Talk:Dutch phonology

Pro-Polder POV

The section on Polder Dutch seems to be worded in a somewhat unbalanced way, with a very, well, pro-Polder bias, suggesting that this pronunciation of the diphthongs is objectively easier, hence progressive and destined to become established ("standard") in all of the Netherlands. The prediction about the triumph of Polder diphthongs in future Dutch is, obviously, just an unsubstantiated guess, unless it is demonstrated that it is, indeed, spreading quickly all over the country in younger generations. As for the assessment of ease and naturalness, it is even more subjective and dubious, since the pronunciation of diphthongs with less differentiated components is favoured by the natural tendency towards assimilation, often ending in complete monophthongisation. This is seen in the development of the original Germanic diphthongs in Icelandic - and, as far as /ai/ is concerned, also in Yiddish - which has had the exact opposite direction compared to the German and English sound change that Stroop apparently views as a universal and timeless standard.--77.85.55.14 (talk) 01:39, 29 May 2020 (UTC)

I can see where you're coming from, though I don't think that's what was intended. I believe the term peripheralization might help here. As I recall, this is when things are made more distinct to increase articulatory distance among otherwise subtle acoustic contrasts. But I don't see the term elsewhere at Wikipedia (even at sound change), so my memory of the name of the process might be failing me. We can describe this process neutrally as a logical process while still being neutral as to whether it's better.

As for the claim that it is becoming more standard, I would suggest that we try to find a recent source that claims this. There are two sources cited at the end of the paragraph in question; even if they are found to back up the claim, it's been nearly a generation since they were published and it's a good idea to confirm claims like this about generational change. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 02:03, 29 May 2020 (UTC)

I agree that "the starting points of /ɛi, œy, ɔu/ are further lowered to [aɪ, aʏ, aʊ] to facilitate the pronunciation" is a very non-neutral wording, since the shift is not related to "natural" universals. Such a reasoning might apply e.g. to [ɪi̯] and [ʊu̯] as "embryonic" stages of diphthongization, which will tend to develop into diphthongs with greater articulatory distance, but not with "ripe" diphthongs like /ɛi̯, œy̯, ɔu̯/. It is better described as a push chain triggered by the diphthongized articulation of the mid long/tense vowels /eː, øː, oː/, as is also mentioned in the following sentences. I suggest to completely scrap the part "to facilitate the pronunciation". –Austronesier (talk) 18:08, 29 May 2020 (UTC)

Sounds convincing, no doubt. But can you present reliable sources that back the push-chain theory? — Anyway, I am in favour of scrapping the "facilitate the pronunciation" claim. Love —LiliCharlie (talk) 18:43, 29 May 2020 (UTC)

@LiliCharlie: This web page[1] has led me to this very interesting thesis[2] by Jacobi, cf. Figure 1.1 on page 2, where an earlier drag chain analysis (by Stroop) is cited. Chain, yes; push, no. :) –Austronesier (talk) 19:16, 29 May 2020 (UTC)

Collins and Mees: copyright status

This article repeatedly cites Collins and Mees's book. The reader is pointed to a PDF of this that was very conveniently available in late 2016 on the website of National Pedagogical Dragomanov University (Ukraine), deleted (or moved) since, but nevertheless even now very conveniently available thanks to Wayback. Offhand I know nothing about the coauthors, but the URL doesn't suggest that one of them was then teaching at the university. Shouldn't the link be removed as a inviting the reader to a likely copyright violation? -- Hoary (talk) 22:58, 28 June 2020 (UTC)

I agree, and turns out the PDF was linked from ~50 articles. I just removed the link in every one of them. I had done the same in one a couple years ago but didn't realize it was used in so many... Nardog (talk) 23:55, 28 June 2020 (UTC)

Excellent work, Nardog! (You've made me feel very lazy.) -- Hoary (talk) 00:34, 29 June 2020 (UTC)

open/close vowels

The Dutch short O is described as being open: ɔ This seems so wrong! The sound, in fact, is very closed, just listen to "jongens", for instance. The sound is not really very different from the open u sound in the German equivalent "Jungen". Perhaps, it depends on the individual speaker, but virtually every time I hear speakers who pronounce a very closed short "o" sound.

