Talk:Dutch phonology/Archive 1

Archive 1

Vowels

I've been trying to understand the Vowel section of this article but fail to follow the counting that is presented. It is stated that Dutch has 14 simple vowels, which suggests that the chart of monothongs contain 14 entries. I count no more than 11. What am I missing? Is there perhaps a difference between simple vowels and monophthongs? Similarly, it is stated that there are 4 diphthongs and that 3 more are included in the chart. That suggest that there should be 7 items on the chart of diphthongs. I count only 6. Again, what am I missing? Should the list of vowel example words not contain at least 18(=14+4) or possibly 21(=14+4+3) entries? I only count 16. Again, what am I missing? If I am misunderstanding something, which is likely, am I the only "idiot" out there?

It seems to me this article and International Phonetic Alphabet for Dutch are covering a lot of the same ground and would benefit from being merged. Since articles on phonologies of languages are generally called "XXX phonology" (see Wikipedia:WikiProject Phonetics#Language phonologies), I say this article is the right place to merge to. Thoughts? User:Angr 16:17, 26 August 2006 (UTC)

There's been some debate about this for other articles, because the IPA articles are often in the business of explaining how the language's orthography translates into IPA, which is really not the same as the language's phonology. So some languages have articles called "X Phonology" and "X Orthography", whereas others have one article called "X Phonology and Orthography".

Honestly, both of these articles are so incomplete and disorganized that it might be better for someone with knowledge of Dutch (...not me, unfortunately) to cannibalize them both to create a new article or articles, depending on what the consensus is on having two articles versus one. --Armchairlinguist 18:01, 19 December 2006 (UTC)

The Dutch IPA article was really just a brief orthography guide, and I don't see much harm in keeping such information here as long as the focus of the rest of the article remains on phonology/phonetics rather than spelling. I've merged the two for now.

Peter ^Isotalo 23:00, 24 December 2006 (UTC)

Why is there no colon after the [o] in the example transcriptions of <goal> and <oven>? This seems inconsistent with the rest of the article. Change?Homun 15:03, 1 January 2007 (UTC)

Homun, you are touching upon a more extensive problem. In Dutch people refer to the a in zak /ɑ/ as de korte a (the short a) and the one in zaak /a/ as 'de lange a'. Unfortunately the difference is really more a matter of laxness or openness and not of length. De length hardly matters. If fact the only words where length as such is phonemic is in pairs like dor - door. /dɔr/ - /dɔ:r/ (and yes this Dutch speaker actually uses /r/...rrroling!). I think it would be better to limit the use of the colon to such cases but I don't think that there is agreement upon that point

The vowel sounds before r are rather different and that is not discussed. —The preceding unsigned comment was added by 152.1.193.137 (talk) 21:49, 30 March 2007 (UTC).

Well, the vowel length is still phonemic in that it can act as the decisive factor where vowel quality does not. An example from myself, coming from Eindhoven (east Brabant), is how I would understand [mɑ:n] (long, but lax a) and [man] (short, but tense a). Neither are found in standard Dutch, but they're found in some dialects (notably of Antwerp). Now, I myself would analyse both [mɑ:n] and [ma:n] as 'long a', therefore <maan>, despite the vowel quality being that of the standard 'short a' in the first case. And the same happens to [man] and [mɑn], which I would understand as <man> with short a. So in this case at leat for me personally, the vowel length is more important in the disambiguation than the vowel quality is, and therefore it would be phonemic. CodeCat 19:43, 13 June 2007 (UTC)

Well, no. When they are six years old, the Dutch learn that their language has long and short vowels. Then as adults, at least some of them have to come to terms with the fact that empirical science says no. It is interesting that Antwerp switches the a-sounds around and attention has to be paid to length. I suspect what is actually happening is that you "hear" them long or short because the context is telling you which it should be. 2A01:CB0C:CD:D800:6C58:E74F:F368:C5D9 (talk) 13:36, 23 March 2023 (UTC)

Flemish/Southern Dutch

I think the article is still a little too much biased towards ‘standard’ Dutch. The word in itself is not even well-defined, because there is a Flemish standard as well. For example:

• In the south, there are (almost) no diphthongs: e: really is a long e, same with ø: and o:, ɛi is ɛ:, œy is œ: and ʌu is ʌ:. This should at least be mentioned below the chart, or even better: in a separate chart.
• The w is pronounced w in some areas.
• Mention that the ‘standard’ Dutch of the south is that as spoken on Belgian television. I do not think this is existent as a real spoken language anywhere. There is a ‘Standard Flemish’, however, but that would be another article, since its status is disputed.

Then I have some comments which are not specific to the south:

• I am not sure whether ʃ is not a native sound: it occurs, infrequently in words like sjaal, sjouwen, sjorren... The sound is not necessarily the same as in meisje, huisje.
• r is sometimes realized as ɹ, the ‘Gooise r’ phenomenon, mostly famous from the music show ‘Kinderen voor Kinderen’.

I made these changes in the Orthography chart before it was merged with this page, but I am not enough of a specialist to describe these phenomena rigourously. Hamaryns 10:15, 15 January 2007 (UTC)

Regarding KvK and the "Gooise r", I found this article from the 15 July 2000 NRC Handelsblad amusing. Unhappily for any effort to create a concise yet accurate description of Dutch phonology is the proposition expressed in the tagline "Het Nederlands heeft twaalf erren, zonder veel samenhang" — "Dutch has 12 R's [i.e. 12 different ways of pronouncing r], without much relation to each other." In the article a professor surmises that Kinderen voor Kinderen sing-along CDs are responsible for spreading Gooise r through emulation and repetition, particularly by pre-teen girls. -- IslandGyrl 14:10, 17 January 2007 (UTC)

Some rewriting needed

The article starts with "Dutch language devoices all consonants at the ends of words (e.g. a final /d/ becomes [t]), which presents a problem for Dutch speakers when learning English". This is true, but hardly a good initial sentence of an article on Dutch phonology. Then it continues with some ortographic facts.

Also, the next paragraph "Because of assimilation, often the initial consonant of the next word is usually also devoiced, e.g. het vee (the cattle) is /(h)ətfe/. This process of devoicing is taken to an extreme in some regions (Amsterdam, Friesland) with almost complete loss of /v/, /z/ and /ɣ/." - devoicing of /z/ and /v/ is indeed very common in certain regions, but has nothing to do with assimilation.

The whole introduction seems like a list of random facts, instead of an introduction to the phonology of Dutch. Could someone rewrite please?

Jalwikip 13:11, 12 April 2007 (UTC)

Katje

The cart of Dutch consonants fails to mention the voiceless palatal fricative [c], as in katje ['kɑcə]. 81.164.101.57

For one, it's a plosive, not a fricative. And the reason it's not included is that it's a phone, not a phoneme. If we were to include [c], we should also include, say [J] (the palatal nasal) in Jantje ([jAJc@]). As I wrote above, this article needs major clean-up, and a section on non-phonemic sounds and allophones should be included, but the phoneme chart is not the right place for that. Jalwikip 12:05, 18 June 2007 (UTC)

/9/

The vowel chart lists /9/ as phoneme (IPA œ), but no examples are given. Being a native Dutch speaker, I cannot think of any word having /9/ as a phoneme. Can anyone enlighten me or remove this as a phoneme (and update 14 to 13 in the 'simple vowel phonemes' above)? Jalwikip 12:07, 18 June 2007 (UTC)

Checked the Dutch Wikipedia, and it lists only 13 vowels. I think the original writer included œ as it was listed as allophone for ø. The latter I changed to ʏ, in accordance with both my own findings (I'm a native speaker) and the Dutch Wikipedia page. Also, I removed the part of the allophones also occuring before /l/ which isn't true (only before /r/, e.g. <deur> /dør/ [dʏ:r] but <beul> /bøl/ [bøl]. Perhaps there's a slight lowering, but definitly not to the values after /r/. I'll update the image in four days (just made a wikimedia account, they have a four day waiting policy). Jalwikip (talk) 15:17, 29 November 2007 (UTC)

I think the actual realisation of 'short u' depends on dialect. People from Noord Holland seem to pronounce dun with a sound close to [ʏ]; that is, like the German short ü in dünn (take a listen to this audio sample). However, further south, the pronunciation seems to be closer to [ɵ] or [œ], which is generally more centred and lower, like the German short ö. Compare Dutch kunnen with German können: as a native of Eindhoven, I pronounce the former very close to this pronunciation of können, only without the aspiration. Now, historically, the 'short u' in Dutch is equivalent not to 'long u' but to 'eu', and it is the result of an original short u [u] or short o being umlauted (making it cognate to German ü and ö). And since the diphthong /œy/ has an unmistakably lower starting point than 'short u', perhaps the sound is best denoted with /̞ø/ (lowered /ø/), or alternatively the diphthong could be denoted /ɶy/ since the start point is in fact rather close to /a/ I think. And as for the pronunciation before r: deur to me has the same sound as 'short u' except longer [dœ:ʀ] or perhaps even a diphthongised R a-la German [dœɐʀ]. --CodeCat (talk) 21:05, 27 May 2008 (UTC)

[β]

In the text below the IPA chart of Dutch consonants the author mentions that the /ʋ/ phoneme is sometimes realized as the voiced bilabial fricative [β]. I think the phone is closer to a bilabial approximant. Ivo von Rosenqvist

I'm not sure in which (supposedly Hollandic) dialects the "w" is pronounced bilabially. The labiodental approximant /ʋ/ seems pretty standard and universal to a Hollandic-speaking person like me. Someone using [β] or the bilabial approximant [β̞,] would sound Spanish to me. Afasmit (talk) 12:12, 26 November 2007 (UTC)

Isn't "w" a bilabial approximant in the Suriname dialect? −Woodstone (talk) 09:32, 27 November 2007 (UTC)

Yes, but I'm not sure if that could be called a 'dialect'. It is a 'variant' of Dutch, and for Surinames living in the Netherlands, it's an accent (sometimes on top of a regional accent). Jalwikip (talk) 15:18, 29 November 2007 (UTC)

In Randstad Dutch and Northern Dutch the labiodental approximant is pretty much standard. The bilabial pronounciation is the standard pronunciation in Belgian Dutch. It also occurs in the Southern parts of the Netherlands (Limburg and Brabant). But as I said, I think the phone is closer to a bilabial approximant than to a voiced bilabial fricative. In Suriname "w" is pronounced as a voiced labial-velar approximant (like in English). Ivo von Rosenqvist (talk) —Preceding comment was added at 16:38, 8 February 2008 (UTC)

Do you have any scientific sources for that? I am pretty sure I pronounce /ʋ/ as a labiodental approximant, and I think that's the normal pronunciation here (I am from Leiden, Zuid-Holland). Ucucha 18:55, 8 February 2008 (UTC)

I am sorry. I made a mistake in my previous post. Instead of voiced bilabial fricative I meant to say labiodental approximant in my first sentence. I corrected it now. So, when living in Leiden, it is normal to pronounce /ʋ/ labiodentally. It is the standard pronunciation in most of the Netherlands. Only Belgium and the South have a bilabial pronunciation.

Listen for instance to the news on Dutch and Belgian television. The Dutch newsreaders will pronounce /ʋ/ more or less like < w > in German Wasser, whereas on Belgian television they will pronounce it more or less like < b > in Spanish trabajar. Whereas in Suriname /ʋ/ will more sound like < w > in English water.

I have studied Dutch at the University of Ghent. I have got this information from an unpublished workbook by the Flemish professor Johan Taeldeman. And I am a Flemish native speaker of the Dutch language myself.Ivo von Rosenqvist (talk) 19:38, 8 February 2008 (UTC)

According to Verhoven (2005), Standard Dutch has the labial-velar [w], not a bilabial approximant. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 20:06, 8 February 2008 (UTC)

True. But a bilabial pronunciation does exist as well. The Flemish and Surinamese /ʋ/ are clearly bilabial. It is just not the Standard pronunciation.

