Talk:Danish phonology/Archive 1

Archive 1

Archive 2

Archive 3

Phonology?

Why does this page not have any information at all about danish phonology? This is only an attempt to equate the graphemes of danish with a (sometimes incorrect)pronunciation. It's below the standard of an average tourist phrasebook. Magnuspharao 05:55, 18 August 2005 (UTC)

Plosives

(Copied from User talk:Enkyklios)

In my own langauge, Danish, the old opposition between voiceless and voiced stops has been given up in favour of an opposition between aspirates [b̥ʰ, d̥ʰ, g̊ʰ] and non-unaspirates [b̥, d̥, g̊] (voice being irrelevant and absent in the normal pronunciation). The stops in question do not have the muscular tension characteristic of the fortis, but at the same time they lack the voice normally associated with the lenis. In Ancient Greek, however, where we have a system of three articulations, voiceless, aspirated and voiced stops, the notation [b̥ʰ, d̥ʰ, g̊ʰ] is probably misleading since it would be natural to articulate the three rows with the greatest possible distance. Enkyklios 07:47, 13 January 2006 (UTC)

I'd like to raise with you the question on whether it is appropriate to use [b̥ʰ, d̥ʰ, g̊ʰ] and [b̥, d̥, g̊] rather than [pʰ, tʰ, kʰ] and [p, t, k] in the article on Danish phonology. As you say, in normal pronounciation these sounds are all voiceless and the distinction between the rows is the presence or absence of aspiration. It strikes me that a Martian linguist would therefore use the unvoiced symbol as a base. This situation is actually similar in Icelandic, where for some reason it has been quite fashonable to use [b̥, d̥, g̊] for the sounds which are usually represented word-initially with b, d and g but are in fact pronounced voiceless. Currently, e.g. in the new authorative book on the subject of Icelandic language, the IPA symbols [p, t, k] are used in their stead. Stefán Ingi 13:14, 13 January 2006 (UTC)

I am not unsympathetic with your point of view. Yet, I am no phonetician, and I feel that the phoneticians must have had an idea when they started to use these letters in their analysis of Icelandic and Danish in the first place. I suppose that a French linguist (my version of your Martian linguist) would hear a slight difference between our [b̥, d̥, g̊] and his own [p, t, k], and that difference has nothing to do with voice but with the muscular tense. Similarly, if I can trust my own ears, French /p,t,k/ are not 100% identical to Danish /b,d,g/. At any rate, even if one acknowledges that the phoneticians have not made up the whole voiceless lenis-phenomenon, it is perhaps reasonable not to notate this fine nuance in a regular IPA translation of Icelandic and Danish. After all, the possible nuances which one could express with special diacritica are legio, and the average reader will soon lose himself in the arcane jungle of phonetic transcription. Enkyklios 14:14, 13 January 2006 (UTC)

I agree with this summary. The question is now how to proceed. Since I am also not a phonetician perhaps it would be best to get more input. I'll copy this to Talk:Danish phonology and try and get some people interested. Thanks, Stefán Ingi 14:46, 13 January 2006 (UTC)

In southeastern German (my mother tongue) b, d, g are voiceless and p, t, k are not aspirated. The fortis-lenis distinction – higher air pressure and often longer duration for the fortes – is alone phonemic. I wouldn't know how to transcribe them other than [b̥, d̥, g̊] vs [p, t, k].

Voiceless lenes are actually quite common worldwide (southwestern German, Mandarin, Navajo...), even though they tend not to occur in the same languages as plain fortes.

David Marjanović | david.marjanovic_at_gmx.at | 13:04 CET-summertime | 2006/3/31

A minor detail

"[a], the regular allophone of /ɛ/ after /r/ is [ɑ] before labials and alveolars in the language of most younger speakers; before labials, it is often realised as a dipthong [ɑɪ̯]; the difference between strække "stretch" and strejke "strike", the only minimal pair, is practically non-existent."

/k/ isn't a labial, it's a velar. Shouldn't it be "before velars"? Mistyped, perhaps? --85.187.44.131 18:07, 18 April 2006 (UTC)

Text sample

Why have you chosen a trext sample in antique orthography? The example should reflect how danish is currently written not how it was written before the spelling reform of 1958.Maunus 06:54, 5 September 2006 (UTC)

"Current Developments"

Let me preface this saying I know nothing about Danish, but the following struck me as odd: In the "Current Developments" section it lists a possible sounds change in progress of a merger between /ø/ and /ö/. The second of these is not mentioned as an already existing phoneme in the table of vowels above, but just below this it says "One often hears [ø] for /ö/" implying that it is an active phoneme. I can only assume that the vowels refer to the same thing, and that /ö/ describes a fronted mid-high rounded vowel, aka /ø/. Is this a case of orthography (again, I know nothing about Danish orthography) being presented as phonology? --Coyne025 13:54, 13 April 2007 (UTC)

I do not know phonologist's jargon or IPA, so I can not give a technical explanation. And I of course do not know what the writer was thinking of. But I do know a bit about the history of Danish spelling that may be relevant.

There was a major spelling reform in Denmark in 1948. In the 100 years or so leading up to that, various people and groups made proposals, and some of these spelling systems were extensively used. One of the most popular alternative spellings was quite close to what was eventually made official. One part of this system that did not make it into the offical spelling, as thus is no longer used, was the addition of the letter Ö.

The letter Ø has several sounds in Danish. The sound that is the name of the letter, the Ø-sound, is the vowel of the the Frensh words for two and blue. I don't think it is used in English at all. For example, the verb dø, die, has this sound, as does its present form, dør, dies.

Another sound, let us call it the Ö-sound, is the vowel of Enlish shirt or hurt. This is also written ø, for example in the noun dør, door.

So does "den næste dør" mean the next one dies, or does it mean the next door? There is no doubt when the phrase is spoken, but when written you need context. Spelling dies "dør" and door "dör" would have solved this, but this is no longer done, and was never the standard the spelling.

My guess is that this "Ö-sound" is what the writer meant. I have no idea what it is called in technical jargon or what its IPA-symbol might be.

However, Ö has also been used for other things. Storm P always used Ö, never Ø, in the hand drawn text on his comic strips. Until recently it was common practice to use Ö on maps. It was tought that an O wich happen to span a line on the map might be confused with an Ø and vice versa.

--Klausok 07:06, 14 April 2007 (UTC)

This has nothing to do with spelling. The phoneme being written with the letter <ø> is in fact IPA [œ] (an open mid front rounded vowel) so what it is saying is that /œ/ and /ø/ (close mid front rounded vowel) is merging for some speakers. The /œ/ phoneme is a marginal phoneme only contrasting before certain consonants (n, m, j). I don't know why they have chosen to write IPA [œ] as <ö> in the vowel tabel instead of using the IPA transcription - but if you look in the long table on the right you will see that the pronunciation of <ö> is described as [œ]. Another thing is that I am not sure the statement is true - i certainly cannot recall having heard any of those words with [ø] instead of [œ]·Maunus· ·ƛ· 12:42, 16 April 2007 (UTC)

Do you mean that The phoneme being written with the letter <ö> is in fact IPA [œ]? BTW I have not heard them merge either.