Or listen to "vol" (= "full"), for instance, at forvo.com: Examples "vol zijn van", "vol vuur", and even more noticeable: "vol in beeld" (really sounds like engl. "full"). — Preceding unsigned comment added by 84.58.206.90 (talk) 17:23, 6 October 2020 (UTC)

Thank you for bringing this up. There is considerable variation in height for /ɔ/ (cf. Booij (1995:7) who cites this source), which should be mentioned here. The entry for /ɔ/ needs to be reworked anyway, since it contains OR about "pharyngealization" which is only mentioned in passing by Collins & Mees (2003), and is falsely attributed to Verhoeven (2005) and Gussenhoven (1999) who don't mention it all (@pagewatchers: have I overlooked something?). –Austronesier (talk) 08:44, 7 October 2020 (UTC)

I am the one who had the IP 84.58.206.90 on October 6 2020 and started that topic... Many thanks also from my side! I thought for sure that someone would argue against it, but it seems that there really is something to what I am observing. - And yes, there are exactly those examples in your link that I had in mind (dom, tong, vol etc.) — Preceding unsigned comment added by 84.58.199.46 (talk) 00:17, 13 October 2020 (UTC)

@Austronesier: I understand we're only citing Collins & Mees (2003:132), but I don't quite comprehend what an "open-mid back rounded [ɔ] ... with pharyngealization" is supposed to be. Cardinal back vowels are defined as becoming fricative sounds if the tongue is retracted any further. So do Collins & Mees assert that the usual realisation of Standard Dutch /ɔ/ has audible friction and is a phonological vowel but not a vocoid? (I speak decent Dutch but I don't recall having heard a fricative /ɔ/ that is similar to [ʁ] or [ʕ] with lip rounding. That's only me, though.) Love —LiliCharlie (talk) 20:43, 6 January 2021 (UTC)

@LiliCharlie: What they actually write is: "There is retraction of the tongue-root towards the pharynx wall". I have rephrased this to "pharyngealization" (following the preceding version), but if you think that this is inapt, I'll be the last one to protest if the "pharyngealization"-thing is completely scrapped. But FWIW, pharyngealized back vowels are not unheard of in the lit, and are e.g. described for Upper Saxon[3], nɔˤ. –Austronesier (talk) 21:04, 6 January 2021 (UTC)

That sounds like "retracted tongue root" to me. A difficult and controversial notion. Love —LiliCharlie (talk) 21:21, 6 January 2021 (UTC)

I am the anon who added the section "short /ɔ/" above, but not the one that added this section. If you scroll up and read my section, then this user really says exactly the same thing that I said back then. So this confirms that the Dutch /ɔ/ tends to be much closer than the German /ɔ/. The question is only whether the Dutch is [ɔ] and the German is [ɒ], or whether the Dutch is [o] and the German is [ɔ]. 78.55.36.66 (talk) 21:13, 29 March 2021 (UTC)

Listen to wikt:stok and wikt:Stock for example. It's clearly a different vowel. And see that it has nothing to do with an underlying dialectal distinction of a-umlaut. It's simply a different realisation of the standard phoneme. -- Here also the unsymmetric nature of the Dutch vowel system must be borne in mind. In the unrounded row there is [i], [ɪ], [ɛ], while the rounded vowels lack on of these steps. Hence also theoretically it makes sense that the more open rounded vowels should move to a middle position, namely [o] and [ø]. That's exactly what I, and apparently other Germans, hear. 78.55.36.66 (talk) 21:21, 29 March 2021 (UTC)

explain North and South

The article mentions "north" and "south" many times, but it is not explained what this means. I don't know. Is Amsterdam already the northern Netherlands? Or is it only Groningen and Friesland that is north? --K. 20:02, 6 February 2023 (UTC)

That depends; in a discussion of Flemish versus Netherlands Dutch, the border forms the obvious dividing line. But within The Netherlands itself, the Rhine/Meuse delta is seen by many Dutch citizens as a major dividing line, both culturally and linguistically. Still, the central, standard dialect associated with the Randstad is not usually thought of as "typically Northern" (even if Amsterdam, Rotterdam, The Hague, and Utrecht are all situated north of the great rivers) --- one would be inclined to regard as such: Gronings, Friesian (a language not a dialect, albeit mostly for linguistic-historic reasons, given that the modern forms of Friesian and standard Dutch have converged in so many respects) and West-Friesian (as the Dutch call the speech of Northern North Holland, i.e. North of the Zaanstreek). — Preceding unsigned comment added by 2A01:CB0C:CD:D800:6C58:E74F:F368:C5D9 (talk) 13:07, 23 March 2023 (UTC)

Phonologically, the border more or less follows the great rivers, with a transition zone. Northern Dutch as used in this article is mostly used in the West of the Netherlands, not in the North. What variety has more prestige is a matter of opinion. PiusImpavidus (talk) 14:22, 20 July 2023 (UTC)