To put it simple. Standard Dutch uses [ʋ]. Belgian Dutch uses [β]. And Surinamese Dutch uses [w]. Ivo von Rosenqvist (talk) 20:24, 8 February 2008 (UTC)

Well then why does this source say that standard Belgian Dutch uses [w]? — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 21:16, 8 February 2008 (UTC)

Verhoeven (1995) probably uses a simplified version of IPA. This is not uncommon. In German phonetics [ʋ] as in Wasser is often simply transcribed as [v] (even though its pronunciation is closer to Standard Dutch water than to Standard Dutch vis).

Anyway, you can ask any Dutch speaker. There is a distinct difference between the Standard Dutch, the Belgian Dutch and the Surinamese Dutch pronunciation of the phoneme /ʋ/. Ivo von Rosenqvist (talk) 22:12, 8 February 2008 (UTC)

Verhoven (2005) explicitely says it's labial-velar: "[w] has a labial-velar articulation rather than labiodental in many accents of Netherlandic Dutch." (p245). If you'd like to look it up, it's in Journal of the International Phonetic Association volume 35. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 23:00, 8 February 2008 (UTC)

Is he perhaps referring to /ʋ/ in final position? The final consonant in words like lauw "between warm and cold" /lɔuʋ/ might indeed be close to /w/ (though I'd say it's rather a bilabial approximant in my pronunciation). In the word klauwier "a bird" it might be even closer to /w/ ([klɔuwir]). Ucucha 10:37, 9 February 2008 (UTC)

Good point. I didn't even think of that.

In words like klauwier and lauwe the standard pronunciation is bilabial. In this case there is no difference between Standard Dutch and Belgian Dutch. But I think that /ʋ/ is only pronounced as labial-velar in Surinamese Dutch. I think any native speaker will agree that there is a difference between the Surinamese pronunciation of /ʋ/ and the Standard Dutch pronunciation of /ʋ/ in leeuwen.

In lauw on the other hand, there is no consonant. It is just pronounced as [lɔu].

So, to put it simple:

[ʋ] -- Standard Dutch water (+ German Wasser)

[β] -- Belgian Dutch water -- Standard Dutch and Belgian Dutch lauwe (+ Spanish trabajar)

[w] -- Surinamese Dutch water and lauwe (+ English water)

Ivo von Rosenqvist (talk) 13:01, 9 February 2008 (UTC)

Are you sure that lauw has no final consonant? I think I can actually hear a difference in pairs like u-uw, jou-jouw and nou-nauw. Ucucha 14:32, 9 February 2008 (UTC)

I think that jou and jouw are supposed to be pronounced the same. And I think in un-monotorized speech jou and jouw really do sound the same. It is only when one really wants to emphasize the difference between the two homophones that one pronounces the /ʋ/ in jouw. It's like verrassing and verassing. Normally the two words are pronounced exactly the same. But one can add a glottal stop to verassing to distinguish it from its homophone. These are artificial differences. Similarly, one can also pronounce /ʋ/ in lauw. But it is not a normal way of speaking. In lauwe and klauwier on the other hand it is normal pronounce /ʋ/. It would sound odd not to pronounce /ʋ/ in those words.

But perhaps it is easier to hear with leeuw [leu] versus leeuwen ['le:βə]. The word leeuwen does not even contain a diphthongue [eu] anymore. No one will ever say *['leuʔə]. Or *['le:ʋə] for that matter. And only someone with a Surinamese accent will say ['le:wə]. But on the other hand it is possible to pronounce leeuw as [le:β]. For instance: Je moet beter luisteren. Ik heb leeuw gezegd. Niet leeg! Ivo von Rosenqvist (talk) 15:31, 9 February 2008 (UTC)

Verhoeven wasn't talking about in final position. He transcribes waren as [waːʀə], wie as [wi], kwam as kwɑm] etc. amd makes no reference to a bilabial approximant. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 20:03, 9 February 2008 (UTC)

Oh hey, I forgot I had some other sources on Dutch. Let's see, according to Peters (2006), Gussenhoven & Aarts (1999), and Verhoeven (2007), Hasselt, Maastricht and Hamont (respectively, all in Limburg) it's a bilabial approximant. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 06:15, 11 February 2008 (UTC)

Perhaps it is better to base this article on more than one source. I think there are still a lot of innaccuracies in the article. Ivo von Rosenqvist (talk) 21:02, 11 February 2008 (UTC)

To start this discussion back up again... My pronunciation of <w> (Eindhoven, i.e. southeastern) is quite clearly different from anything here. It is not [ʋ] because it is bilabial, it is not [β] because it does not fricate, but it is also not [w] as it is clearly different from the Suriname pronunciation (which is clearly rounded and resembles the vowel [u]). Reading Voiced labio-velar approximant and Roundedness, I realised that my pronunciation of <w> is actually compressed, and not protruded like the usual [w]. It seems to stand between the approximants [w] and [ɥ] and somewhat resembles the central vowels [ʉ] or [ɵ]. I am not entirely sure how it is pronounced in standard Belgian, but most likely it is indeed more like a compressed [w] rather than a real fricative ([β]). --CodeCat (talk) 00:23, 1 February 2010 (UTC)

You haven't eliminated a bilabial approximant as a possibility, which could resemble a compressed [w]. Do we have a sound file from e.g. Spanish that CodeCat could listen to for comparison? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 05:07, 1 February 2010 (UTC)

As an example of the difference between accents: the sentence
Wat hebben ze gedaan met mij mooie Nederlandse taal!
pronounced by an Amsterdammer visiting Brussels, sounds to my Belgian ears like
[vɔ,t hɛ'bə·zə, χə,da:'a·n mɛ·t mə,n mo:'jə, ne:'də·ɹlɑ,nsə, ta'a:,l]
where I use (non-standardly) apostrophe, middle-dot and comma to represent three levels of the (to me extreme) musical pitch variation of the immediately preceding vowel, instead of
[ɥɑt hɛbəzə ʝəda:n mɛt mən mo:jə ne:dərlɑntsə ta:l]
(with [ɥ] for "non-syllabic [y]", trilled (apical or laminar) alveolar r, and ʝ or maybe ɣ as "the voiced counterpart of the German ich-laut" where the "Hollandic g" sounds to me like a German ach-laut) as it would be said with much less musical variation in a Belgian (or maybe Southern Dutch) accent. I'm not writing the stresses because they are the same. Tonymec (talk) 23:13, 19 August 2012 (UTC)

/ɣ, x, χ/

First, I don't think I know anyone that uses a /ɣ/ in Dutch. I know only /x/. Comments?

Second, the article says that "The phoneme /ɡ/ became a voiced velar fricative /ɣ/ [...]" which seems to contradict what is being said in High_German_consonant_shift: "The West Germanic voiced velar fricative /ɣ/ shifted to /g/ in Old High German in all positions. [...] Dutch has retained the original Germanic /ɣ/".

129.27.202.101 15:01, 26 September 2007 (UTC)

the latter is true, removing the wrong quote. Jalwikip 13:55, 13 November 2007 (UTC)

The difference between /x/ and /ɣ/ for <g> is probably dialectal.Cameron Nedland (talk) 00:04, 19 December 2007 (UTC)

Yes, <g> is /x/ for all Northern, Western and Eastern varieties of Dutch, and /ɣ/ for Southern varieties. All varieties have a voiceless fricative for <ch>, although /ç/ can be heard in the Southern varieties instead of /x/. Jalwikip (talk) 15:13, 7 January 2008 (UTC)

According to the Wikipedia article Dutch <ch> is pronounced as the voiceless velar fricative [x]. I think however that this pronunciation only occurs in Belgium and the Southern Netherlands (Brabant and Limburg). In Standard Dutch the pronunciation is closer to the voiceless uvular fricative [χ]. Ivo von Rosenqvist (talk) 21:02, 11 February 2008 (UTC)

I've cleaned up the article a bit. I'm kind of confused by /ɣ/. Verhoeven (2005) says it's not present in Netherlandic Dutch but I don't have the first page of Gussenhoven (1992/1999). — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 22:59, 11 February 2008 (UTC)

I am not entirely sure myself. The Belgian Dutch situation is easy. The phoneme /ɣ/ is pronounced as the voiced velar fricative [ɣ]. The phoneme /x/ is pronounced as the voiceless velar fricative [x], unless the phoneme is preceded by a front vowel and followed by the palatal approximant [j], then the phoneme gets palatalized. The word wieg [βix] becomes wiegje ['βiçjə] in Belgian Dutch. I am less sure about Standard Dutch. The phoneme /x/ is pronounced as the voiceless uvular fricative [χ]. Word initial the phoneme /ɣ/ is nowadays also pronounced as [χ]. Only old newsreaders and actors still make a distinction between gloor and chloor. In the middle of a word (like in wagen) it is still voiced. But I am not entirely sure what sound it is. I think the sound that comes closest is the raised voiced uvular fricative [ʁ̝]. But I don't know any scientific source, that backs up my theory. Ivo von Rosenqvist (talk) 13:08, 12 February 2008 (UTC)

I would say it's safe to state that <ch> and <g> (as fricative) are always [χ] in western Dutch. −Woodstone (talk) 15:14, 12 February 2008 (UTC)

Well, there must be some voiced variety. How about the <g> in legde? Is it also voiceless? And is there no difference at all between kachel en waggel? Ivo von Rosenqvist (talk) 18:39, 12 February 2008 (UTC)

Sure, by assimilation it goes back to voiced in <legde>. I hear and say no difference between the ch/gg in <kachel> and <waggel> in standard Dutch. In the southern parts they are different. −Woodstone (talk) 19:59, 12 February 2008 (UTC)

Now, let's say that <g> and <ch> have completely merged in the language of younger speakers of Standard Dutch. Nonetheless, I still think that among older generations there are some speakers of Standard Dutch who make a dictinction between <ch> and <g>. When I listen to newsreaders from the fifties and the sixties I clearly hear a difference between the two sounds. If you don't know what I'm talking about, just listen to some videoclips on this website. Perhaps this distinction is artificial, élitist, and only made by language purists, but it does exist. And it is not the same sound as the [ɣ], which is used in Belgium. This apparently obsolete pronunciation of <g> has been the standard pronunciation for a long time. So, I think we should at least mention it. Ivo von Rosenqvist (talk) 20:32, 12 February 2008 (UTC)

For what it's worth: the IPA handbook chapter on Dutch has only [χ] using <gat> as example. It has no mention of [ɣ] or [x]. Of course sometimes <ch> can be [ʃ] and <g> can be [g]. −Woodstone (talk) 21:19, 12 February 2008 (UTC)

Well, for modern Standard Dutch that one sound [χ] is sufficient. But in older Standard Dutch and Belgian Dutch, there are apparently more sounds. And even modern Standard Dutch speakers treat /ɣ/ and /x/ as different phonemes. Else they would say hij legte (by analogy with hij lachte).

You know what. I'll ask one of my professors what he thinks about it. I'll keep you posted. Ivo von Rosenqvist (talk) 21:28, 12 February 2008 (UTC)

The spelling convention is fossilised and harkens back to a phonemic distinction, as you say. But the sounds have merged in modern Dutch north of the border (or at any rate the great rivers). Dutch spelling also maintains the ei/ij distinction (maintained as an ei/ie distinction in some dialects) and the au/ou distinction. 2A01:CB0C:CD:D800:6C58:E74F:F368:C5D9 (talk) 13:29, 23 March 2023 (UTC)

I think that the current article is fine with regards to /ɣ/, correctly stating that it has been merged with /x~χ/ in the north (although that geographic specification is perhaps a bit more vague than desired).