--Klausok 16:51, 16 April 2007 (UTC)

Yes, that is what the tables say. I still don't know why they use that letter though.·Maunus· ·ƛ· 21:00, 16 April 2007 (UTC)

Symbol

I can't find any explanation for the symbol n̩. Can anyone explain to me the pronunciation of it?

It represents a syllabic n, like the "on" of button.Cameron Nedland 22:24, 7 September 2007 (UTC)

Glottal stop?

The article (as well as the various transcriptions around Wikipedia) use a superscript <ʔ> (<ˀ>) to denote what looks like a form of elision or glottal replacement. But I'm not really sure. Is that supposed to be a full glottal stop and if so what's wrong with using <ʔ>? Either way, the article, I think, should explain this a little bit. Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 01:43, 8 September 2007 (UTC)

it's the stød-·Maunus· ·ƛ· 09:16, 8 September 2007 (UTC)

Ahh, my bad. Thank you. Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 18:33, 8 September 2007 (UTC)

Sound sample

It would be very nice if a good danish speaker would create a sound file with the read text sample :)

I will try to make one I am jusst not sure how to make .ogg files.

---Yeah, that's a very good idea, I'd do it myself but has no idea how to speak danish. Tobias_shlfr (talk) 07:58, 7 August 2008 (UTC)

Orthography

It took me ages to decode what the algebra for letter juxtapositions meant, so I wrote it in human-readable language. Honestly, who thinks that "/ _ r; / r _" is simpler than "next to r"? Feel free to change if I've misunderstood the symbols – I would be surprised if I've comprehended them all correctly! 212.137.63.86 (talk) 12:25, 2 September 2008 (UTC)

I've just noticed a was missing. I think I've got it from the vowel-phoneme table (I don't speak Danish by the way...) 212.137.63.86 (talk) 12:35, 2 September 2008 (UTC)

Ø

Perhaps I’m misunderstanding or misreading the “Phoneme / Pronunciation” vowel chart, but some aspects of the various realisations of ø (i.e., /ø/ and /œ/ and their long counterparts) seem rather confusing and not always correct to me.

I. First off, I cannot—off the top of my head—think of any examples where /øː/ becomes [œː] before /r/. All the examples I can think of have [øː], such as møre, køre, etc. have [øː]. The change to [œː] seems to me to happen only after /r/, not before it. I presume this is a case of a misplaced colspan declaration: it should have gone to the previous cell in the table.

II. Secondly, writing [ø] > [œ] in the leftmost column of the phoneme /ø/ to me indicates that where no /r/ is present, [ø] becomes [œ], which seems entirely inaccurate. Should we not rather write [ø] / [œ] and add a note that [ø] is lowered to [œ] before /m/ (e.g. sølle [ˈsølə] vs. sømme [ˈsœmə]), except where written <y> (e.g., lymfe [ˈlømfə]); and similarly that, after /r/, [œ] is further lowered to [Œ] before /m/ (e.g. sømme [ˈsœmə] vs. rømme [ˈʁŒmə]), except where written <y> (e.g. krympe [ˈg̊ʰʁ̥œmbə])?

III. Thirdly, there seems to be a rogue Note 3 in the chart that does not have a corresponding reference. I’m not sure it even belongs to the <ø>’s, but it seems a likely candidate. Can’t figure out what it’s supposed to be for, exactly …

IV. Lastly, the presentation/description of the phonemes /œ/ and /œː/ seem rather muddled and incomprehensible to me. Note 4 (for /œː/) says the phoneme appears “(1) [o]nly before /r/ and (2) perhaps /n, v/ (only høne, bøvet)”; and Note 5 (for /œ/) says that this appears “only before /n/”.

Now, as regard Note 4(2), I consider it somewhat misleading to lump /n/ and /v/ together into this phoneme, since bøvet (and similar words, such as møve) are clearly the result of a subconditioning of /øː/, which is realised as [œː] between a labial and /v/. As [øː] and [œː] are in complementary distribution in these surroundings and there are no minimal pairs, I don’t see the necessity of ‘inventing’ a different phoneme to explain the difference in pronunciation. In Note 4(1), since høne is the sole example of [œː] which is not explained by either this previous subconditioning or the general conditioning that /øː/ is lowered to [œː] after /r/, it could be analysed either as an analogous exception to regular /øː/ (analogous since all other words that contain /øːn/ in fact happen to contain /røːn/) or as containing the phoneme /œː/. In this case, it seems to make more sense to include the example under the phoneme /œː/, though there is—rather unfortunately—no minimal pair to lean upon.

In the chart, /œː/ before /r/ is written to be realised [œː]. Of this, however, I cannot find a single example; on the other hand, I can find a few examples of [Œː] before /r/: gøre [ˈg̊Œːɐ], røre [ˈʁŒːɐ]), smøre [ˈsmŒːɐ]. This appears to be a simple mistyping that should be corrected in the chart, and has no bearing on the status of the phoneme, as far as I can tell. Therefore, I posit that /œː/ does not occur “[o]nly before /r/ and perhaps /n, v/”, but rather that /œː/ occurs only before /r, n/.

Similarly, the chart states that /œː/ does not occur after /r/, only before it; I should think this is incorrect. /œː/ after /r/ is, in fact, one of the few places where we have an actual minimal pair: røve (‘steal’) [ˈʁœːʊ] vs. røve (‘arses’) [ˈʁŒːʊ]. In this case, the former is clearly a case of a regular development of /øː/ to [œː] after /r/, whereas the latter must be analysed as being /œː/, though once again presenting us with [Œː], not [œː].

The short counterpart of /œː/ is stated in Note 5 to appear only before /n/. However, the imperative and singular forms, respectively, of the previous minimal pair also constitute a minimal pair: røv (‘steal!’) [ʁœʊˀ] vs. røv (‘arse’) [ˈrŒʊˀ] (if we accept this, despite the phoneme being now part of a diphthong), which would mean that /œ/ also occurs before /v/.

For these reasons, I would suggest changing the relevant parts of the chart to the following:

Phoneme	Pronunciation
	-	Before /r/	After /r/
/øː/	[øː]		[œː]
/ø/	[ø] / [œ]3	[ɶ]	[œ] / [ɶ]3
/œː/4	[œː]	[ɶː]
/œ/5	[œ]	-	[ɶ]
Before labials and alveolars Before labials and velars Before /m/, except where /ø/ = <y> Only occurs before /v, r, n/ (the latter of which has just one example: høne) Only occurs before /v, n/

(Finally, apologies for the length of this, and for the rather ugly hack of using [Œ] to represent the open front rounded vowel. It is, unfortunately, necessary, because of the bug in the standard OS X version of Helvetica, which renders [œ] and [ɶ] the same.) 89.236.56.150 (talk) 16:05, 22 October 2009 (UTC)

Stræde and gade

I have taken out this bit:

E.g. the vowels of stræde "alley" and gade "gade", which are pronounced identical in Modern Standard Danish, are transliterated æ: and ɑ: respectively since they are allophones of different phonemes.

