(d)" means, as it doesn't seem to appear in that book. Maybe we are talking about different books after all? The book I'm talking about is Lithuanian Grammar by Ambrazas et al., and its ISBN is 9986-813-22-0. Peter238 (talk) 22:36, 11 August 2015 (UTC)
 * Yes, we are talking about the same book, just different language versions. The source states that the velarization concerns every non-palatalized consonant, not just those three, and the allophonic labialization occurs before every labialized vowel (see two pages later, point (d)). The source sais nothing about [t͡ʃˠ] [d͡ʒˠ], nor does it offer any phonetic representation of this phenomenon.--Ąžuolas (talk) 18:13, 13 August 2015 (UTC)
 * Ok, done. The labialization mark $⟨⟩$ says nothing about the amount of the labialization, if that's what you're talking about. If you're talking about using $⟨⟩$ in general, there is nothing OR about it. We're using IPA, and that's how it transcribes labialization. Peter238 (talk) 20:04, 13 August 2015 (UTC)
 * The reason why one has to be finicky and careful about these things is that they might lead to a confusion. The current notation "Hard /ʃ, ʒ/ are velarized [ʃˠ, ʒˠ]" gives an impression that Lithuanian has some kind of a velar offglide (see Russian phonology, Irish phonology). Besides, the book states that every hard consonant is velarized, just [ʃ] and [ʒ] slightly more than others.--Ąžuolas (talk) 20:38, 13 August 2015 (UTC)
 * No, the transcription does not imply any kind of an offglide. You might be thinking that because of the rather broad transcription systems of Russian and Irish that we use on Wikipedia. In official IPA,  stands for plain velarization, nothing more. And you're right, the source does say that all of the hard consonants are velarized. I'll update the article accordingly, thanks. Peter238 (talk) 21:01, 13 August 2015 (UTC)

Reverts
If we want to create a separate section on the orthography on this page, I am fine with that, but the reason why I am doing this is because that is a phonology section only. Fdom5997 (talk) 04:34, 26 August 2020 (UTC)
 * Take a look at Kazakh and Romanian phonologies. They have same thing. Why don't you revert those changes then? IdkGoodName (talk) 14:43, 26 August 2020 (UTC)
 * An interface between phonology and spelling is always useful information, either in the phoneme chart itself, or in a separate table. People want to relate the transcribed phonemes with the words they know. Lithuanian spelling is quite regular, so the phoneme chart does not get overblown by it. (Obviously we couldn't do such a thing with English.) –Austronesier (talk) 14:55, 26 August 2020 (UTC)
 * True. I support the addition. Sol505000 (talk) 10:36, 2 September 2020 (UTC)

Pitch accent
Can someone go through the article and add standard pitch accent markings where they're missing? I'm deaf to pitch accent myself and I'd probably make some mistakes. Sol505000 (talk) 10:13, 2 September 2020 (UTC)

Vibrant R
I dunno if it's my brain being stupid and trying to learn consonants and vowels of other languages made me pronounce R as tap R in certain words, or Lithuanian indeed has vibrant R. I live in Northern part of Lithuania, so Polish influence to my dialect is out of question. Anyone also noticed tap R in their speech? Most likely fast speech, but I am not certain. IdkGoodName (talk) 18:18, 2 September 2020 (UTC)
 * I use both to be honest, with the tap being predominant in my speech (AFAIK, I don't monitor it that carefully). Either way, again, we need a source. Sol505000 (talk) 08:09, 3 September 2020 (UTC)
 * I mostly use short trill, while tap being a bit more rare for me. IdkGoodName (talk) 13:48, 3 September 2020 (UTC)
 * A variation like this is anything but unusual in languages that don't contrast trills with taps. Even Spanish doesn't fully distinguish them; in a preconsonantal position both can occur in free variation. Sol505000 (talk) 14:29, 3 September 2020 (UTC)
 * The quotes I mentioned here may apply in Lithuanian too. When a phonemic trill is realized with only one contact, the articulatory mechanism may still be like a trill rather than a tap (although such a sound may still be referred to as a tap or ⟨ɾ⟩ in literature as it's often difficult or irrelevant to differentiate the two). Nardog (talk) 16:03, 3 September 2020 (UTC)