If no one has any objections, I believe this topic should be archived. — Preceding unsigned comment added by Barefoot Banana (talk • contribs) 14:53, 28 August 2023 (UTC)

We normally archive discussions starting from the ones that haven't received replies for the longest. So you just prevented (or stalled) that. Nardog (talk) 15:28, 28 August 2023 (UTC)

The comment before mine was from 2023, and that was a reply to a comment from 2008. Why hasn't it been archived after 15 years of silence? Barefoot Banana (talk) 19:23, 28 August 2023 (UTC)

The special 's' sound in Dutch

I think that specifying /s/ for the Dutch 's' is completely counter-productive as it does not reflect the real sound in any way. Have you guys ever heard a native Dutchman/-woman pronounce the /s/ as in mens? It almost sounds like [ʃ] and also a little bit like the [ɕ] in Polish. But it definitely does NOT 100% sound like /s/ in English 'sea'!! -andy 84.149.64.73 01:29, 15 November 2007 (UTC)

Perhaps in some dialects (Amsterdam's comes to mind) the s could be [sʲ] or [ʃ], perhaps by yiddish influences. However, in standard Dutch s is simply [s]. Americans often pronounce Dutch names like Smit, Snel, or Mens with a [ʃ] because they are thinking of the German equivalents Schmid, schnell and Mensch. Afasmit (talk) 12:32, 26 November 2007 (UTC)

Agreed, as a native speaker I hear people say /s/ quite regularly, and it is definitly a [s]. Jalwikip (talk) 15:19, 29 November 2007 (UTC)

Oh my bad, completely forgot about this topic! Thanks for the answers. Well, seems you guys are right, it's mainly the people in Noord-Holland province who pronounce it that way. What is supposed to mean as well that bos (woods) and bosch (bush, etc.) sound EXACTLY the same when these people pronounce it. I really ought to be more specific! I have been hearing this distinctive "s" sound with Dutch presenter Rudi Carrell all those many years! And lo' and behold, he's from Alkmaar. That should make a perfect explanation. :) -andy 217.50.43.11 (talk) 23:08, 23 December 2010 (UTC)

'Bos' and 'Bosch' are not contrasting words. 'Bosch is the old spelling (before 1948, I believe), retained in family and place names. 'Bos' is the current spelling. In either case it means forest/woords (or also 'bunch'). Eti erik (talk) 13:59, 17 December 2017 (UTC)

I believe it's a type of Voiceless alveolar retracted sibilant and I have reason to believe it is general in Dutch (and also other languages, such as Icelandic, Faroese and Finnish apart from those languages where it is already described to be present in that article), perhaps apart from those few (southern? Limburgish?) dialects that have developped a separate phoneme /ʃ/. Most people fail to pick up the subtle difference, however, apparently even many phoneticians.

I have discussed this issue with Peter Schrijver, as regards Dutch and far beyond (he believes in a rule or strong tendency for sibilants to be apical – or perhaps better: retracted – when there is only a single phonetic one in a language, disregarding voice here, and I might add especially when it contrasts with an interdental place of articulation), and I have attended (much later) a talk by Aurelius Vijūnas, who presented evidence for its occurrence in various Indo-European languages and argued that it was the realisation of Proto-Indo-European */s/ (which would fit Schrijver's rule, as there is – as far as I can see – unanimous agreement that PIE is to be reconstructed with a single sibilant phoneme). --Florian Blaschke (talk) 08:33, 16 October 2012 (UTC)

Couldn't agree more, Florian. It's just that IPA sucks hard when it comes to sibilants, even the dental (or denti-alveolar) ones, which are neither like English /s/ and /θ/, are described as "alveolar". It's bullshit, cause Russian or Polish does not articulate it the same way standard German or British English does. But anyway, regarding what the IP said more than two years ago: "What is supposed to mean as well that bos (woods) and bosch (bush, etc.) sound EXACTLY the same when these people pronounce it." <bos> and <bosch> are always homophonous in standard Dutch, the <-sch> ending is NEVER [ʃ], but [s]. It's not German. --Ahls23 (talk) 12:35, 14 March 2013 (UTC)

Aurelijus Vijūnas's paper is available online by now, by the way, on Academia.edu. --Florian Blaschke (talk) 03:47, 18 June 2015 (UTC)

Netherlands Dutch /s, z/ are simply pronounced as alveolar (and in the TV Dutch accent /z/, can even become an alveolar semi-constrictive, especially in order to avoid the merger with /s/); there is no such thing as /ʃ/ or palatalization for it, it's just an impression caused by the different articulation  ; in Flemish Dutch however they are dental (or in a middle position between dental and alveolar in the Flemish Tussentaal) --Maury

Agreed. This /ʃ/ impression is reinforced among anglophones by being a staple of the "comedy dutch accent" i.e. the accent anglophone actors put on to convey Dutch (the usual result is not really a Dutch accent, but rather something like South-African English spoken by a garden gnome). 2A01:CB0C:CD:D800:99A1:850A:6F05:EB97 (talk) 14:39, 20 September 2020 (UTC)

Vowels and diphthongs

I am not familiar with this particular work (I could look it up in my library), but I see that on this Wikipage page < u > in Dutch hut has been transcribed as [ʏ]. This is also a simplification. The sound is much closer to [ø]. The Wikipedia page also fails to mention that [o:] as in Dutch boom is only pronounced as [o:] in Belgian Dutch. In Standard Dutch it is pronounced as the diphthongue [ou].

This is how Taeldeman classifies the Standard Dutch vowels and diphthongues:

Dutch vowels and diphthongues

Symbol	Example
IPA	IPA	Dutch
e	bet	bit
i 1	bit	biet
ø	høt	hut
y 1	fyt	fuut
ɛ	bɛt	bed
eː 2	beːɾ	beer
ə	də	de
øː 2	døːɾ	deur
ɑ	bɑt	bad
aː	zaːt	zaad
ɔ	bɔt	bot
oː 2	boːɾ	boor
u 1	ɦut	hoed
ɛi	ʋɛin	wijn
œy	œy	ui
ɔu	zɔut	zout

Notes:

1) pronounced longer before the phoneme /ɾ/.

2) diphthonguized when not before the phoneme /ɾ/.

Belgian diphthongs

You mention in a footnote that Belgians pronounce bout as [bɔut] rather than [bʌut], which is correct as far as I know. Most Belgians I know also mangle (uh, I mean, pronounce differently in a value-neutral descriptive way) other dipthongs. In particular, they pronounce bijt as [bɛːt] rather than [bɛit] and buit as [bœːt] rather than [bœyt]. (I'm not a linguist, so I may be doing the IPA wrong here, but there definitely is some difference in the way that Belgian Dutch speakers pronounce the diphthongs spelled as "ei" and "ui".) These two shifts occur also in the dialect of The Hague, leading to funny shibboleths like "broodje ei met ui".

I modified your IPA so you know how to do the formatting in the future. I have the vowel quadrangles of several Dutch dialects, all from the journal of the IPA, the most notable being Standard Belgian Dutch (which, I think has a place here). For the vowels spelled ij and ui respectively, this source shows almost no difference between the two styles but remember that this is Standard Belgian, which could be quite different from a more common Belgian variety. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 06:44, 19 February 2008 (UTC)

I wouldn't mention any differences between Belgian Dutch and Standard Dutch concidering the diphthongues. Those differences are not considered Standard Belgian Dutch. In correct Belgian Dutch (as is spoken on Flemish television) the diphtongues are exactly the same as in the Netherlands (apart from [e:], [o:] and [ø:] of course). Ivo von Rosenqvist ( talk) —Preceding comment was added at 23:17, 19 February 2008 (UTC)

You're also excluding the subtle rounding difference in ou, right? — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 00:05, 20 February 2008 (UTC)

To be fair, I am not entirely sure about this either. Nowadays, many young people in the Dutch Randstad pronounce their diphthongues slightly different. The diphthongues get lowered. These people pronounce the diphthongue ij for instance as [ai]. This phenomenon is called Poldernederlands, but so far it is not recognized as Standard (Dutch) Dutch. I think the suble rounding difference in ou is also a Poldernederlands phenomenon (and therefore not recognized as Standard Dutch). But I am not entirely sure. Where did you get this information? Ivo von Rosenqvist (talk) 11:05, 20 February 2008 (UTC)

Verhoeven (2005) (cited in the article). Remember, though, that the [ɔ] part of [ɔu] is lower than the /ɔ/ monophthong. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 19:35, 20 February 2008 (UTC)

Okay. Then I guess it will be all-right. Ivo von Rosenqvist (talk)

Southern dialects

As for the /ɣ/ sound: the difference between it and /x/ is made in the south of The Netherlands and in Belgium, and it corresponds more or less to the orthographic distinction g vs. ch. This is a feature that occurs even in parts of the country where the distinction between standard language and dialect is weak, and speakers of more distinctive dialects like Limburgish usually retain it when speaking the standard language. For what it's worth, the parts of The Netherlands where the /ɣ/ form is used coincide roughly with the majority-Catholic parts of the Dutch Republic that were governed as territories ("generaliteitslanden") by the States-General and that resemble Belgium (the old Spanish Netherlands) in culture and religion.

As for the shift of long /i:/ to /ɛi/: this creates many spelling headaches because the shifted words are still spelled with ij while the old words with the same sound are spelled with ei. In some dialects, the distinction is retained, but these are all dialects that are perceived as very distinct from the standard language (Frisian, Lower Saxon, Lower Frankish, etc.), and speakers will use the standard pronunciation when they intend to speak the standard language. One of the more remarkable of the dialects that retain the difference is that of the island Ameland, because unlike the other dialects that retain the difference, it is very close to the standard language.

Historical sound changes

The sentence: Dutch (with the exception of the Limburg dialects) did not participate in the second Germanic consonant shift is a little bit wrong. Limburgish only participated in a few of those soundshifts, compare:

• German machen /-x-/ Dutch maken, English make, Limburgish make
• German Pfanne /p͡f-/, Dutch pan, English pan, Limburgish pan (does mean something different, a good translation should be kille)
• German zwei /t͡s-/, Dutch twee, English two, Limburgish twee, twie, twieë

Limburgish only has a few examples of the shift: ich (nl: ik, de: ich, en: I) for example. Only Kerkraads has had more shifts. --OosWesThoesBes (talk) 17:30, 3 April 2008 (UTC)

Since German itself has different borders for different types of shifts (see High German consonant shift, the southern Limburg dialects are affected differently than the northern ones. There are also many German dialects that have "ik" for example. Jalwikip (talk) 15:15, 8 May 2008 (UTC)

Still, mache and pfan and tswai are only found in Kirchroajs, which is a Ripuarian language and not Limburgish. --OosWesThoesBes (talk) 13:27, 10 May 2008 (UTC)

Does the Kerkrade dialect really have an affricated initial /p/?! I thought it applied only to the former /t/. Mr KEBAB (talk) 03:08, 25 March 2017 (UTC)

Retroflex approximant

I didn't see anything about the use of the retroflex approximant in many varieties of northern Dutch, particularly at the end of syllables. Anyone knows anything about that? Aviad2001 (talk) 07:22, 30 April 2008 (UTC)

That's what Dutch people call the 'Leidse "r"' (Leiden-style r), it is associated with 'posh' people or with people having an American accent. This kind of 'r' is only heard after a vowel like 'markt'->/mɑɻkt/; 'kakker'->/kɑkɘɻ/.SuperMidget (talk) 23:02, 24 May 2008 (UTC)

It'd be good to add a few words about that in the article. Is it really a Leiden thing? I lived in Flanders and speak Dutch (though not quite fluently), and whenever I go to Holland it seems to be all over the place. Aviad2001 (talk) 20:36, 25 May 2008 (UTC)

It is widespread in the so-called Randstad region. Usually the Dutch call it the "Gooise r". A "Leidse r" is slightly different and it is, in the Leiden dialect, used before vowels as well.