It is just plain wrong. The vowels of stræde and gade are quite different. They are the sounds used as names for the letters æ and a. To my untrained ears they sounds exactly like the vowels in English get and hat respectively (except for length). the previous unsigned comment was written by User:Klausok

I am a native speaker and I pronounce them exactly the same. And so does everybody I know. I am afraid your untrained ears are wrong. Only elderly speakers and speaker of nonstandard varieties distinguish the pronunciation of those two. Maunus 11:52, 8 January 2007 (UTC)

I am a native speaker, 47 years of age, grew up in North Zealand and now live and work in greater Copenhagen. I know nobody who pronounce them the same. Are you saying that glade and glæde sound the same to you? Or vade and væde?--Klausok 14:27, 8 January 2007 (UTC)

No, glade and glæde and vade and vade do not sound the same. The vowel in stræde however is influenced by the /r/ in front of it causing a lowering of the vowel to coincide with trhe pronunciation of the vowel in gade, vad, lade and glade. For speakers of modern standard danish stræde rhymes with those words and not with glæde, vædde, æde and smæde. This is exactly what the passage you removed says: even though most danes pronounce them the same they are in fact allophones of each a different phoneme, but the r-colouring of the æ:in stræde causes it to be pronounced the same as the fronted ɑ: in gade. Also I don't know which sociolect/dialect you speak but some kinds of north zealandic do still distinguish the two. so it may be a question of your pronunciation varying from standard danish (the queen also distinguishes stræde and gade)Maunus 14:33, 8 January 2007 (UTC)

Being a native speaker, I can confirm that the vowels of stræde and gade are virtually identical in my own speech and in the speech of most of my contemporaries (I'm 35): [ˈg̊æːð̩] and [ˈsd̥ʁæːð̩]. I am not able to detect any significant difference in my tongue position when pronouncing the central vowels of the two words. Similarly, drille and falde rhyme in contemporary Danish: [ˈd̥ʁæll̩] and [ˈfæll̩].

It is, of course, not the case with glade and glæde. The first one has the same vowel as stræde and gade, i.e. a vowel closer to English bad than to English bed. On the other hand, glæde has a vowel, which is even more closed than the vowel of English bed. Especially in the modern variety of Copenhagen, [ɛ(ː)] tends to merge with [ɛ(ː)]. Enkyklios 11:39, 21 January 2007 (UTC)

I am very surprised at this. I would say that "drille" rhymes with "fælde", not "falde"--Klausok 07:58, 22 January 2007 (UTC).

Somewhat late to the party, this very much surprises me. Drille and falde do not quite rhyme for me, either (I would say the [æ] in falde is a bit lower than the [æ] in drille, though they’re close); but drille and fælde most certainly do NOT rhyme, at all. Fælde has [ɛ] as its vowel, which is far higher up than either [æ] in drille or falde. 89.236.56.150 (talk) 16:15, 22 October 2009 (UTC)

Then you are most certainly speaking a socio- and or dialect that is different from what is called "modern standard danish" - retaining the original vowel qualities in examples like these is common in a few speech varieties, for example many by older speakers or speakers form conservative language groups. To me rhyming those words sounds like the dialect of Karen Blixen or the Queen. If you want I can provide examples of the forms with merged vowels stated as being the dominant form if you want to check yourself you can look in Ida Grønnums book "Fonetik og Fonologi" theres a chapter on vowels in modern standard danish, and she also mentions the variation among differen age groups. ·Maunus· tlahtōlli 11:15, 22 January 2007 (UTC)

[ð] vs. [ɹ̪]

Why is the eth ‹ð› being used to reprisent the alveolar approximant instead of the upside-down r with the dental mark ‹ɹ̪›? Wisapi (talk) 23:50, 19 January 2010 (UTC)

Probably because that's how the literature represents it. It's also simpler to type and may be acoustically similar to a dental approximant. Plus, there's a voiceless one, which is even harder to represent using the alveolar approximant character. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 07:50, 20 January 2010 (UTC)

This voiceless variant must be then included to the article because it's not there. Just out of curiosity: if we were to write that phoneme as it is, which one would it be: [ɹ̪̊ ɹ̪̥]? Wisapi (talk) 19:26, 22 January 2010 (UTC)

There are multiple ways of representing it. As the article states, there is sometimes frication, though that may happen with approximants in emphatic pronunciation. [θ̠ [ð̠] could represent alveolar non-sibilant fricatives (as they do in Icelandic). Of the characters you list, the latter isn't a good way to represent anything because the diacritics get cluttered. The first one is a voiceless dental approximant, which is more easily rendered as [θ̞]. Taking out the dental diacritic would be an accurate choice for a voiceless alveolar approximant. I think that's how it's done in English. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 19:59, 22 January 2010 (UTC)

Dubious edits

I've considered reverting the following unsourced edits: [1] They contradict Danish language#Phonology, for example by denying the existence of the pharyngeal realisation of <r> in Danish, see Talk:Danish language/Archive 1#Pharyngeal??. I suspect that the user changed the description to conform to his own (possibly non-standard or conservative) dialect. However, unfortunately, neither is that section sourced, nor was the original version of the page sourced, either. Any comments? --Florian Blaschke (talk) 14:12, 22 April 2011 (UTC)

I didn't pay close attention to the edits, since I don't know a whole lot about Danish phonology. I do know that Sounds of the World's Languages says that "Similarly, the Danish 'r' sound in words such as raad ('council') is not a uvular approximant as some textbooks describe it, but a pharyngeal approximant with an articulatory position similar to that in a low back vowel." p. 323. This is why I put Danish in the table over at Approximant. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 17:23, 22 April 2011 (UTC)

Since these are abstract representations I feel the need for at separate column for a single phoneme is exaggerated as long as it's based purely on articulatory arguments. /r/ have different allophones, some maybe more uvu-pharyngeal than others depending on the speaker, but claiming it to be a pharyngeal phoneme is not conforming with Danish textbooks.

If so, what about many of the other phonemes? /p t k b d g/ are not alwayes plosives, /n/ may be labial and velar too depending on surroundings, /e ø o/ are not always close or even mid, /ɛ œ ɔ/ are not always mid, /a/ may be both front and back, and /ə/ can be just about anything, even contoid, as long as it syllabic.--Schwa dk (talk) 10:37, 23 April 2011 (UTC)

Ladefoged could have specifically been talking about a specific allophone. Like I said, I don't really know enough about Danish phonology. I suppose at some point I might know more and edit the article to provide it with actual citations for its claims. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 14:59, 23 April 2011 (UTC)

Ladefoged definitely talks about a specific allophone, namely [ʁ]. In the example given, råd, that is in initial position before an open back vowel, it is very likely more pharyngeal than in say a word like grise [gʁiːsə] ('pigs'). In coda it may very be as high and fronted as [ə̯] or a mere lengthening of the preceding vowel in a word like kirke [kie̯gɪ] ('church').