I agree with the above. I also think that this should be incorporated into the article. The current article contrasts "Belgium" with "north". Assuming that the sound files are from a speaker of Standard Belgian Dutch, I think "Belgium" doesn't need to be changed. But "north" should be specified further, and in some places should be replaced by (Standard Netherlandic Dutch). Barefoot Banana (talk) 12:05, 30 August 2023 (UTC)

stress

I feel that intercalated schwa's, so typical of the Dutch language, have something to do with this. 2A01:CB0C:CD:D800:6C58:E74F:F368:C5D9 (talk) 11:01, 23 March 2023 (UTC)

What is this about? Can you expand so that people can discuss? Barefoot Banana (talk) 12:01, 30 August 2023 (UTC)

Broad vs narrow phonetic transcriptions

@Nardog Thanks for the feedback on my last edit!

I'm by no means an expert in phonetics, so take this with a grain of salt, but: Even in a broad phonetic transcription, aren't you supposed to avoid allophones (since that's strictly a phonemic concern)? E.g., even though /r/ can be realized as anything from [r] to [ʀ], would you still be able to write [blɛːr] in a broad phonetic transcription if a narrower transcription would actually be [blɛːʀ]? I guess I'd imagined that, in phonetic transcriptions, the broad/narrow distinction would be about secondary features (like length or phonation) rather than the actual phones themselves (if that makes sense).

Either way, this gives me a good opportunity to actually sit down and read the Handbook of the IPA, so I thank you for that :-) BalinKingOfMoria (talk) 15:24, 17 August 2023 (UTC)

would you still be able to write [blɛːr] in a broad phonetic transcription if a narrower transcription would actually be [blɛːʀ]? Unequivocally yes. See Handbook, pp. 28–30. Nardog (talk) 23:15, 17 August 2023 (UTC)

Today I learned! (Along those lines, I've also undone an earlier edit of mine.) BalinKingOfMoria (talk) 23:32, 17 August 2023 (UTC)

@Nardog A quick question about p. 29 of the Handbook, which says:

These last two transcriptions look superficially very like a phonemic transcription, but they are in principle different because information has been included (albeit sparingly) which is not required for the unambiguous representation of the words.

Is it fair to assume that this is the case in (e.g.) [blɛːr]? (I don't know anything about Dutch phonology, and I don't want to jump to conclusions in case I've misunderstood this article.) BalinKingOfMoria (talk) 23:40, 17 August 2023 (UTC)

Sure, basically any phonemic transcription can be put in square brackets and it would be correct, because, as the Handbook puts it, a transcription can be devised which includes any number of additional symbols to indicate the phonetic realizations of the phonemes (which means that number can be zero; also a phonemic transcription usually includes at least some symbols that would match the realizations). This is one of the strengths of the IPA: if we want to focus on the realization of e.g. the vowel in blèr but not on the rhotic, we can be agnostic about the latter by representing it with the phonemic symbol, ⟨r⟩. Nardog (talk) 23:51, 17 August 2023 (UTC)

[ˈneidəɹˌlɑndəɹs]

There was one [ɹ] missing. Added it! Not sure whether a secondary stress should be indicated here though.

Wathiik (talk) 08:43, 4 October 2023 (UTC)

/ʃ ʒ/ or /ɕ ʑ/ ?

The use of /ʃ ʒ/ vs. /ɕ ʑ/ is inconsistent in the article. They refer to the same phoneme. Gussenhoven (1999) and Verhoeven (2005) use /ʃ ʒ/. Exarchus (talk) 11:38, 23 December 2023 (UTC)

Brabantian pronunciation of vowels

The article mentions:

"in Antwerp /ɪ/ is closer, more like [i]" and "In Antwerp, /ʏ/ may be as high as /y/ and the two vowels may differ in nothing but length."

According to my understanding, this is not specific to the city of Antwerp, also not to the province of Antwerp (which I assume Verhoeven (2005) means with 'Antwerp area'), but to the whole of the Brabantian region.

Roughly speaking, the Brabantian pronunciation (of 'polished' Dutch, that is) is like this: /ɪ/ = [i], /i/ = [iː], /ʏ/ = [y], /y/ = [yː], /ɛ/ = [ɛ], /eː/ = [eː], /øː/ = [øː], /ɑ/ = [a], /aː/ = [aː], /ɔ/ = [ɔ], /oː/ = [oː] and /u/ = either [u] or [u:] (two different phonemes, an understudied phenomenon).