I was told the Gooische R is something which is spreading throughout the Netherlands..I know I don't use it..but the lady had graduated on 'the' Dutch r. I put 'the' in quotation marks because she concluded there were as many as 27 different R's in the Netherlands and perhaps more. :) 81.68.255.36 (talk) 12:34, 21 February 2010 (UTC)

I am surprised to read that this sound occurs at the end of a syllable. That's not how it works: The 'Gooise R' that is gaining terrain is heard anywhere except before a vowel. So both at the end of a word and before a consonant. It is certainly heard in words such as 'barst', 'werk' etc. where it not at the end of a syllable. Eti erik (talk) 14:02, 17 December 2017 (UTC)

@Eti erik: In these words it's in the syllable coda, so in that sense it is at the end of a syllable. It's just not the very end of it. Mr KEBAB (talk) 14:05, 17 December 2017 (UTC)

@Mr KEBAB:That is not how I interpreted 'syllable end', but when you read that as 'in the coda' it is pretty much true, yes. I think the real mechanism is that the sound occurs anywhere except before a vowel, though. At least I think in the combination 'er is' it is not heard normally because r is followed by a vowel (in the next word). I would have to pay attention when listening to people with a Gooise R to be sure about this, though.... Eti erik (talk) 14:50, 17 December 2017 (UTC)

@Eti erik: I guess er is is rarely said with retroflex r. But that may apply only if you say it erris, as one word. Which is probably the most usual pronunciation (but that may depend on the speaker). If you put a glottal stop before is, then the sound following /r/ is no longer a vowel but a consonant. I don't know. Mr KEBAB (talk) 15:00, 17 December 2017 (UTC)

Haarlem

The spoken language of Haarlem is not Standard Dutch. Altough a wellknowm myth says so, it is untrue! —Preceding unsigned comment added by 87.210.117.240 (talk) 23:24, 18 October 2008 (UTC)

I have always heard this myth as "the least discernible accent is that found among speakers from Haarlem" that is, whereas speakers from any of the four big cities are readily recognised as such by almost all native speakers, the Haarlemmers are the hardest to place geographically. In other words, it is said to be the most non-descript, or perhaps neutral, accent. 2A01:CB0C:CD:D800:99A1:850A:6F05:EB97 (talk) 14:44, 20 September 2020 (UTC)

IJ, OU, and General Organization of "Standard" vs Variants

Something I am certain of is that the pronunciation of "ij/ei" is varies distinctively by class and region in the Dutch-speaking area. There's no mention of this.

And I thought that it was not pronounced [ɛi] in "good" Dutch.

But this raises the question of what "good" Dutch is -- and where if at all it is uniformly defined. (Government language academy? Leading dictionaries which give IPA?)

Likewise, my understanding was that a posh Holland accent has "ou" as [ɔu] not [ʌu]. So (as in England) can one speak more poshly than "standard"?

In general the article would benefit from a more systematic presentation of pronunciation groups: "Standard Holland Dutch" presented (with that name) first as the model, and then each regional (and ideally class) variant discussed, in its own section, in contrast to this. There are several very vague statements like "In some dialects..." -- without specifying which.

In Limburgish Dutch ij and ei are pronounced as [æɪ̯], ou and au are pronounced as [ɑu̯] and ui as [œɪ̯]. --OosWesThoesBes (talk) 15:32, 5 December 2008 (UTC)

ch

I changed

• The sound spelled <ch> is a uvular fricative in Standard Dutch[2] and velar in Belgian dialects.[3]

to

• The sound spelled <ch> is a voiceless velar fricative in Northern Dutch[2] and a voiceless palatal fricative in Southern Dutch.[3]

Because Standard Dutch pronunciation varies considerably between Belgium and the Netherlands. The uvular fricative is not part of the standard language in Belgium, the velar is also part of the standard language in the Netherlands for speakers coming from the southern provinces, so it's not exclusive to "Belgian dialects". The source is obviously biased to Randstad Dutch. I cannot check it because it's not a url link. So how will I ever be entitled to fix this then? This way anyone could invent a name and pretend it to be an authority.--Hooiwind (talk) 08:56, 8 December 2008 (UTC)

The source may be biased (I wouldn't know, I'm not familiar enough with Dutch), but you ought to find a reliable source yourself. The way you've edited it makes it seem as though your claims are sourced when they are not. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 09:00, 8 December 2008 (UTC)

Your vigil is appreciated, but the only thing I have really changed is the fact that Standard Dutch is not a synonym for Northern Dutch. (would be just like saying that Standard English equals British English and then say that "some American dialects use"...) But I will try to find a different source, which everybody can access.--Hooiwind (talk) 09:25, 8 December 2008 (UTC)

But that's not the only thing you've changed. If standard Dutch and northern Dutch are different, then don't assume that a claim about one actually refers to the other. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 17:52, 8 December 2008 (UTC)

I already acknowledged my mistake. Finding a correct, accessible source is harder than I thought though.--Hooiwind (talk) 19:52, 8 December 2008 (UTC)

ʝ?

I very much doubt the sound /ʝ/ appears as a phoneme, at least not as the sound sample sounds. Also, the remark: “4 The sound spelled <ch> is a uvular fricative in northern Standard Dutch[2] and velar in Belgian Standard Dutch.[3]” doesn’t really make sense: it is supposed to apply to /ɦ/, /ç/, /x/, /χ/, /ʝ/ and /ɣ/, but these sounds are so wildly different, I just can’t imagine. As a native speaker, I’d say the possible phonemes for <ch> are /ç/, /x/ and /χ/ (and certainly no voiced sounds, like /ɣ/!). I think they wanted the remark 5 to apply to some of these? Seems so. The relevant diffs are [1], [2] and those before that.

And this edit introduced the sound. I’ll make a note on User:Hooiwind’s talk. H. (talk) 13:30, 23 February 2009 (UTC)

I tried to fix the notes, looks good now.

Also note that there are no example words in the second table with ʝ in them. H. (talk) 13:47, 23 February 2009 (UTC)

I think [ʝ] is more to bee seen as the pronunciation of <g> in Belgian Dutch only, where the sound is most certainly pronounced in a voiced way. However, - as a native Dutch speaker and Belgian - I really doubt that the place of articulation is palatal. I'd rather consider it a [ɣʲ], so palatalised. Anyone? — Preceding unsigned comment added by Massy (talk • contribs) 10:43, 1 December 2012 (UTC)

Edit: Graag verwijs ik naar het volgende artikel door van Reenen en Huijs, waarin wordt gesproken over post-palatale fricatieven: http://www.meertens.knaw.nl/taalentongval/artikelen/Reenen_Huijs.pdf . Misschien schrijven we beter de [̠ʝ] dan beter met een mintekentje eronder? — Preceding unsigned comment added by Massy (talk • contribs) 11:54, 2 December 2012 (UTC)

ʌu

I changed the ʌu examples from 'bout/faun' to 'fout/nauw'. The latter are far more common words (I don't think I've ever used 'faun'...). Jalwikip (talk) 14:12, 2 April 2009 (UTC)

<g>

I don't speak Dutch, but I am attentive to pronunciation and deal with phonetics pretty often. Listening to singers, commentators and the like, I've always had the impression that: 1. Speakers from the Netherlands always pronounce <g> as a voiceless uvular or velar fricative ([x], [χ]); 2. Speakers from Belgium always pronounce <g> as a voiced velar fricative ([ɣ]). Nobody I heard ever pronounced <g> as a palatal ([j]).

I see more or less the same has already been stated by other users in the section titled /ɣ, x, χ/ "above". It has also been stated that the <g> is pronounced as a [ɣ] or a similar voiced sound by Netherlands speakers only on old recordings etc.. Nevertheless the article currently asserts that Northern Dutch pronounces <g> as [ɣ] and Southern Dutch pronounces <g> as [j], and only mentions the [x] and [χ] pronunciation of <g> in a footnote as typical of "some dialects".

The sound files given for <gaan> in this article are given as examples of the alleged regularity, and nevertheless they confirm my impression instead. The Netherlands example is supposed to be a [ɣ], but to me it sounds like a [χ] or even a uvular vibrant (definitely voiceless in any case). The Belgian example is supposed to be a [j], but to me it sounds like a [ɣ].

Hopefully, the confusion is not so great as it seems. Possibly part of the problem is the difference between traditional dialect features and current educated speakers' practice in the two countries. Even then, the latter should be featured more prominently, especially as it is the reality many people are likely to encounter. --91.148.159.4 (talk) 14:13, 23 May 2009 (UTC)

• is supposed to be a [ɣ], but to me it sounds like a [χ] or even a uvular vibrant (definitely voiceless in any case): I agree completely the [ɣ] example is most definitely a voiceless sound. The Dutch Consonants with Example Words table should be corrected. Fnugh (talk) 14:07, 5 February 2010 (UTC)

It's voiceless in the north where voiced fricatives become voiceless, right? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 18:17, 5 February 2010 (UTC)

I live and grew up in Eindhoven and I pronounce it closer to [ʝ] or perhaps [ɣ], but certainly not voiceless. There is a clear distinction for me between the consonants in lachen [lɑçə]/[lɑxə] and lagen [lɑːʝə]/[lɑːɣə]. So you should be careful in making assuptions that appear to be valid for you or your area. Dutch is less homogeneous than you'd think. --CodeCat (talk) 23:31, 5 February 2010 (UTC)

Well, i really don't know, tbh, I'm only saying that the example from the Dutch Consonants with Example Words table is wrong as it doesn't actually have a [ɣ]. —Preceding unsigned comment added by Fnugh (talk • contribs) 15:23, 9 February 2010 (UTC)

Historical sound changes

Dutch (with the exception of the Limburg dialects) did not participate in the second Germanic consonant shift except for the last stage - compare

/-k-/ > /-x-/: German machen vs. Dutch maken (help·info), English make /-p-/ > /-f-/: German Schaf vs. Dutch schaap (help·info), English sheep /-t-/ > /-s-/: German Wasser vs. Dutch water (help·info), English water /-θ-/ > /-d-/: German das, Dutch dat (help·info) vs. English that

I don't understand this list. It is mentioned that Dutch did not participate in the sound changes except in the last stage..but the last example shows according to the comparison /-θ-/ > /-d-/ that Dutch actually didn't participate. Even if they did participate it should be /-θ-/ < /-d-/. This is a little confusing to me. Someone care to explain please? 81.68.255.36 (talk) 14:22, 20 February 2010 (UTC)

From studies of older Germanic languages (including Old Dutch itself) it is known that the words in question contained /θ/ and not /d/ originally. So we can rule out the possibility of /d/ > /θ/ in favour of the opposite. What may be confusing is that the change /θ/ > /d/ is commonly thought of in context of the High German consonant shift, but it affected ALL continental West Germanic languages, not just High German. So calling it a High German shift in this case is an unfortunate (historical) misnomer. --CodeCat (talk) 10:53, 21 February 2010 (UTC)

What I find confusing is that it seems the sound that is 'greater' (k > x) prevails in Dutch while in the last case it is the opposite. I am no expert on this sound change so perhaps I'm not seeing something perfectly obvious for other people, but I'd like to understand. 81.68.255.36 (talk) 12:31, 21 February 2010 (UTC)

Oh! Noo it doesn't work like that. In linguistics, > means 'becomes' or 'changes into', and < means 'originates from'. Read it like an arrow. It has nothing to do with any sound being less/greater than another! --CodeCat (talk) 12:37, 21 February 2010 (UTC)

Ah, of course. Thanks for reminding me :P I would have had absolutely no problems with it if I was reading some etymology, but this...oh well, thanks again! 81.68.255.36 (talk) 12:43, 21 February 2010 (UTC)

Other diphthongs

I'm surprised there is no mention of several additional (although rare) diphthongs in Dutch. /iu/ nieuw, /ui/ moeite, /eːu/ leeuw, /aːi/ kraai, /oːi/ mooi. Is there any reason why these aren't listed in the table? CodeCat (talk) 17:46, 20 May 2010 (UTC)

The current article assumes they are phonemically sequences of vowels and approximants, even though phonetically they may be realised as your transcriptions suggest.