Short A
How do we know that it's near-open? In other words, on what grounds are we using the symbol ⟨ɐ⟩ instead of ⟨a⟩? Which source uses it? If it's OR, I suggest that we use a plain ⟨a⟩ instead, both here and on Help:IPA/Lithuanian (and transcriptions linking to that, of course). I understand allophonic ranges of vowels, but we need a source. Sol505000 (talk) 08:07, 3 September 2020 (UTC)
 * The vowel section is indeed badly undersourced. This paper looks perfect for supporting citations, also for ⟨ɐ⟩. –Austronesier (talk) 14:38, 3 September 2020 (UTC)
 * That's what I meant. Thanks! Sol505000 (talk) 15:06, 3 September 2020 (UTC)

Wrong R
Copied from Help talk:IPA/Lithuanian as this is more likely to be noticed here. I am Lithuanian. We absolutely do not use the voiced alveolar trill. Most of us don't even know this sound exists. If you find a peasant Lithuanian using this sound, that's an extreme edge case. We use the voiced uvular trill. This is not "differences between different parts of Lithuania", etc. If you try to speak Lithuanian with an alveolar R, you'll sound ridiculous. Can we fix this? And the Help:IPA/Lithuanian page? Also, the Lithuanian row should be moved from the Voiced dental, alveolar and postalveolar trills page to the Voiced uvular trill page. 37.157.147.100 (talk) 10:38, 26 April 2021 (UTC)
 * That's a bold claim. Please provide a reliable source. Until then, we follow the existing sources (e.g. Augustaitis 1964), corrobrated by the judgement and expertise of native editors like . –Austronesier (talk) 11:03, 26 April 2021 (UTC)
 * Well this is something of a glitch in the matrix situation. Initially I didn't look for a source because I didn't learn this from one. For us this is common knowledge. I've looked around the Internet and almost every resource says alveolar. Everyone on TV uses uvular. Uvular is what we are taught in school. Every Lithuanian I've met uses uvular. The only thing that I found that you might consider reliable is this on page 192 (PDF page 210) where it sort of says that uvular is also allowed. Could then perhaps the voiced uvular trill be added as an alternative pronunciation? Do note that we have this minor "meme" in Lithuania where most people have always used pronunciation x for something, then our linguists say that's incorrect and that we should use pronunciation y, then most people laugh at the linguists because y sounds silly and continue to use x. This might be such a case. My opinion is that Wikipedia should represent how native speakers actually speak a language and only mention barely used rules. In practice the voiced alveolar trill is the alternative that almost no one knows about, but I don't expect to convince you just like this. Hopefully more Lithuanians will add to this thread in the future. Also,, what do you think of this discrepancy between the supposed rule and how we actually say R? 37.157.147.100 (talk) 13:23, 26 April 2021 (UTC)
 * That's a good one to start with. Page 44 (original pagination) would probably more apt, as it makes a statement about actual produced speech (some use [ʀ], most use [r]). I take this as a descriptive statement, with no prescriptive intentions. P. 192 just describes how a hypothetical sound production is phomenically perceived.
 * I have taken a closer look at Augustaitis (1964) in the meantime and found on p. 37 (my own bad translation): "An individual variant is the so-called uvular R (Zäpfchen-R), which is pronounced by some people in place of the usual apical dental R (Vorderzungen-R)". Augustaitis's description is more than 50 yrs old, Girdenis's book first appeared in 1995. Both say that uvular R is only used by some speakers. Things might have changed since then, but with these sources, we could at least add a word that uvular R exists. –Austronesier (talk) 15:32, 26 April 2021 (UTC)
 * Thank you for looking into this. Sorry I didn't check page 44 as it's a big book and not being a regular editor I have limited time for this. I should stop using the word "absolutely". It appears that Lithuanian certainly used (still uses in the countryside?) [r]; that's news to me but I won't deny it. Although it's surprising to me that [r] was dominant in 1995. We do see people from the countryside on TV and I've never heard them use [r] so that's why I thought it's not used. My experience being born in Kaunas and living here for 23 years is that everyone in the cities uses [ʀ]; sadly there doesn't seem to be a modern source confirming that. This sentence on page 44 caught my eye: "The situation is similar in French, only here the apical [r] is seen as a kind of phonetic provincialism. The norm of an elegant Parisian pronunciation gives preference to the uvular [ʀ]". After learning about [r], that is how I see Lithuanian as well.
 * Not being a regular editor I could easily swap out the IPA characters, but I'm not sure how exactly to add a mention of an additional [ʀ]. For now it can say that [ʀ] is used by some speakers because there's no source stating that [r] is archaic. 37.157.147.100 (talk) 16:50, 26 April 2021 (UTC)
 * Yesterday I talked to my family and they use [r] but in casual speech it usually sounds like [ʀ] to me. I must have forgotten what I learned in school, scratch that. No one ever told me that my [ʀ], which sounds like [ʀ], sounds like everyone's [r] which, to me, usually sounds like [ʀ]. I guess this misperception is what page 192 was about. Sorry for being annoying with my "alternative facts".
 * Still, according to both sources, and me being a live example, [ʀ] is a valid option. Learners of Lithuanian should probably know that they can use [ʀ] if they find it easier than [r]. Later today I'll try to edit [ʀ] in as an alternative used by some native speakers. 37.157.147.100 (talk) 05:07, 27 April 2021 (UTC)
 * Lithuanian is velarized (with a voiced frictionless coarticulation at the place of articulation of, so ) like its Russian equivalent, so no wonder it sounds like  to you. The recording on alveolar trill is not velarized because it shouldn't be, as it's meant to represent the default IPA value of the symbol ⟨r⟩. Sol505000 (talk) 06:03, 27 April 2021 (UTC)
 * I've added this to the bullet list below the consonant table (sorry, forgot the edit summary). I referenced Augustaitis but not Girdenis because I couldn't find the Lithuanian version and the English translation probably has different pagination. If this looks fine, I'll also add this to the table in Voiced uvular trill as individual phoneme pages seem to mention variations. 37.157.147.100 (talk) 13:04, 27 April 2021 (UTC)
 * @37.157.147.100: Yes, an entry in the bullet list is perfect, since the table itself lists phonemes (not sure why the [ŋ] is there; better should be a another bullet list entry as allophone of /n/). Don't worry about the misunderstanding, in the end we have built in a puzzle piece to improve the article. Thanks! –Austronesier (talk) 16:31, 27 April 2021 (UTC)
 * I'd love to help, but I can't. I haven't done any research into the phonetics of Lithuanian, apart from the sources listed on this page. I left the country many years ago and I only speak to my family in Lithuanian (via Skype), except for a very occasional trip back home when I always try to interact with strangers. My wife is French so (un)fortunately (depending on how you look at it) she taught our child that instead of Lithuanian, which we don't use. I use the alveolar trill, mostly with one contact (an alveolar tap), which perhaps shows my age. The standard laid out in Augustaitis (1964) is a good description of how I speak, AFAICS. I am aware of the uvular variants, but I'm not sure how widespread they are. Again, apologies for my ignorance. I mostly learned phonetics to improve my understanding of spoken English, because all the accents I've encountered were an absolute nightmare to deal with in the UK. TEFL (especially in public schools) is an effing joke when you're not taught how to understand the language. Sol505000 (talk) 18:59, 26 April 2021 (UTC)
 * Sorry, I was mostly wrong, see above. 37.157.147.100 (talk) 05:07, 27 April 2021 (UTC)