All of this with the caveat that I'm writing this from a Belgian point of view and can't confirm (or deny) this for North Brabant. Exarchus (talk) 17:08, 27 December 2023 (UTC)

Related to this is the statement: "/i, y, u/ are frequently longer in Belgian Standard Dutch and most Belgian accents than in Northern Standard Dutch in which the length of these vowels is identical to that of lax vowels."

It is specifically in Brabantian that these are long vowels, see appendix 2 of the paper "Akoestische kenmerken van de Nederlandse klinkers in drie Vlaamse regio's" by Verhoeven & Van Baal (2002). From the three mentioned regions (Antwerp, Limburg, East Flanders), only Antwerp has a long /i/ and /u/ (but as mentioned, there is also a short [u], ignored in the paper, but the words used in the test (spoed, Loes) both have long [uː], so the results are valid on that point). All three regions have longer /y/ than /ʏ/, but I'd say in Limburg it's only half-long. Exarchus (talk) 20:58, 27 December 2023 (UTC)

By the way, I wonder if the long /y/ in Limburg and East Flanders is a consequence of the phoneme /y/ (specifically when written 〈uu〉) mostly occurring before /r/, where it is always long. (In the test, the nonsense words 'puut' and 'luus' were used.)

Also by the way, the paper talks about the Antwerpian /i/ becoming long to differentiate itself from /ɪ/. When I look at Middle Dutch phonology, it is rather the reverse: long /i:/ has become short (except in Brabant) because the duration contrast was no longer needed as there was a difference in quality. (edit: what specifically happened in Middle Dutch was apparently this: "monophthongisation of opening diphthongs: /iə/ > /iː/, /uə/ > /uː/. The result might have also been a short vowel (as in most Dutch dialects today), but they are known to have remained long at least before /r/.") Exarchus (talk) 21:56, 27 December 2023 (UTC)

To show how understudied (there is this paper from 2021) the phenomenon of the short vs. long /u/ in Brabantian is: the website of the university of Ghent tells us about Brabantian that 'ie' and 'oe' are always pronounced long, but then they manage to give an example where this is definitely not the case: boek [buk].

This phonemic inventory of the dialect of Asse also clearly distinguishes [u] and [uː]: http://www.woordenboekvanhetasses.be/uploads/docs/personalsite/Asseswoordenboek_Foneeminventaris.pdf Exarchus (talk) 23:17, 27 December 2023 (UTC)

I hadn't talked about the diphthongs yet, I'd say they are roughly like this: /ɛi/ = [ɛː], /œy/ = [œː], /ɔu/ = [ʌu] Exarchus (talk) 11:37, 28 December 2023 (UTC)

The current article text contains:

”* The native tense vowels /eː, øː, oː, aː/ are long [eː, øː, oː, aː] in stressed syllables and short [e, ø, o, a] elsewhere. The non-native oral vowels appear only in stressed syllables and thus are always long.”

With respect to the claim about the pronunciation of the occurrences in non-native words, I cannot hear any long versions in e.g.:

• /e/: energie;
• /ø/: reumatiek;
• /o/: onanie;
• /a/ (twice): analyse.

The current article text also contains:

”* The non-native /iː, yː, uː, ɛː, œː, ɔː/ occur only in stressed syllables. In unstressed syllables, they are replaced by the closest native vowel. For instance, verbs corresponding to the nouns analyse /aːnaːˈliːzə/^ⓘ ('analysis'), centrifuge /sɛntriˈfyːzjə/^ⓘ ('spinner'), and zone /ˈzɔːnə/^ⓘ ('zone') are analyseren /aːnaːliˈzeːrən/^ⓘ ('to analyze'), centrifugeren /sɛntrifyˈɣeːrən/^ⓘ ('to spin-dry'), and zoneren /zoːˈneːrən/ ('to divide into zones').”

I do not hear long versions of:

• /o/ in zoneren;
• /a/ (twice) in analyse or analyseren,

any more than I hear long versions of:

• /i/ in analyseren;
• /y/ in centrifugeren.

[A case could be made for long /aː/ in the second syllable in the non-standard verb analiseren /anaːliˈzeːrən/ ('to analise'), from the adjective anaal /aˈnaːl/ ('anal').]

Moreover, I cannot see any backing for the claim that /o/ in zoneren and either /a/ in analyseren are pronounced as long vowels.Redav (talk) 17:00, 12 March 2024 (UTC) Redav (talk) 17:00, 12 March 2024 (UTC)