They are mentioned explicitly now, though. I think the current article is fine like this. Barefoot Banana (talk) 12:14, 30 August 2023 (UTC)

Dutch "ER"

Are there any dialects of Dutch that pronounce final "er" just like in German, which drops the "R" sound? —Preceding unsigned comment added by 70.112.242.125 (talk) 22:24, 4 March 2011 (UTC)

I think the southeastern dialects, mostly Limburg and the far southeast of Noord Brabant, have a pronunciation like that. CodeCat (talk) 10:52, 5 March 2011 (UTC)

In Den Haag you can hear that quite a lot! --84.10.140.55 (talk) 21:24, 21 March 2012 (UTC)

There are some dialects that can have a R-vocalization similar to the one found in German (along with R articulated similarly to German); it's often found in Southern Gelderland (Nijmegen), The Hague area - Maury

I think that both of you are overestimating the presence of [ɐ] in the Hague. It's the traditional dialectal pronunciation, but recent studies (see e.g. Sebregts (2014)) have shown that it's very rarely used (in 2% of cases or less). Unless, of course, you visited a neighborhood in which the broad form of the dialect is still spoken. [ɐ] is a very marked realization of /ər/ and it's dying out. I don't know about Limburg and other accents near the German border, but I wouldn't expect to hear it in the Hague very often. Or at all, depending on who you're talking to. Also, remember that /r/-dropping after /ə/ (which can happen occasionally in Standard Dutch) and /ər/-vocalization to [ɐ] are two entirely different processes, not least because their outcomes sound completely different. Mr KEBAB (talk) 03:24, 25 March 2017 (UTC)

It is one of those things where a phenomenon has become so notorious that speakers become hypersensitive to the issue. In this case ground zero is the infamous character Haagse Harry, whose many invectives include kankɐ hoeɐ. Not wanting to sound like him, more and more hagenezen are consciously steering clear of ɐ. When academics such as Sebregts aren't listening, I suspect ɐ is still going strong. 2A01:CB0C:CD:D800:6C58:E74F:F368:C5D9 (talk) 13:23, 23 March 2023 (UTC)

About the /ʝ/ sound...

I'm confused. I had thought that this sound: /ʝ/, sounds rather like the "g" in gin, and not like the gutteral sound in the Scottish "loch". This sound also appears in Spanish with that sound. Can someone verify this? —Preceding unsigned comment added by 70.112.242.125 (talk) 01:48, 6 March 2011 (UTC)

Since it's a sound that doesn't occur in English, it's hard to describe it. [ʝ] is a lot like the y of English, though it's more constricted so that it sounds a little like a soft g sound. To make it more confusing, there's also some dialectal variation, though I'm not sure what dialect you're learning. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 02:16, 6 March 2011 (UTC)

Well, I read about in the article on Dutch Phonology for Wikipedia that in southern dialects and in Belgian Dutch, there is a distinction between the Dutch "g" and dutch "ch". It said that the "g" in those areas has a /ʝ/ sound to it. But that would mean that it's not guttural at all, and so I was confused about that. —Preceding unsigned comment added by 70.112.242.125 (talk) 02:34, 6 March 2011 (UTC)

Yeah, in those contexts it's palatal. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 02:57, 6 March 2011 (UTC)

I believe this is a mistake on the part of some linguists. The /ç/ and /ʝ/ in all the recordings are clearly front-velar, not palatal. See comments above. Benwing (talk) 08:10, 8 February 2012 (UTC)

@Benwing: [ç, ʝ] are perfectly acceptable broad transcriptions (with retracted diacritics implied by the context) of front-velar fricatives. Mr KEBAB (talk) 11:54, 18 December 2017 (UTC)

IJ and EI

I am currently learning Dutch, and from what I've listened to, both of these letter combinations sound more like the English word "eye", and rather like the "ei" in German. Is anyone else confused about this? —Preceding unsigned comment added by 70.112.242.125 (talk) 01:50, 6 March 2011 (UTC)

Well yeah, /ɛi/ is certainly not the way it's pronounced in the Netherlands. It's either /æi/ or /æɪ/. The same way, /œy/ is not so accurate. It's more like /ʌy/ or /ɐy/. /ʌu/ is also not the way it's pronounced, /au/ is used by the most Dutchmen. That said, of course that's the way people speak here, and it's different from Standard Dutch pronunciation. It doesn't mean that this article is wrong.--84.10.140.55 (talk) 21:22, 21 March 2012 (UTC)

The recordings

'ch' and 'g' are described as velar in the Netherlands and palatal in Belgium, but what is recorded is clearly uvular for the former and velar for the latter. Also, many of the files seem to be cut too closely at the onset, to the point where initial [p, t, k] sound like [b, d, g] (my native language is Polish, so I don't mistake the lack of aspiration for voicing). 46.186.37.98 (talk) 01:26, 19 May 2011 (UTC)

As example for the 'ui' sound, the vowel table contains 'buit' (translated as 'booty', meaning loot). The sound file is also named Nl-buit.ogg. However, the word pronounced in the sound file is clearly 'bui' (rain shower); there is no silent 't' in Dutch. To correct this, the sound file should be renamed. — Preceding unsigned comment added by 99.65.138.88 (talk) 23:37, 20 October 2011 (UTC)

I listened to the recording and clearly hear the final 't' pronounced. −Woodstone (talk) 05:43, 21 October 2011 (UTC)

Different treatment for common Dutch and common Belgian pronunciation

I know the issue of Belgian vowel pronunciation has been discussed several times. But to me it hardly seems fair that the diphtongal quality of the vowels oo, ee & eu, which is common practice in the Netherlands is mentioned, be it with a footnote like this "Pronounced as long vowels in Belgium, but as narrow closing diphthongs in the Netherlands. The transcription /eɪ øʏ oʊ/ for this diphthongal pronunciation is non-standard and used here for the sake of clarity.", but the monophtongal pronunciation of the vowels ij, ei, ui and au/ou, which is common in large portions of north Belgium is not. I mean: if non-standard, but common dutch pronunciation is mentioned, shouldn't the same be done for common, non-standard Belgian pronunciation. A similar footnote could easily be made for these sounds, just my two cents Nychus (talk) 15:59, 1 November 2011 (UTC)

Sounds reasonable. You got a source? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 02:02, 26 December 2011 (UTC)

I think this distinction could be phonemic as there are a few (non-native) words that have /ɛː/ in all dialects, distinct from /ɛi/, /ɛ/ and /eː/. It is found in words like "crème", which certainly does not rhyme with "hem" in the north due to the distinctive vowel length, nor does it have the vowel quality of "heem" or the diphthong of "lijm". If /ɛi/ becomes /ɛː/ in the south, then presumably the phonemes /ɛi/ and /ɛː/ merge there, but I don't know if this is true. Does "crème" rhyme with "lijm" in Belgium? CodeCat (talk) 18:16, 9 January 2013 (UTC)

Yup, in the dominant pseudo-standardlanguage it almost always is. (/krE:m/ en /lE:m/). In dialects there will be more variance (as lijm may be pronounced /lim/ etc, in undiftongising areas) but that would complicate things too much and not add much to the article either. But in casual 'tv-flemish' both will usually sound identical as /E:/. 109.132.200.202 (talk) 14:35, 12 February 2013 (UTC)

I think the mouth opens slightly wider in words with IJ, so that some distinction in sound quality is maintained. 2A01:CB0C:CD:D800:6C58:E74F:F368:C5D9 (talk) 13:18, 23 March 2023 (UTC)

ʃ as a native phoneme?

It may be (once again) a difference between Southern and Northern accents, but I've always heard the letterpair -sj- realized as [ʃ] in e.g. meisje "girl" [mɛːjʃə]. Maybe with no lip rounding though, unlike in French; but definitely farther back in the mouth than [s]. The presence of the -ei- diphtong just before may be significant: the sound seems different to me in kaasje "little cheese". Tonymec (talk) 23:26, 19 August 2012 (UTC)

You may be partially correct. In words as sjouwen and sjaal I also hear [ʃ], but not in all those diminutives with -sj- (huisje, baasje, etc.). It might be because the j is added as part of a regular suffix. Meisje is a bit of a special case. It is a diminutive, but no longer recognised as such (it comes from maagd > maagdje > meisje, compare german Magd > Mägdchen > Mädchen). I think that as long as the form ending on -s (like with kaas > kaasje) still occurs it is more or less pronounced as [sj] or [ʃj], else as [ʃ]. PiusImpavidus (talk) 15:51, 20 August 2012 (UTC)

It's also possible that you 'hear' it that way because your brain recognises how the word is split up. [sj] natively only occurs across morpheme boundaries, and is fairly marginal otherwise, it only occurs in loanwords word-initially. I would say that the difference between [sj] and [ʃ] is allophonic, so it's a matter of individual pronunciation and not one of phonemic contrast. There is a wide range of pronunciations too... I wouldn't be surprised if you would hear [sj], [sʲj], [sʲ], [ɕ] or [ʃ] depending on the speaker. CodeCat (talk) 00:36, 21 August 2012 (UTC)

Meisje is not the diminutive of maagd, but of its doublet meid. 77.248.144.142 (talk) 20:19, 14 April 2022 (UTC)

I agree with @CodeCat, [ʃ] only occurs in loanwords (the fusion of /s/ and /j/ doesn't count). Moreover, it does not have the distribution that true native phonemes have. I can't readily think of examples in word-medial or word-final position.

It is true that it does not alternate in words such as meisje or chocola, but I would wager that children, during acquisition, interpret this sound as a consonant cluster of /s/ and /j/. Barefoot Banana (talk) 11:32, 30 August 2023 (UTC)

Voiceless palatal stop info needed

If articles such as Tjeerd Oosterhuis and Trijntje Oosterhuis are going to use transcriptions with the IPA [c], could someone with the requisite knowledge please add the relevant info about this phone to this article (whether in an "Allophones" section or in the chart) and/or the Help:IPA for Dutch and Afrikaans page? Alternatively, if the sound is not really a [c], could someone then change the transcription in the two aforementioned articles? Thanks. —  AjaxSmack  04:08, 24 February 2013 (UTC)

I see no reason to use the symbol [c] for these sounds. Rendering by [tj] is adequate, similar to the use of [tʃ] in English. −Woodstone (talk) 17:19, 24 February 2013 (UTC)

I don't know anyone who doesn't pronounce it as a dental, with at least some palatality or frication. [tʲ] or [tɕ] would be more accurate. CodeCat (talk) 18:01, 24 February 2013 (UTC)

Agreed. I will change the transcriptions of these two names to having [tɕ], because there is definitely some frication in the ordinary pronunciation in Standard Netherlandic Dutch. Barefoot Banana (talk) 12:19, 30 August 2023 (UTC)

On second thought, I actually think [tʃ] is more appropriate, but that's lacking in this article as a realisation. Postalveolar pronunciation are given for /sj/ clusters, though.

And another issue is that the transcriptions do not seem to be as phonetically accurate as possible anyway, e.g. using [r] in the coda despite that being very rare in the Netherlands. Barefoot Banana (talk) 12:23, 30 August 2023 (UTC)

Changing the ad hoc symbols

If it is already stated in the article that the symbols used are assembled ad hoc, and are not the standard, but probably should be, why not change the /œy̯/ diphthong into /ɐɵ̯/? Belgian Dutch /œy̯/ is OK for the in transcribing the starting point if it is rounded, but perhaps the end point should again be /ɵ̯/. /ʌu̯ / is also somewhat debatable, but OK, if the starting point is somewhere halfway between /ɐ/ and cardinal 14. Change the /i̯/ into /ɪ̯/ if the vowel is more open and not as tense. --79.175.111.158 (talk) 16:11, 24 March 2013 (UTC) Statistics

Truth about Dutch G and more..

It's all here, with the much improved Canepari's CanIPA:

http://venus.unive.it/canipa/pdf/HPh_17_Europe.pdf

Go to Dutch section:

http://venus.unive.it/canipa/pdf/HPh_10_Consonants.pdf , the consonants and respective IPA symbol

http://venus.unive.it/canipa/pdf/HPh_08_Vowels.pdf , vowels and respective IPA symbol

It's true what many of you stated, G voiced is not anymore in Dutch, and since a long long time!

G is /X/, while CH is closer to /x/ (though not exact) in all the dialects now, and no, they haven't merged,or at least yet.. check Canepari's CanIPA, normal IPA is just out of date, and it has nowadays lots of mistakes, while this one is a very well made one

Also you can see there the truth about their R, the fact that the normal R realisation is not a trill but an uvular flap, but in the mediatic accent we find /ʁ/ word initial, and uvular /ɹ/ in other positions /s/ and /z/ are dental! No like English way, and that explains the doubt many had about it and the many real realisation of the vowels also are to sort out... A Dutch speaker will surely confirm what I'm telling you and Canepari about it, there's lots of things wrong in this article..

Explenation: In the Dutch language there are only 5 vowels; A E I O U. (short vowels) You can combine characters to make them longer (it sounds a bit louder and a bit longer); AA EE IE OO UU (long vowels) And final, you can combine letters to get a few more; AU EI IJ OU(the 'ei' and 'ij' sound the same and the 'au' and 'ou' too) (combined vowels) When counting those (combined) characters you get 14 vowels, but only the first vife count as a vowel, so there are only 5!