Anyway, my comment above was more directed at the original poster. Sorry if it came out wrong. I have no problem calling it pharyngeal or uvular. It just seems rather arbitrary to choose one label over another. I've made a lot of much needed edits lately, and I wouldn't like to see it reverted because one minor dispute.--Schwa dk (talk) 16:34, 23 April 2011 (UTC)

A bit on long vowels?

The section on 'short/long vowels' should be simplified and clarified. The use of an -e after a consonant and an open vowel is not very well-explained: is kage pronounced kayg, kaguh, kageh, kag, kayguh, or kaygeh? (sorry, I don't know IPA) I'm teaching myself Danish, and this has become rather a painful block. Could this section be clarified? 217.20.20.85 (talk) 19:30, 7 June 2011 (UTC)

Strictly speaking, this has nothing to do with vowel length, but the realization of the phoneme /ə/. The article certainly lacks information about this, and I would be happy to add it at some point.

With regards to kage it is pronounced [kʰæːæ ~ kʰæːɪ], see Help:IPA for Danish for closest English equivalents.--Schwa dk (talk) 14:47, 12 June 2011 (UTC)

Phoneme inventory

Does Danish really have no phonemic 'ng' as the consonant section suggests? What about all the words that end in -ing? — Preceding unsigned comment added by 72.89.239.8 (talk) 21:56, 31 May 2011 (UTC)

The phone [ŋ] is traditionally interpreted as /ng/ in the litterature, and this is what is reflected in the article. The phonemic string /ng/ will always be realized as [ŋ] by all danes, so you wouldn't gain anything by postulating a new phoneme /ŋ/. The distribution of [ŋ] in the language corresponds perfectly to a consonant cluster like /nd/, e.g. it only occurs in coda, there are no long vowel before [ŋ], in monosyllabic syllabic words it always has stød and so forth.--Schwa dk (talk) 13:43, 12 June 2011 (UTC)

Personally I will have to object to some simplifications in the phoneme inventory. Being a native Danish speaker, I know that [ð̞] is not merely an allophone of /d̥/ since the word rådden ['ʁʌð̞n̩] "rotten" is clearly different from the word rotten ['ʁʌd̥n̩] "the rat". Similar minimal pairs can be found for /ŋ/ and /ng/ (since synger ['søŋʼɐ] "sing(s)" is not the same as synker ['søŋʼg̊ɐ] "sinks". [ŋ] does not always occur before a [g̊] as the article claims). Finally, [a] and [ä] are different phonemes since kræmmer ['kʰʁämɐ] "merchant, shopkeeper" is different from the word krammer ['kʰʁamɐ] "hug(s)", at least in the standard dialect. I also doubt that [æ] and [ʌ] are not phonemic, but I can't think of any minimal pairs. 80.165.160.150 (talk) 16:18, 8 June 2011 (UTC)

Being a language user doesn't automatically grant you insight of phonological theory. According to the traditional analysis on which this article is based, rådden and rotten are /rɔd.ən/ and /rɔt.ən/ respectively. Likewise, synger and synker are /søng.ər/ and /sønk.ər/.

The article doesn't claim that [ŋ] only occurs before [g̊] but before /g/ or /k/. Beware of the square brackets versus the slashes. /g/ is a phoneme and [g̊] is an allophone.

Likewise, [ä] and [a], or rather [a̝] and [ɑ̈] as your examples seem to imply, are not different phonemes, but different allophones. They may both stem from different phonemes as in your example, or be allophones of the same phoneme in different contexts, e.g. in /kat, tak/ [kʰa̝d̥, tˢɑ̈g̊] "cat, thanks".--Schwa dk (talk) 14:38, 12 June 2011 (UTC)

I definitely did not mean to pretend to be some kind of expert on phonology just because I speak Danish. I just tried to point out that this article seemed to lack some distinctions that I have always thought were quite clear. Though I must admit that can't argue against a professional linguist, since I am just a linguistically interested amateur and probably missed something in my research. 80.165.160.150 (talk) 17:10, 12 June 2011 (UTC)

Clearly absurd table of allophones.

Please stop re-adding the table, it's clearly incorrect. I think it's extremely unlikely that such analysis can be justified. For example why should be [ɑ] an allophone of /ɛː/ after /r/ and before /d/; allophone of /ɛ/ after /r/ and before labials and alveolars, except /d/ and allophone of /a/ everywhere else? And what could possibly justify the claim that [œː] is an allophone of /øː/ after /r/ while /œː/ does not occur after /r/? Maybe something in the danish grammar suggests this, but then it would be almost certainly better described as vowel alternations or sandhi.--90.179.235.249 (talk) 17:44, 8 June 2011 (UTC)

The table is correct according to the bibliography mentioned in the article. If you think there are incorrect elements, please provide a source, not just your opinion. For an explanation, the uvular /r/ affects many vowels to become more open and retracted in certain positions. Please be aware that phonological traditions differ, and your disagreement with the article may stem from a different phonological approach altogether. The Danish tradition is to do a deep analysis with fewer phonemes and more rules. This is reflected in the article since it's based on litterature with this approach.--Schwa dk (talk) 10:26, 11 June 2011 (UTC)

You might want to use inline citations to clarify for readers where exactly the information is coming from. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 14:10, 11 June 2011 (UTC)

The information is in agreement with Basbøll (2005) p. 52 and Grønnum (1998a) 3rd ed. (2005) p. 287-288. I'm not sure where and how to put citations in the article. Help will be appreciated.--Schwa dk (talk) 11:42, 12 June 2011 (UTC)

All right, I've added inline citations using <ref></ref> tags, which create superscript links that go to the bottom. If you're worried about the complicated {{harvcoltxt}} template, it's no big deal; you can just put e.g. "Basbøll (2005) p. 52" and it will be clear to readers and other editors. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 12:53, 12 June 2011 (UTC)

Great, thanks!--Schwa dk (talk) 14:48, 12 June 2011 (UTC)

I'm not supposed to find sources that the information is incorrect, you are supposed to find sources it's correct. You have finally specified your sources, but I insist it canot be left like this. I will check if I can find these books, but are you sure it is really described as allophony? Because this really makes no sense. Even if it is described like this in the books, the reasons for such wierd analysis should be explained in the article. Especially these rows seem to be clearly wrong:

Some vowel allophones

Phoneme	Pronunciation
	Default	Before /r/	After /r/
/ɛː/	[ɛː]	[æː]	[æː] / [ɑ]1
/aː/	[æː]	[ɑː]
/øː/	[øː]		[œː]
/œː/	[œː] ~ [ɶː]	[œː]	NA

Why it is claimed that [æː] is an allophone of /ɛː/ when it's next to /r/ and an allophone of /aː/ in other positions? And why is [œː] analysed as an allophone of /øː/ after /r/ and said that /œː/ does not occur after /r/? Why not simply say that /øː/ does not occur after /r/? This should be explained in the article. And what are notations like [ɛ] ~ [æ] suposed to mean? And why there is [ɛ] in this table and not in the first one? Sourced or not, these issues have to be dealt with.--90.179.235.249 (talk) 10:27, 19 June 2011 (UTC)

While I don't challenge the confusing presentation, the allophony described is a little unusual but an analysis that postulates this type of allophony is not unusual or extreme. Either way, as you say, sourcing should be clearer. I think that Schwa dk will be getting to that soon. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 14:40, 19 June 2011 (UTC)

I'm rather new to wikipedia editing, but I believe you're supposed to ask for a citation for questionable content rather than just removing the parts that you disagree with. The table was there before I started editing, and anyone familiar with Danish phonology is also familiar with tables of vowel allophones like this. I guess whoever put it there in the first place didn't think of making an inline citation.