IPA transcription: recent recommendation regarding the Lithuanian language.
I invite you to read the following paper entitled "Recommendation on the adaptation of the International Phonetic Alphabet (IPA) to represent the sound units of the Common Lithuanian language" which was published in 2021. Here's the link: http://www.vlkk.lt/vlkk-nutarimai/protokoliniai-nutarimai/rekomendacija-del-tarptautines-fonetines-abeceles-pritaikymo-bendrines-lietuviu-kalbos-garsiniams-vienetams-zymeti — Preceding unsigned comment added by GuillaumeLDY (talk • contribs) 01:24, 30 January 2022 (UTC)

h
Sorry if this isn't the correct place to leave this kind of comment, but as a native speaker of Lithuanian I was quite surprised that the table of consonant phonemes includes the voiced velar fricative as the pronounciation of the letter "h", but not voiceless glottal fricative, which I and I would say most Lithuanians, use. Although I don't have a source I think this could be something that has changed over time, but either way "h" is most definitatly usually pronounced as a voiceless velar fricative. — Preceding unsigned comment added by 194.103.189.12 (talk) 15:26, 28 February 2022 (UTC)
 * Are you saying orthographic h is usually or ? You say glottal first but then velar. Given ch is [x] according to the article, I assume you meant to say glottal for both times. Is that correct? Nardog (talk) 06:03, 2 March 2022 (UTC)

ę
According to this article, ę is pronounced as /ɛ:/, but I'm Lithuanian, and never heard anyone use /ɛ:/ for ę, not even in formal speech. Muonium777 (talk) 10:14, 25 June 2024 (UTC)