In dutch there are a few more ways to speak those a diffrent way: - When there is only 1 non-vowel character next to the single fowel (A E I O U) you speak it out as a short vowel. - When there is only 1 non-vowel character next to the vowel and then again a vowel, you pronounce it as a long vowel like this: A = AA, E = EE, I = IE, O = OO, U = UU - When there is no character next to a vowel you say it as a long or short vowel: A = AA, E = U, I = IE, O = OO, U = UU - When a vowel is used as second letter with a single non vowel character next to it, you speak it out as a long vowel: A = AA, E = EE, I = IE, O = OO, U = UU exceptions: First/third character is a vowel: long vowel / combined vowel (see all vowels) Combined words: 'betekenis' (meaning) be-teken-is = bU-tEEkUn-Is / 'verlaten' (leaving) ver-lat-en = vUrlAAtUn

A few more examples: bot (bone) = bOt, short boot (boat) = bOOm, long boten (boats) = bOOtUn, long and short

very rare there are the following characters used: ë = E (ëe = EE), ü = U/UU, é = EE, è = E With these you have 18 characters (some people see the I-grec as a vowel as well)

There are many exceptions where the rules of speaking are not correct, don't claim those statements wrong by this text! I am Dutch, please do not mind my fails in English. — Preceding unsigned comment added by 213.126.0.218 (talk) 14:50, 3 May 2013 (UTC)

I don't think any source that considers G to be devoiced in all Dutch varieties can be considered reliable. I am a native speaker, who pronounces G as voiced /ɣ/, and I am sure that many Belgians will agree. The fact that the document treats Flemish as distinct from Dutch is also very suspect, considering that they are a unified language and there is a dialect continuum between them. Treating the national border as a linguistic border strikes me as very un-scientific. CodeCat (talk) 15:17, 3 May 2013 (UTC)

Need help with pronunciation of two names

I need help with this, can you fill in these two names for me?

Anest Tadari (nl

Piet Nekilt (nl

Razdower (talk) 16:56, 10 May 2013 (UTC)

The only Dutch part of that is /pit/... the rest seems foreign to me. CodeCat (talk) 01:56, 11 May 2013 (UTC)

The first one may be [ˈaːnəst tɑˈdaːri], but I'm not sure. Peter238 (talk) 18:04, 10 December 2015 (UTC)

How to keep the information synchronised with Dutch language?

The table of vowels and consonants was recently cleaned up in Dutch language to show only the phonemic contrasts. The same should be done here too. But what can be done to prevent the two articles from drifting apart in the future? CodeCat (talk) 18:15, 3 November 2013 (UTC)

Short vowels in northern dialects

Northern /ɑ ɔ/ have some sort of a back secondary articulation. Most likely they're pharyngealized. In addition, /a/ has a somewhat nasal quality. --Helloworlditsme (talk) 09:23, 28 November 2013 (UTC)

What do you mean by "northern"? To me it means anything above the river Meuse, but I can't really connect what you say to anything I'm familiar with. Can you provide a source for this? CodeCat (talk) 14:45, 28 November 2013 (UTC)

I mean most provinces excluding Zeeland, southern North Brabant and probably a big part of Limburg. I don't know the farthest north and farthest east accents. Listen to at least the first minute of this video: http://www.youtube.com/watch?v=HLcb4EkWGLw . Some of the commentator's pronunciations are: [ˈvrɛi̯s̠xɔˤp] [ˈbɔˤs̠fɛɫt] [z̠oʊ̯ɫɑˤŋ] [xãˑt]. I'm not sure about the stress in the third example, but it's not relevant here. There are more instances of [ɑˤ ɔˤ ã] in the first minute, but since my Dutch is bad, I won't attempt to transcribe them. If I try to imitate the first three pronunciations, /ɑ ɔ/ must have a secondary articulation from the same place as Russian /ɫ/ (which is pharyngealized). The last must be nasalized. There's no way to achieve the same sounding vowels otherwise. But it's true that Eddy Poelmann (the commentator in the video) is somewhat inconsistent with pharyngealizing low back vowels, but on the other hand, his /a/ nearly always has a nasal quality. If I had a source I would've already put it in the article :) But that's what the talk page is for, to discuss things that should be added to the article, but which need to be sourced first. --Helloworlditsme (talk) 17:49, 28 November 2013 (UTC)

Inline citations

While I agree with most of the article, it really needs more inline citations. Peter238 (talk) 12:58, 16 October 2014 (UTC)

Short /ɔ/

Isn't this commonly [o], at least in Hollandic Dutch? I'm German, fluent in Dutch, and I hear the Dutch short "o" almost as close as my personal realization of German /ʊ/. It may be that I use [o] for German /ʊ/, and [ɒ] for German /ɔ/, but the thing is that I don't perceive the same difference between both languages in front vowels. So, to put it another way, isn't Dutch /ɛ/ much more open than Dutch /ɔ/? Now, the speaker who says "bot" in our example words uses /ɔ/. I think my German /ɔ/ is indeed even more open, but it's open all right. However, I think that many speakers use a much close vowel instead. (My main experience is with Hollandic Dutch, South Holland in particular.) — Preceding unsigned comment added by 92.204.88.105 (talk) 10:02, 24 January 2017 (UTC)

I grew up in an area in the east of the country where people still distinguish between /ɔ/ and /o/. Basic words such as 'hok' and 'bok' are pronounced /hɔk/ and /bok/, respectively. There are some minimal pairs, such as pop, pronounced /pɔp/ when it's music and /pop/ when it mans doll . But in standard Dutch (at least in the Netherlands) there is no short /o/ at all, it is always /ɔ/. But it is not as open as the o in English 'lot', which is /ɒ/. Eti erik (talk) 14:46, 17 December 2017 (UTC)

In standard Dutch there is very little independent vowel length. So usually the contrast is only between [oː] and [ɔ]. And similarly with [aː] and [ɑ]; [iː] and [ɪ]; [eː] and [ɛ]; [ʏː] and [ə];. A change in quality is accompanied by a (slight) change in length. Explicitly long [ɔː] is limited to loan words. −Woodstone (talk) 16:29, 17 December 2017 (UTC)

@Woodstone: Small correction: the contrast is between /øː/ and /ʏ/. [ʏ] and [ə] can be considered allophones of one phoneme, but they usually aren't. Mr KEBAB (talk) 11:50, 18 December 2017 (UTC)

"Standard Dutch"

This article should cover Dutch, not just whatever "Standard Dutch" might be. The actual Dutch standard, as managed by the Dutch language union, is strictly a spelling standard. There is no standard for Dutch pronunciation whatsoever. Note that many variations in pronunciation are discussed in this article, and I strongly oppose removing those. In particular, there's a paragraph in the Vowels section that covers the pronunciation of Antwerpian. CodeCat (talk) 21:18, 24 March 2017 (UTC)

(Thanks for starting the discussion yourself, even though you didn't have to do it.) Before I start I must say that this article has severe sourcing issues, anyone can remove unsourced content, and it's on the person who wants that content to stay to find a source themselves (see WP:OR). I'm sure you're aware of that, I'm just reminding it for the sake of clarity.

You're wrong in saying that there's no spoken Standard Dutch. There very much is such a thing; in fact, there are two prestige varieties of spoken Dutch: Standard Netherlandic and Standard Belgian. See Collins & Mees (2003:5–6) (2003:4–5) and also Booij (1999:2), who says that he describes 'the phonology of standard Northern Dutch' in his book.

The answer to 'what is spoken Standard Dutch?' is simple: it is what reputable sources consider it to be. When they disagree with each other, all that we need to do is to point that out. The fact that there's no institution that regulates Dutch pronunciation is irrelevant; there's also no such institution in the case of German, yet there obviously exist at least three prestige varieties of spoken Standard German, with the 'Northern German' one being obviously at least somewhat more prestigious than the other two (Swiss and Austrian).

We should not cover Antwerpian or any other non-standard dialect in this article, only Standard Dutch and divergent regional pronunciations. If you have a source for the Antwerpian accent in Standard Dutch, by all means use it. We cannot be sure at all that when the source for the Antwerpian dialect writes [ɛi̯~ɛə̯~eə̯], that that is how people from Antwerp pronounce Standard Dutch /eː/. The same is true of standard /ɛi̯/ - we don't know whether Antwerpian Standard Dutch uses [aə̯] for it or not (I guess it doesn't)... unless the source does clarify the issue?

Chief, relevant references to non-standard regional dialects are perfectly acceptable. We can create Regional Dutch phonology or a similar article and discuss non-standard regional dialects there. We need a separate article for spoken Standard Dutch itself and not to pretend that it doesn't exist when it obviously does. Wikipedia goes with what reputable sources say. Mr KEBAB (talk) 22:13, 24 March 2017 (UTC)

We're also listing two pronunciation dictionaries in the Further reading section. If spoken Standard Dutch didn't exist (which I've already proven to be a false statement), it wouldn't be possible to list the most accepted pronunciations of thousands of words. Of course, the most recent one that we list is 17 years old; there also are some words (e.g. loanwords, foreign names) which don't have very well established pronunciations, which by necessity makes pronunciation dictionaries partially prescriptive. But, AFAIK, such words are in the minority. Mr KEBAB (talk) 23:29, 24 March 2017 (UTC)

I'll add even more: Gussenhoven (2007) lists more pronunciation dictionaries, whereas Verhoeven (2005) titled his article Belgian Standard Dutch. Both are available for free online. Mr KEBAB (talk) 00:14, 25 March 2017 (UTC)

There's nothing that suggests that this article can be only about standard varieties. As I said before, the name of the article is "Dutch phonology". So then if there is such a thing as Standard Dutch pronunciation, why don't you create a Standard Dutch phonology article and put the information there? I strongly oppose the removal of information about dialectal differences that is currently present in this article. CodeCat (talk) 01:24, 25 March 2017 (UTC)

Because we already have this article that is already mainly about Standard Dutch, it also has sourcing problems. It's better and more logical to improve this article, rename it to 'Standard Dutch phonology' (following Standard German phonology) and move inappropriate dialectal stuff (there's really not a lot of it here) to Regional Dutch phonology (or however you want to call it). Mr KEBAB (talk) 01:47, 25 March 2017 (UTC)

Because this article doesn't contain a lot information about non-standard regional dialects, the reason for which I want to restrict its scope now is to prevent that from happening in the future. There are good reasons to do so, which are readability and not confusing laymen. Also, lumping the standard language and non-standard dialects together just doesn't happen in case of other 'X phonology' articles on Wikipedia, we keep them separate. I don't see why Dutch should be a special case. Mr KEBAB (talk) 14:42, 25 March 2017 (UTC)

There'd also be a problem of WP:CONTENTFORK if I followed your advice and created Standard Dutch phonology myself instead of renaming this article. Mr KEBAB (talk) 14:47, 25 March 2017 (UTC)

I don't want this article to be only about "Standard" Dutch, but about Dutch in general. That's what its name is, after all. So if you want to focus exclusively on "Standard" varieties, you'll have to make your own article. CodeCat (talk) 17:59, 25 March 2017 (UTC)

Can you actually address what I'm saying? "I don't want that" is not a valid argument; also, please don't behave as if you WP:OWNED the article. If I were to create a separate article, at least 30% (probably more) of this article would have to be moved there in order not to violate WP:CONTENTFORK. Then, this article would have to be renamed in order not to confuse people (we have other articles, e.g. French phonology, in which only the standard pronunciation is discussed, yet the word 'standard' is not included in the title). Perhaps we're starting to split hairs? Mr KEBAB (talk) 18:21, 25 March 2017 (UTC)

The argument I gave is that this article's title necessitates the scope: "Dutch phonology" implies Dutch in general, not only certain varieties of Dutch. I don't own the article, but there's currently no consensus for making it only about "Standard" Dutch. I suggest starting a discussion in a wider forum such as Wikiproject Languages, in order to find a consensus on what to do. CodeCat (talk) 18:50, 25 March 2017 (UTC)