I'm not sure why you find the analysis so problematic. It seems to me that you might be accustomed to a different phonological approach than what's presented in this article, but it's not like that there is one correct way of doing phonological analysis. It would be nice if all wiki articles on language specific phonology had the same approach to phonology, but I don't believe it would be possible at the moment. If you're unhappy with the current analysis you can probably just stick with the allophone table and call them phonemes instead, but you'd be missing a lot on the distributional pattern of the sounds.

However, if you find it hard to believe that a language can have some twenty distinctive vowel qualities, the table shows that the vowels can be attributed to just nine different phonemes in a way that is probably 99.9% consistent within the language. This is still quite many vowel phonemes compared to most languages.

With regards to [æː], the table shows that [æː] and [ɑː] can be mapped onto /ɛː/ and /aː/ before /r/ since [ɛː] and [aː] doesn't occur before /r/. Otherwise, you would have to establish two more vowel phonemes, /æ/ and /ɑ/.

With regards to [œː], you could map it onto /œː/ instead of /øː/ without complications. I'm not sure of the authors' intentions, but it may be that /ø/ is preferred to /œ/ as the more common phoneme, or that's it's more consistent with speaker intuition or orthography or whatever. Either way, that's how it's done in the litterature.

With regards to the notation [ɛ] ~ [æ], it seems to signal some uncertainty or discrepancies in the litteratures or some speaker dependant variation. As far as I can tell in [x] ~ [y], x is the more conservative pronunciation and y is the more recent development. I didn't invent the tildes but I've seen them used in different contexts in phonetic transcriptions to signal e.g. a range between two phones. I'm not sure everyone is familiar with this notation though. I have no problem with it being there, but there might be better ways.

With regards to [ɛ] not being in the other table, it seems that someone edited the phone symbols in that table and not the other to a more narrow IPA transcription. I think they should both use the standard that is used in IPA for Danish throughout wikipedia. I'll do that.--Schwa dk (talk) 10:15, 20 June 2011 (UTC)

Per, WP:BURDEN, citation request tags are a nice tool to identify uncited material, but they are not a replacement for actual citations and can be removed when a reasonable amount of time has been given. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 11:48, 20 June 2011 (UTC)

Compressed vowels?

We say that Danish /u/ is compressed, so I transcribed it ⟨uᵝ⟩. Is this correct, and are any of the other back or mid vowels compressed? — kwami (talk) 22:34, 5 June 2012 (UTC)

Do you have a source for this? I have never seen it transcribed this way. It seems like an overspecification since /u/ may be pronounced in many different ways depending on the speaker. Also [ᵝ] is not an IPA chaacter, is it? --Schwa dk (talk) 05:43, 6 June 2012 (UTC)

Yes, β is IPA. (And any IPA letter may be used as a diacritic.) We list Danish /u/ at Close back compressed vowel. Is that wrong? — kwami (talk) 08:36, 6 June 2012 (UTC)

Oh I see. No, that doesn't seem right. I think maybe someone confused Danish with Swedish. The Swedish and Japanese /u/'s doesn't sound like anything you would normally encounter in Danish. You can listen to some sound samples of Danish /u/ on my website here: http://schwa.dk/filer/udtaleordbog_danpass/#bue (sorry for the lack of English translation - if you have any questions, don't hesitate to contact me). --Schwa dk (talk) 09:37, 6 June 2012 (UTC)

Okay, thanks. I'll remove the Danish example. — kwami (talk) 15:58, 6 June 2012 (UTC)

Jens eats a child

Perhaps we could find a better example of a Danish phrase...?Jeppiz (talk) 19:24, 15 October 2009 (UTC)

Done. Changed to "Jens eats a bread loaf". Pronunciation of "brød" copied from Wiktionary; I hope it is correct. --Jhertel (talk) 15:58, 21 March 2013 (UTC) (Thanks to Klausok for correcting bread to loaf. --Jhertel (talk) 03:09, 22 March 2013 (UTC)).

/a/->/æ/ transition

Well, it seems that a lot has been edited out since I last visited here. Especially one essential thing is now missing that was previously there. Now you can read: Before labials and velars, /a/ is [ɑ] in most varieties: in other positions, it is [a] in the conservative speakers. But look here:Danish language (Vowels section): For example, /ø/ is lowered when it occurs either before or after /r/, and /a/ is pronounced [æ] when it is long. YES! When long, /a/ is indeed pronounced /æ/. But why isn't this mentioned anywhere in the phonology article?? -andy 77.191.222.178 (talk) 06:03, 6 August 2013 (UTC)

"Learned" Suffixes

I can read something about learned suffixes. Does this happen to mean loaned suffixes, i. e. those borrowed from other languages? (cf. the term "loanword") It's funny that the mistake (if it's one) does not only appear once in the article. In my German tongue, it reminds me of people who ALWAYS write "Gesetzt" if they mean "Gesetz" (law). :) -andy 77.7.121.55 (talk) 20:47, 28 May 2013 (UTC)

Learned as in book learning. Suffixes imported from the learned languages Latin and Old Greek, as opposed to from languages people actually speak, like German or French. Klausok (talk) 07:18, 29 May 2013 (UTC)

I've never heard of the concept of "learned suffixes", also several of them are frmo french and not from latin, none of them are greek. Wha is meant is clearly borrowed suffixes from romance languages.User:Maunus ·ʍaunus·snunɐw· 23:31, 6 August 2013 (UTC)

/y/

Is Danish /y/ pronounced with protruded lips, like Swedish and Norwegian /y/, or with compressed lips, like German and French /y/? kwami (talk) 12:43, 2 December 2008 (UTC)

I could ask the same about /u/: protruded like German and English, or compressed like Norwegian and Swedish? kwami (talk) 06:19, 11 May 2009 (UTC)

/y/ is pronounced with compressed lips, just rounded and with advanced tongue root. /u/ is just compressed.·Maunus·ƛ· 09:59, 11 May 2009 (UTC)

Okay. At close front rounded vowel Danish y is now listed with German ü and Swedish u as compressed, while Swedish and Norwegian y are listed as protruded. At close back rounded vowel, Danish u and Swedish/Norwegian o are listed as compressed, vs. German u as protruded. kwami (talk) 17:10, 11 May 2009 (UTC)

I checked up on /u/ and u is exolabial i.e. protruded. y is still compressed though.