Yes, I already did. I'm not doing anything unless there's a consensus. Mr KEBAB (talk) 18:56, 25 March 2017 (UTC)

Am I not invited? CodeCat (talk) 19:21, 25 March 2017 (UTC)

It's just a link to this discussion (which IMO should be kept in one place). Here it is. I just forgot to update the description. Mr KEBAB (talk) 19:38, 25 March 2017 (UTC)

I don't know whether it is better to move this article to Standard Dutch phonology/Standard Dutch phonologies or retain its name and treat nonstandard phonological systems as well. In case of the latter choice I suggest that nonstandard sytems be discussed in a separate section near the end of the article. Love —LiliCharlie (talk) 19:33, 25 March 2017 (UTC)

Thanks for the input. The thing is that the article simply doesn't discuss non-standard dialects in any depth. There are a few sentences and that's all. The article also lacks inline citations in many places. That's why I'm proposing what I'm proposing. Mr KEBAB (talk) 20:55, 25 March 2017 (UTC)

I'm not sure what you mean. There's lots of detail about all kinds of phonetic variations. CodeCat (talk) 21:57, 25 March 2017 (UTC)

I was only able to find those two sentences about Antwerpian which you say shouldn't be removed. The rest could be easily interpreted as describing both Standard Dutch spoken with a local accent (we should cover those here) and non-standard dialects, which is fair enough to me, though maybe some people would want to rephrase some of them. I wouldn't remove e.g. the part about the "sharp-long" and "soft-long" e and o because it's relevant to the spelling reform from nearly 100 years ago. So the article is already almost entirely about SD with non-standard dialects being mentioned here and there, mostly in the context of SD. Mr KEBAB (talk) 00:35, 26 March 2017 (UTC)

The distinction between standard and regional dialects can be confusing for lay readers. Perhaps the thing to do is focus primarily on Standard(s) and include sourced information about regional dialects (or even sociolects) where relevant, but also treat Regional Dutch phonology as a sort of subpage to this one. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 23:44, 30 March 2017 (UTC)

/ɦ/

My edit to replace /ɦ/ with /h/ in the phonemic transcription of two Dutch sentences was just reverted. "ɦ" has been introduced at some point based on Gussenhoven's contribution on Dutch in the 1999 Handbook of the IPA and someone then "corrected" all the "h"s in Dutch pronunciations in wikipedia. However, Gussenhoven himself in 2007, in an article entitled "What is the best transcription for Dutch", already referenced in the article, simply has "h" in table 2. In the (English language) abstract he writes the guiding principles: "First, the phonetic principle dictates that the phonetic value expressed by the symbol should not deviate unnecessarily from the sound it is meant to represent." I had never thought the Dutch and English (or German!) "h"s different, and after 30 years in the US with a most horrible Dutch accent no one seems to have noticed that my "h" is off. Thus, "h" approximates the Dutch pronunciation just fine. Gussenhoven's third principal is key: "the typographic principle requires that an alphabetic symbol is to be preferred to an exotic one." ɦ is obviously exotic, and there is no reason to use it in phonemic transcriptions. It's the principle that allows Prinknash to be transcribed as /ˈprɪnɪdʒ/ rather than /ˈpʰɹ̠ɪnɪdʒ/ (not that the first one makes more sense). Afasmit (talk) 03:44, 9 December 2017 (UTC)

Afasmit wrote: “It's the principle that allows Prinknash to be transcribed as /ˈprɪnɪdʒ/ rather than /ˈpʰɹ̠ɪnɪdʒ/ ...”

This is laid down on page 159 of the Handbook of the International Phonetic Association as principle 4 (a) of The Principles of the International Phonetic Association:

“When two sounds occurring in a given language are employed for distinguishing one word from another, they should wherever possible be represented by two distinct symbols without diacritics. Ordinary roman letters should be used as far as practicable, but recourse must be had to other symbols when the roman alphabet is inadequate.” Love —LiliCharlie (talk) 06:05, 9 December 2017 (UTC)

Indeed, in Standard Dutch in both the Netherlands and Belgium there is only one way to pronounce "h", with, as always, differences in pronunciation between individuals and dialects. There are no words that have distinct meanings by the use of [ɦ] or [h]. (An unreferenced bit in the text says that in Zealandic and West Flemish the "harde" en "zachte" g are rendered as [ɦ] and [h] respectively.) Our current text says, again without a reference, that "Some speakers pronounce /ɦ/ as a voiceless [h]" (no idea why that is not written [ɦ] and [h]). It would be interesting to know if this is dialectical or personal, like the pronunciation of /r/ in Dutch is.

Thanks for bringing this to the talk page Afasmit. The typographic principle that Gussenhoven puts forth is perhaps just his personal opinion. Here at Wikipedia, where typographic constraints are much less stringent than the context that he was writing in, there is no reason to be constrained by Gussenhoven's typographic principle if we can be more phonetically accurate.

The fact that native English speakers, who are not primed to notice the distinction between the two sounds, have not taken note of whether your h's are strange doesn't really amount to much evidence. If someone can provide phonetic sourcing that shows that this sound in Dutch is typically pronounced voiceless, as it is in English, that would help your case that h is more appropriate than ɦ. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 14:23, 9 December 2017 (UTC)

Gussenhoven seems to be the one that introduced the idea that we pronounce /h/ as [ɦ]. Perhaps that is his personal opinion too? And Dutch people, perhaps with the exception of those still speaking the Zealandic and West Flemish dialect, are also not primed to hear a difference. At any rate, the main point here was not the debatable phonetic [ɦ], but the use of "ɦ" in phonemic transcripts. English wikipedia has adopted the standard to give a /phonemic/ transcript for names pronounced in English and [phonetic] transcripts of names as pronounced in other languages; that's why I used the Prinknash example, transcribed as /ˈprɪnɪdʒ/ even though most native and foreign speakers can hear that it is not pronounced with an [r]. Afasmit (talk) 22:06, 10 December 2017 (UTC)

I'm at a loss as to why you would think an analysis that appears in the Journal of the IPA might have something inaccurate simply because, in another article, the author has the opinion that a phonemic representation might best stick to letters one can type on a keyboard, rather than strive to be completely phonetically accurate at the expense of typographic expediency. That is shoddy analysis on your part and I think you can do better than that.

Then again, maybe you can't. If you can't tell the difference between an evidence-based phonetic study and an opinion, and if you can't provide the necessary sourcing that I've called upon you to provide to help us settle the issue, then there's no point in continuing this exchange. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 00:15, 11 December 2017 (UTC)

Here's hoping that someone who can read and use logic will join the debate at some point. Afasmit (talk) 02:16, 11 December 2017 (UTC)

@Afasmit: There's nothing wrong with using /ɦ/ in phonemic transcription. Your comparison with English /p/ and /r/ being transcribed as /pʰ/ and /ɹ̠/ is inaccurate because /ɦ/ is a diacritic-free symbol, whereas /pʰ/ and /ɹ̠/ aren't. There also are problems with using /pʰ/ as a phonemic symbol, since aspiration is lost in many positions (indeed, in many accents such as GA it's present only syllable-initially in stressed syllables). Using /ɹ/, on the other hand, is perfectly appopriate in English (indeed, to me it's much more appropriate than /r/).

I had never thought the Dutch and English (or German!) "h"s different, and after 30 years in the US with a most horrible Dutch accent no one seems to have noticed that my "h" is off. Thus, "h" approximates the Dutch pronunciation just fine. This isn't Help:IPA/Dutch or your personal blog but an article about Dutch phonology. There's nothing wrong with using /ɦ/ as a phonemic symbol because we have at least one source to back it up (well, two - Verhoeven (2005) is another one) and because it's correct phonetically (see Collins & Mees (2003)). Also, this insistence on using 'simple' symbols for phonemes strikes me as very old-fashioned. It's 2017.

Provide a source that states that Dutch /ɦ/ is voiceless and/or a list of sources that use /h/ in phonemic transcription and then we can talk. Replacing phonemic symbols with 'simpler' ones just because they're easier to type is a bad idea. Mr KEBAB (talk) 09:02, 11 December 2017 (UTC)

I agree. We should use the symbols that actually represent the phoneme. Rua (mew) 12:50, 11 December 2017 (UTC)

Thanks Mr Kebab for reading and understanding my arguments. I’ll be glad to give up my quixotic attempt to address an insignificant issue, though I still believe that the use of ɦ to explain how Dutch words with an h are pronounced almost always will confuse rather than help Wikipedia users. My own confusion stemmed from a few things, including: 1) [h] and [ɦ] are allophones in standard Dutch (and English), and Dutch speakers don’t perceive the /h/ in English or German different than in their own language, and vice versa. The latter does not count as it is my and anyone I asked their personal experience. It’s challenging to support this with articles as it is hard to find articles just using “h”! If I read table 4 of this article correctly, my experience may not be unfounded, as American men typically fully voice their h’s (i.e. say [ɦ]) when embedded in a short carrier utterance and a minority of American women do as well. The primarily sex-linked difference was confirmed for initial [ɦ] in this article, though the fraction is much lower (30% of initial instances are voiced by men). Similarly, there is high variability of the use of [h] and [ɦ] in Dutch according to our article here. 2) One of the foremost Dutch linguists introduced the [ɦ] in the IPA handbook and years later wrote a [peer reviewed] article on the preferred transcription of Dutch in [Nederlandse Taalkunde], in which he advices the use of /h/. 3) I was misled by the universal use of /r/ instead of /ɹ/ in English phonemic transcripts, which is apparently inconsistent with the rules of phonemic transcription. If so, perhaps one of the earlier discussers will feel obliged to correct this;-) 4) The Wikipedia description of phonemic transcription states “One particular form of a broad transcription is a phonemic transcription, which disregards all allophonic difference, and, as the name implies, is not really a phonetic transcription at all (but at times coincides with it), but a representation of phonemic structure.” It uses the example /ˈlɪtəl/ for little as the phonemic version of the phonetic [ˈlɪɾɫ̩]. (The standard /ɫ/ at word endings could be another very common, non-diacritically encumbered symbol entirely skipped in English phonemic transcriptions.) This "broadening" I (apparently) have always misinterpreted to mean that simplification is part of the difference between phonetic and phonemic description.
There probably is a way to make this match with what you all wrote above, but it is a bit confusing to a non-linguist like myself. Cheers! Afasmit (talk) 02:25, 12 December 2017 (UTC)

@Afasmit: (I'm not a linguist though, just like you). They could be allophones in Dutch (they are in some varieties of English), but it's the frequency with which h is voiced is what's important. On the other hand, I've never heard of h-voicing in German. The fact that native speakers of Dutch don't perceive their h as different is irrelevant, given the multitude of issues Dutch people have with German phonetics (aspiration, assimilation, differentiating /ʏ/ from /œ/, correct pronunciation of /l/, consistency in vocalizing /r/, etc.) How many of them speak with a native-like accent? Exactly. The same applies to English-speaking Dutchmen.

Yes, but that doesn't render what he wrote before incorrect. It's just a change of opinion. Both /h/ and /ɦ/ are defensible choices as far as Dutch h is concerned, but /ɦ/ is more correct phonetically. Again, I see no good reason to change it to /h/.

Maybe it is, I'm not sure.

Actually [ɫ] is an allophone of /l/, not a phoneme. Dutch is similar in that regard, though in it, [ɫ] has a different distribution than in English. In neither case can that symbol be used in phonemic transcription, because it's not a phoneme. Mr KEBAB (talk) 13:15, 13 December 2017 (UTC)

I am a linguist and my opinion is that /ɦ/ is the best transcription for phonology and phonetics. The claim that Dutch people cannot hear the difference between [ɦ] and [h] is incorrect. When Dutch people use [ɦ] in English it's a tell-tale sign of their native language, and I immediately recognise it as a Dutch native speaker. When foreigners try to learn Dutch, they also typically struggle with acquiring the /ɦ/ because it's realisation is simply not [h].

Although I have read that some English speakers pronounce /h/ as [ɦ] in words such as behind, this constitutes a marginal difference and actually still does not sound like the Dutch [ɦ], which is actually phonetically more breathy than simply modally voiced.

In the current article, it states that Dutch /ɦ/ is sometimes pronounced as [h]. Someone should find a source for it, because I wouldn't know where they do this. The h-dropping in certain dialects is something distinct from pronouncing it as [h]. Barefoot Banana (talk) 11:52, 30 August 2023 (UTC)

Phonemic diphthongs and tense close-mid monophthongs

I think we should make a small exception to the rule that we should only use the most common phonemic symbols. IMO, the diphthongs /ɛi, œy, ɔu/ should be written with ⟨ɛɪ, œʏ, ɔʊ⟩ for the following reasons:

• In Northern SD, there's no appreciable phonetic difference between the ending points of /ɛi, œy, ɔu/, the ending points of [ei, øy, ou] (which, should we change ⟨ɛi, œy, ɔu⟩ to ⟨ɛɪ, œʏ, ɔʊ⟩, should be changed back to [eɪ, øʏ, oʊ], which was the original transcription in the article) and the starting points of [ɪə, ʏə, ʊə], with all of them being more or less [ɪ, ʏ, ʊ]. The vowel charts from Collins & Mees (2003) that show the pre-/r/ allophones of /eː, øː, oː/ seem to be a bit inaccurate and so I'd trust the written descriptions more than the images, at least in the case of [ɪə, ʏə, ʊə]. The similarity of the ending points of /ɛi, œy, ɔu/ and [ei, øy, ou] is confirmed by sources other than Collins & Mees (2003) so that's no problem.
• Because there's little difference between them on a phonetic level, transcribing /ɛi, œy, ɔu/ with ⟨ɛɪ, œʏ, ɔʊ⟩ and [ei, øy, ou] with ⟨eɪ, øʏ, oʊ⟩ will save some space (look at their phonetic descriptions now). This isn't very important but given the fact that we already transcribe [ɪə, ʏə, ʊə] as we do it seems to be a reasonable thing to do.
• ⟨ɪ, ʏ, ʊ⟩ look a lot like ⟨i, y, u⟩ anyway so it won't be a very drastic change.
• This is totally subjective but [ei, øy, ou] look a bit ugly to me and I'd like to change their transcription back to [eɪ, øʏ, oʊ]. That change would be more logical if we also changed the transcription of /ɛi, œy, ɔu/ to /ɛɪ, œʏ, ɔʊ/. Also, even though [ei, ou] are IPA symbols, they perfectly match the digraphs ⟨ei⟩ and ⟨ou⟩ which denote the phonemic diphthongs /ɛi, ɔu/ in Dutch. This isn't very problematic but something to be aware of.

As you can see, using a slightly more complex transcription of ⟨ɛɪ, œʏ, ɔʊ⟩ would actually increase readability in some aspects. And, should we ever introduce that transcription to Help:IPA/Dutch, it'd also tell our readers that they can safely end their /ɛi/ and /ɔu/ where English /aɪ, aʊ/ or /eɪ, oʊ/ end without worrying that the resulting sounds will be too lax, as would be the case in Russian or Serbo-Croatian.

I haven't been able to find a source that transcribes /ɔu/ with ⟨ɔʊ⟩ (nor ⟨ɑʊ⟩, ⟨ʌʊ⟩), but /ɛi/ is sometimes transcribed with ⟨ɛɪ⟩ in Verhoeven (2005) and De Schutter (2013:446) transcribes /œy/ with ⟨œ^ʏ⟩ (⟨œ⟩ followed by a superscript ⟨ʏ⟩). So switching to ⟨ɛɪ, œʏ, ɔʊ⟩ could be considered a slight violation of WP:OR, but in a good faith. It's definitely not a common set of symbols, that's for sure. Kbb2 (ex. Mr KEBAB) (talk) 18:24, 19 August 2018 (UTC)

I think we should stick with the most common symbols, at least for phonemic representations. Phonemic representations are meant to show the contrasts, and don't have to be 100% accurate. Phonemes are abstract representations of contrastive units, and they could use any symbol in theory as long as the same contrastive unit is always represented with the same symbol. Usually, symbols are used that are a fairly close match to the most common phonetic realisation, but this isn't an absolute rule, and in fact I'm not aware of any rules that say how phonemes must be represented. Sometimes, people take short-cuts. For example, a bunch of linguistic sources are still in the habit of transcribing /r/ for English for example, even though it's no longer a trill for most people. And for many languages you'll see /a/ when the phonetic realisation is [ɑ], if the backness is not a distinctive feature of the phonology. Rua (mew) 20:15, 19 August 2018 (UTC)

@Rua: Fair enough, I have a better idea than this. Kbb2 (ex. Mr KEBAB) (talk) 05:42, 20 August 2018 (UTC)

I suppose this is the best we can come up with at the moment, though it'd be good to change the phonemic transcription as well should reputable sources start using it (which, I guess, won't be soon). Kbb2 (ex. Mr KEBAB) (talk) 12:23, 22 August 2018 (UTC)

Epenthetic consonants in hiatus

Some speakers allophonically insert consonants such as [j] in the gap between two adjacent vowels. This should probably be mentioned in the article, but I'm not sure in which situations this applies, which consonants are used, and whether there is dialectal variation. Rua (mew) 20:08, 19 August 2018 (UTC)

Pre-/r/ allophones in the Netherlands

I've changed [iə, yə, uə, ɪə, ʏə, ʊə] to [iː, yː, uː, ɪː, ʏː, ʊː] in most of the article. I've rarely, if ever, heard the centering diphthongs used in the intervocalic positions and would personally always pronounce dieren, uren, boeren, heren, keuren and oren as [ˈdiːɾə, ˈyːɾə, ˈbuːɾə, ˈɦɪːɾə, ˈkʏːɾə, ˈʊːɾə], in accordance with the mainstream NSD pronunciation (I'm not a native speaker though). When it comes to /i, y, u/, the most important thing seems to be that they're elongated before /r/, whereas when it comes to /e, ø, o/, the most important thing is that they're not closing diphthongs and that they're raised and slightly centralized. AFAICS only /a/ can't have the schwa offglide before /r/, all other vowels can and the prominence of the schwa element seems to be dependent on the type of the postvocalic /r/ you use. It seems to be the most prominent before [ʁ] or [ʀ], intermediate before [ɻ], weak or absent before [ɾ] and non-existent before [ə̯] and [ɪ̯]. As far as I can hear, [iə, yə, uə, ɪə, ʏə, ʊə] can appear intervocalically, but only before [ʁ] or [ʀ]. Kbb2 (ex. Mr KEBAB) (talk) 10:48, 22 August 2018 (UTC)

/sɛntriˈfyːzjə/

I noticed that this article transcribes centrifuge phomenically as /sɛntriˈfyːzjə/ in the Monophthongs section and phonetically as [sɛntriˈfyːʒə] in the Example words for vowels and diphthongs section. What surprises me is that I can't imagine Belgian speakers who usually realize /sj/ and /zj/ not as simple fricatives but as fricative+[j] clusters to pronounce this word with such a cluster. Is there literature on this? Do some linguists postulate the (rare "xeno"-) phonemes /ʃ/ and /ʒ/ as being different from phonemic /sj/ and /zj/ sequences? Love —LiliCharlie (talk) 15:11, 6 October 2018 (UTC)

P.S.: The current article mentions /ʃ, ʒ/ as "not native phonemes ... [d]epending on the speaker and the position in the word." However my impression is that there is a set of words that have /ʃ, ʒ/ for all speakers and irrespective of "position in the word" — whatever that means. Love —LiliCharlie (talk) 16:06, 6 October 2018 (UTC)

Are non-native /iː yː uː/ generally accepted?

There are three sources for the non-native vowels. The first one (Gussenhoven 1999) appears to be based directly on the second one (Booij 1999), which has /iː yː uː/. The third one (Collins & Mees 2003), however, does not have /iː yː uː/, even though it includes the other non-native vowels (cf. p. 127).

I think there are two reasonable ways of representing this difference in the sources:

1. Keep /iː yː uː/ in the table and add a note that Collins & Mees (2003) do not include them.
2. Drop /iː yː uː/ from the table and add a note that Gussenhoven (1999) includes them.

It boils down to the question whether /iː yː uː/ are generally accepted (→ prefer first way) or not (→ prefer second way). Opinions/other sources? Personally, I would rather prefer the second way since the explanation in Booij (1999, p. 15s.) appears to be rather theory-driven, though I do not have any expertise with Dutch. --mach 🙈🙉🙊 13:31, 1 February 2019 (UTC)

monophtongized diphthongs

In this article, I miss the monophtongizing of diphtongs /ɛi, œy/ to /ɛ:, œ:/ before L in closed syllables. I believe this is standard in the north (i.e. the Netherlands north of the rivers) due to the thin syllable-final L. So 'pijl' en 'keil' to me have the same vowel as 'crème' or the final vowel in 'elitair', but 'pijlen' and 'keilen' have diphthongs. 'Vuil' and 'uil' to me have the same vowel as 'oeuvre' or 'freule', but 'vuile' and 'uilen' have diphthongs. I believe most speakers in the Northern Netherlands do this. Has anybody got a source that confirms this? — Preceding unsigned comment added by Eti erik (talk • contribs) 02:11, 20 December 2019 (UTC)

@Eti erik: I realize that the vowel section is imperfect. Collins & Mees (2003) say that both elements of /ɛi/ (which is normally [ɛ̞e]) are centralized and the second element is lowered [ɛ̈ë̞], so that 'pijl' and 'keil' do approach the non-native /ɛː/: [pɛ̈ë̞ɫ, kɛ̈ë̞ɫ]. I believe that this is more similar to the final vowel in 'elitair' when it's pronounced with a uvular /r/: [ɛliˈtɛəʁ]. This is another allophone we don't mention: in a closed syllable, /ɛː/ is a centering diphthong before /r/, especially when it's uvular.

If your observations are accurate, the reason 'pijl' is considered to have /ɛi/ rather than /ɛː/ is because it's a native word and because if we were to consider the vowel in that word to be /ɛː/, then 'pijlen' would have /ɛi/ anyway: [ˈpɛelə] (narrow transcription). There's no such alternation in Dutch phonology (AFAIK). Furthermore, the phonological diphthongs /ɛi, œy, ɔu/ can't appear before /r/ in the same syllable, but the non-native /ɛː/ and /ɔː/ can - which means that they're monophthongs like /eː, øː, oː/ (neither /œy/ nor /œː/ appear before /r/).

However, with that being said, if we get our hands on a WP:RS that says that speakers from Northern Netherlands feel the vowels in 'pijl' and 'vuil' to belong to the /ɛː, œː/ series, we should definitely put that in the article. There's no such thing as one, "correct" phonemic analysis of a language, at least not in the case of Dutch. The intuitions of native speakers are also important. However (another one :P), Booij's analysis of Dutch takes both Northern and Southern Standard Dutch (which in many aspects is just the older Northern standard, save for some phonetic details such as soft G) into account. In Belgium, /ɛi, œy, ɔu/ are often monophthongal and therefore also indistiguishable from /ɛː, œː, ɔː/. But in the Netherlands they're not, especially in the North (the local standard of The Hague is also moving away from that realization, AFAIK). In the Netherlands, members of the former set can appear in stressed syllables and they can't appear before /r/, whereas /œː/ is a marginal vowel that occurs only in a few words. On the other hand, /eː, øː, oː/ are monophthongal in Belgium and diphthongal in the Netherlands - but their second element is a schwa before /r/, not a close vowel, and Standard Dutch doesn't possess phonemic centering diphthongs in its inventory (unlike regional dialects/languages like Limburgish), only closing ones. Also, diphthongizing /eː/ and /oː/ in the sequences /eːw/ and /oːj/ is non-standard.

When you take all of that into account, /ɛi, œy, ɔu/ are phonological diphthongs, /eː, øː, oː/ are phonological close-mid monophthongs (further proven by the fact that /eː/ and /oː/ are shortened to /ɪ/ and /ɔ/ in some unstressed syllables - the phonological diphthongs undergo no such process) and /ɛː, œː, ɔː/ are phonological open-mid monophthongs that are marginal as they can only occur in stressed syllables of loanwords. Kbb2 (ex. Mr KEBAB) (talk) 07:58, 20 December 2019 (UTC)