·Maunus·ƛ· 18:15, 11 May 2009 (UTC)

d commonly corresponds to other phonemes, notably /-/ and /ð/ - shouldn't this be listed in this section? — Preceding unsigned comment added by 2.105.37.95 (talk) 19:00, 21 January 2012 (UTC)

[ð] is an allophone of /d/, not a phoneme. Peter238 (talk) 17:46, 14 April 2015 (UTC)

Allophonic rounding of high vowels

I've heard that unrounded high vowels become rounded (maybe also central?) before [ð]. If anyone is aware of a source that talks about it, please put this information in the article. Peter238 (talk) 17:51, 2 April 2015 (UTC)

Not in any variety I know. /bide/ is [b̥iːðə], /byde/ is [b̥yːðə]. The i is quite close to the cardinal i in my pronunciation and most ones I know. ·maunus · snunɐɯ· 02:40, 10 May 2015 (UTC)

Ok, thanks. Peter238 (talk) 12:01, 3 July 2015 (UTC)

Still no Phonology...

Ten years ago I pointed out that there is no actual phonology here - just a lot of very detailed phonetics. That is still the case.·maunus · snunɐɯ· 02:42, 10 May 2015 (UTC)

I wouldn't say that there's no phonology here, but yes, most of this article covers Danish phonetics. We do cover vowel phonemes, stress and stød though. I'd say that's a good start. Peter238 (talk) 12:10, 3 July 2015 (UTC)

Faroese and Greenlandic accents?

Does anyone know anything about these? Peter238 (talk) 01:19, 4 July 2015 (UTC)

Not much work on that. These are the two most useful texts I could find on google scholar. Unfortunately none of them are available. Kühl, K. (2013). Færødansk og dansk på Færøerne. Maal & Maele, 35(4), 15-22. (here is a preprint[2]). Jacobsen, B. (2000). Sprog i kontakt. Er der opstået en ny dansk dialekt i Grønland? En pilotundersøgelse. Grønlandsk kultur-og samfundsforskning, 98(99), 37-50.·maunus · snunɐɯ· 06:35, 4 July 2015 (UTC)

This one may work: [3]

Nice, thanks! I myself found [4]. Peter238 (talk) 01:57, 5 July 2015 (UTC)

Loss of the voiced velar fricative

On the talk page for that sound, I discuss the environments where some Danish speakers still use it. Shouldn't the orthography section also mention that g may be pronounced [ɣ] in those contexts (as long as we keep the "older speakers" side-note)? I'd add that myself, but the notation in the orthography table is a bit confusing! --Ingeborg S. Nordén 15:16, 2 August 2007 (UTC)

I added that about two months ago. Peter238 (talk) 18:41, 5 July 2015 (UTC)

Visual depiction of relations between letters and sounds

As a learner of Danish, I found the image on this page really helpful in approaching IPA notation for Dansh as it breaks down the enormous amount of information on the different sounds and is a much easier reference when relating what I read to what I hear: http://www.omniglot.com/writing/danish.htm. In fact a reverse diagram of the same information would be useful too, listing the sounds and the letters that can make them. I think this diagram could be incorporated as is - see http://www.omniglot.com/faqs.htm#use - but I guess it may need to be expanded to match the article. 82.71.38.251 (talk) 17:34, 29 July 2015 (UTC)

/tˢ/?

I'm confused, shouldn't /tˢ/ represent the Voiceless alveolar affricate, not the Voiceless alveolar stop?Cameron Nedland 21:41, 27 October 2007 (UTC)

I think there's some sort of acoustical relationship between an aspirated dental/alveolar plosive and a slightly affricated one. So if I were making a phonemic representation of Danish I would gloss over this affrication as it parallels the aspiration of the labial and velar stops. Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 21:52, 27 October 2007 (UTC)

Okay, just a little confused.Cameron Nedland 20:35, 28 October 2007 (UTC)

Good observation but it doesn't give any direct clue if /tˢ/ is Voiceless alveolar affricate or a Voiceless alveolar stop. I ask you to be more direct please. —Preceding unsigned comment added by 201.80.11.190 (talk) 19:37, 31 March 2008 (UTC)

tˢ is not an affricate. Affricates are written t͡s with an arch. tˢ is a voiceless alveolar stop with a slight friction. It should only be transcribed as tˢ in a close transcription but as a plain "t" in a broad transcription. ·Maunus· ·ƛ· 19:58, 31 March 2008 (UTC)

Well that makes three people who have been confused about this. We should edit the article to clarify this. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 21:12, 31 March 2008 (UTC)

Actually, Grønnum (2005) describes [tˢ] as a strongly affricated voiceless apical alveolar stop, so it is an affricate. Peter238 (talk) 12:11, 3 July 2015 (UTC)

ARe you sure an affricated stop is an affricate? I am pretty sure both IPA and Grønnum distinguishes between a stop with some friction and an actual affricate.·maunus · snunɐɯ· 21:23, 3 July 2015 (UTC)

An affricate is nothing more than a stop with a strong fricative release. It's up to the individual to define the boundary where the stop has a strongly enough fricative release to merit it being described as an affricate (even then, as you see, Grønnum (2005) chose to call it a strongly affricated stop, rather than an affricate). The best source for that that I could dig up is [5]. The relevant part is on the first half of page 37. Peter238 (talk) 21:36, 3 July 2015 (UTC)

Another citation: "Affricates are stops in which the release of the constriction is modified in such a way as to produce a more prolonged period of frication after the release. As with many of the types of sounds we have discussed, the class of affricates has no sharp boundaries. Affricates are an intermediate category between simple stops and a sequence of a stop and a fricative. It is not always easy to say how much frication should be regarded as an automatic property of a release; some places of articulation seem to be often accompanied by a considerable frication (see chapter 2)." (SOWL, page 90). Peter238 (talk) 23:36, 3 July 2015 (UTC)

I think it makes more sense to say that it is a stop that tends to be affricated, both because the degree of frication depends on the following vowel, and because some traditional dialects don't affricate it but aspirate it instead. ·maunus · snunɐɯ· 06:54, 4 July 2015 (UTC)

The latter part of the sentence is backed up by Grønnum (2005), but have you got a source on the degree of frication? Grønnum (2005:120) just says that it is "kraftigt affrikeret". On the other hand, Thorborg (2003, don't remember the exact page) says that it is weakly affricated, but I'm much more inclined to believe Grønnum in this case (especially since Thorborg describes the coronal stops as laminal denti-alveolar, something I haven't seen done by other scholars. Maybe that book is describing Queen's conservative sociolect or something? But then... why would you teach that to foreigners.). Peter238 (talk) 09:34, 4 July 2015 (UTC)

Grønnum specifically describes low Copenhagenian I believe, and that is generally described as the one with most frication. I will see if I can find a source for it - unfortunately I dont have Grønnum with me. Here is one source mentioning the sociolectal significance of the degree of frication of initial t.[6]·maunus · snunɐɯ· 09:42, 4 July 2015 (UTC)

That might do, thanks! We need to recheck Grønnum (2005) though. Try [7] for a free online access. And just one thing... what is low Copenhagenian? Does it mean Broad Copenhagen, Posh Copenhagen, or something else? I guess it's the first one. Peter238 (talk) 01:56, 5 July 2015 (UTC)

It is young, lower and middle class Copenhagenian. It has some traits from the traditional working class sociolects of the city and also some influence from immigrant Danish (where t is extremely affricated almost t͡ɕ before front vowels).·maunus · snunɐɯ· 06:33, 17 July 2015 (UTC)

Thanks! Peter238 (talk) 15:16, 17 July 2015 (UTC)

To be more exact, Grønnum (2005: 208) describes her linguistic data as representing “rigsmål as it is spoken by young [footnote: i.e., born around 1985] middle-class adults from the Greater Copenhagen area speaking spontaneously, but clearly”. She does also state, however, that “I will—if it can shed light on certain data—include my own knowledge of Danish as spoken by the older generations, and I will make use of the relationship to our closest neighbouring languages if convenient”, so it’s not purely Low Copenhagen or Young Standard Copenhagen. (A good example of this is her ‘fourth ø’, the sound she describes in grynte: it is separate for her from [œ] in general, but among the staff at the linguistics department at the University of Copenhagen—Grønnum’s old haunts—there is a general consensus that this is not phonetically distinguished by anyone still there. It is mostly based on Grønnum’s own idiolect.) Kokoshneta (talk) 19:33, 16 August 2015 (UTC)

Thanks for the input. I'm aware of the [œ̞–ɶ] (well, [œ–ɶ̝] to be completely exact) issue in Danish phonetics. Peter238 (talk) 19:42, 16 August 2015 (UTC)

@Maunus: any luck with Grønnum (2005)? I haven't found anything, but that may be either because Google Books limits its previews, my reading skills of Danish are questionable, or both. Either way, I've just found this: "Recently, though, the traditionally esteemed 'High' Copenhagian has given way to what Tore Kristiansen (2001) describes as 'Low' Copenhagian, a variety that is spoken in the outskirts of Copenhagen. This is the variety that Danes generally seem to orient towards nowadays." Source: [8]. Peter238 (talk) 23:34, 16 July 2015 (UTC)

Here is another link at dialekt.ku.dk - Low copenhagenian has indeed replaced high copenhagenian as the de facto standard variety. There is also this: [9]. And this one describes ts/t as a sociolinguiistic variable in youth speech.[http://curis.ku.dk/ws/files/34335460/Fonetisk_variation_som_social_praksis.pdf Maegaard, M. (2010). Fonetisk variation som social praksis: belyst ved nyudviklinger i ungt københavnsk. NyS,

39, 33-63. ]·maunus · snunɐɯ· 06:45, 20 July 2015 (UTC)

No, I havent actually thought more about looking for it. I should have it in a bookbox in storage somewhere.·maunus · snunɐɯ· 06:33, 17 July 2015 (UTC)

Ok, no rush. Whenever you have time. Peter238 (talk) 15:16, 17 July 2015 (UTC)

I’m not sure where to add this, so I’ll just add it at the bottom here. The distinction between [tˢ] and [t͡s] is relevant for Danish, since they are phonemically distinguished by at least some speakers (myself included): [tˢ] is syllable-initial /t/ before a full vowel, while [t͡s] is /ts/. Minimal pairs are not easy to come by, but the initial consonant in tæt [tˢεd̥] and tsetse(flue) [ˈt͡sε.t͡sε] are different—the latter has much longer affrication than the former. This does not mean that Danish has a phoneme /t͡s/, of course—it’s still much easier to assume that [t͡s] is just /t + s/ phonemically. But it does mean that there is a phonemic and phonetic distinction between a stop with a relatively short period of affrication and a stop with a relatively long period of affrication, which makes it at least highly justifiable to indicate the ‘strong’ allophone of /t/ as [tˢ] and the combination of /t + s/ as [t͡s]. Kokoshneta (talk) 19:33, 16 August 2015 (UTC)

This is easily solved by transcribing the sequence /ts/ as [t͡ss], or simply [ts], without the tie-bar. In Polish IPA, the latter solution is implemented, so czysta (clean (f.)) is transcribed /ˈt͡ʂɨsta/, whereas trzysta (three hundred) is transcribed /ˈtʂɨsta/. Peter238 (talk) 19:42, 16 August 2015 (UTC)

Inline citations

Parts of the article are really poorly sourced. Peter238 (talk) 00:39, 18 October 2014 (UTC)

I've removed much of the challenged material. Editors who wish to reincorporate the material with proper sourcing can look through the article's edit history to see the claims being made. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 21:46, 19 August 2015 (UTC)

Thanks! Peter238 (talk) 22:18, 19 August 2015 (UTC)

Horats

This name, which is shown (correctly, I think) as being stressed on the second (i.e. last) of its two syllables, is supposedly an example of Greek and Latin words being stressed on either the penultimate (i.e. second last) or the antepenultimate (third last) syllable. So this last 'rule' evidently isn't one, and is invalidated by the example that's given to illustrate it!213.127.210.95 (talk) 14:42, 16 April 2016 (UTC)

The sentence "Danish is a Scandinavian language related closely to Swedish and Norwegian, and more distantly to Icelandic and Faroese as well as to the other Germanic languages."

Danish and Swedish are descended of Old East Norse, whereas Norwegian, Icelandic and Faroese are descended of Old West Norse. Thus, (Danish;Swedish;Norwegian) is not a natural class within (North-Germanic languages).

Granted, Norwegian is much more similar to the other two than it is to its closest genealogical sisters, but this doesn't have any bearing on actual relatedness, which is the study of last common ancestors. I think that either it needs to be corrected to

"Danish is a Scandinavian language related closely to Swedish, and more distantly to Norwegian, Icelandic and Faroese as well as to the other Germanic languages."

or the word "related" has to change.

Shuntooth (talk) 05:30, 12 September 2016 (UTC)

• There is no need to add the distinction between East and West Norse in that sentence. First of all readers are unlikely to read it as a statement of genealogy, and more likely to read it as a statement of general similarity - in which case Danish is arguably closer to Norwegian (at leats to Bokmål) than to Swedish. Secondly since the genealogical relations are given in the main article it is unnecessary to get into those details here.·maunus · snunɐɯ· 06:28, 12 September 2016 (UTC)

"Foreign" phonemes?

Recent English borrowings are typically pronounced as approximated English, such that one might postulate phonemes like /θ/ (thriller, thousand island-dressing) or /ɹ/ (browser, rap). Similarly, Liechtenstein is pronounced with /ç/ by some speakers, and some preserve nasalization in French rendezvous, impromptu. Should this page include some discussion of at least the most common of these?__Gamren (talk) 14:50, 26 October 2016 (UTC)

Sure, go ahead and include them. I suppose the source is Den Danske Ordbog? Mr KEBAB (talk) 14:59, 26 October 2016 (UTC)

I would advise against discussing them unless the source is one of the major works on Danish phonology. Per WP:OR and WP:UNDUE.·maunus · snunɐɯ· 15:06, 26 October 2016 (UTC)

Well, DDO is pretty notable I'd say. Mr KEBAB (talk) 15:13, 26 October 2016 (UTC)

The above is my own analysis, but that is obviously not a source. I do not know if the literature bespeaks this. I may examine the latter, if noone beats me to it.__Gamren (talk) 17:13, 26 October 2016 (UTC)

There you go: [10] Mr KEBAB (talk) 18:58, 26 October 2016 (UTC)

DDO is not a work about Danish phonology.·maunus · snunɐɯ· 18:59, 26 October 2016 (UTC)

It's one of the major online dictionaries for Danish, it also provides IPA for many (most?) of its entries. One could say that one of its purposes is to be an online pronunciation dictionary (kind of like Lindsey's CUBE for the General British accent of English), as, AFAIK, apart from it, there's no such thing. You can always check Basbøll (2005) and Grønnum (2005) to see whether they have anything to say about that. Mr KEBAB (talk) 19:04, 26 October 2016 (UTC)

But it gives pronunciation (phonetics) not a phonological analysis that would allow us to posit new phonemes without incurring in original research.·maunus · snunɐɯ· 07:16, 4 November 2016 (UTC)

Fair enough. Their phonemic status is unclear, that's true. Maybe we need a better source after all. Mr KEBAB (talk) 14:11, 4 November 2016 (UTC)

Double brackets

Sorry, I reverted again. Not using double brackets in the vowels section makes it a lot less readable. Besides, it's not OR, as it's used by e.g. Basboell (2005). I see no problem whatsoever with using it, especially considering how bad the (pseudo-)phonetic IPA for Danish normally is (at least in case of unrounded front vowels). Mr KEBAB (talk) 21:20, 3 November 2016 (UTC)

What do double brackets mean? There's no mention of a phonetics- or phonology-related use in in the section of the Brackets article. — Eru·tuon 22:12, 3 November 2016 (UTC)

'Narrow phonetic transcription'. Mr KEBAB (talk) 23:35, 3 November 2016 (UTC)

That's what brackets are for. I don't think it's a good idea to have a special notation like that. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 23:46, 3 November 2016 (UTC)

Ok, why not? Danish transcriptions make extensive use for phonetic brackets, but the symbol choices for vowels are sometimes poor and need clarifying. I don't think that differentiating between single vs. double brackets does any harm in this case. It's not even OR. Mr KEBAB (talk) 01:37, 4 November 2016 (UTC)

I also think it is to specialized a usage to be useful in a general encyclopedia - I think it is highly unlikely to aid anyone and quite likely to confuse. Also I don't see why we would ever need to use narrow transcriptions really. In general I think this article suffers from an exaggerated degree of detailed information while being extremely weak on presentation of basic information for a lay audience. We ought to be very judicious in deciding when additional informaiton is helpful.·maunus · snunɐɯ· 07:14, 4 November 2016 (UTC)

I object to removing narrow IPA from the article. The fact that we have weak presentation of one area does not mean that we have to 'dumb down' what is already good and detailed - it means that we have to improve what needs improving. Mr KEBAB (talk) 14:13, 4 November 2016 (UTC)

Removing narrow transcriptions is not "dumbing down". It is focusing on the topic and avoiding informaiton that is off topic. The article is about phonology and not phonetics. Narrow trancription does not describe "Danish" but describes a specific instance of some individual pronouncing a Danish word - that is out of scope for this article. The weakness of the article is exactly that overzealous phoneticians have tried to include every aspect of pronunciation instead of sticking to describing the phonology (which includes basics of pronunciation such as allophony, but does not include detailed transcription of spcieific pronunciation that requires narrow transcription). Narrow transcriptions are not necessary when describing the phonology of a standard language. Using narrow transcriptions does should not be used unless they contribute information that is necessary for understanding the language's phonology. ·maunus · snunɐɯ· 12:08, 5 November 2016 (UTC)

We both know that "X phonology" is just a convenient shortening of "X phonetics and phonology" (so maybe a change of title is called for). Not many of our articles titled "X phonology" are actually mainly about phonology, they're equally about phonology or phonetics or (as is this article) mainly about phonetics with phonology being mentioned as a minor topic.

Yes, Danish does require narrow transcription to clarify e.g. the serious misuse of the symbols [ɛ, e], which actually are [e, e̝]. To even consider removing such info... to me is madness, and even e.g. Basboell in his book (titled Phonology of Danish, not Phonetics and Phonology of Danish) transcribes every Danish sound in narrow transcription whenever it's appropriate.

We're not 'respelling' every single word in narrow IPA (in such case, your complain would be valid). We're saying one time what the symbol actually represents, and then continue using it.

I consider it absolutely unacceptable to remove narrow transcription from the article. The fact that this article is about Danish phonology does not mean that we shouldn't cover its basic phonetics, no matter whether they're relevant to the phonology - they're relevant to people who actually want to know how Danish is pronounced, something the pseudo-IPA used by e.g. Basboell or Groennum won't tell them (they avoid using the "raised" diacritic on vowels at all costs, but then they use unnecessary complex symbols, e.g. [b̥] instead of [p], which is total hypocrisy). Mr KEBAB (talk) 15:18, 5 November 2016 (UTC)

I think I might have a sort of compromise solution. Kebab, you seem to be under the impression that we must replicate the symbols used in academic literature, even if they are phonetically imprecise. But we are under no real obligation to do so. In Russian phonology, a number of the sibilants are typically represented with the basic postalveolar symbols (e.g. ⟨ʃ ʒ⟩) but, based on phonetic literature on Russian, we have opted to use more precise symbols. With that understanding, we can simply use the more accurate symbols in phonetic brackets in as narrow a representation as we would like.

I would like to echo Maunus's concern that we not be too phonetically detailed in our transcriptions. Spanish phonology might be a helpful guide. The phonetics are covered in well-researched detail for a number of the sounds, but these details aren't necessarily represented in all the transcriptions. There's a balance that should be struck, but a lack of precision in some of these transcriptions is no reason to make notational parsing of two different types of phonetic transcriptions. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 17:44, 5 November 2016 (UTC)

I don't know. Maybe I misunderstood him to a certain extent, but I do think he's being too rigid with wanting to restrict the scope of this article to phonology alone. There's simply no good reason for that, not in the case of Danish at least...

Just to clarify: in the latest post, touching on the subject of double brackets was not my intention. I stand by what I wrote no matter whether we decide to keep the double brackets or not.

Sorry Aeusoes, but the postalveolar sibilants in Russian are a bad comparison. In Danish, we have a number of overlapping symbols, e.g. semi-narrow [a, æ, ɛ, e] are narrow [æ, ɛ, e, e̝] - 3 out of 4 symbols overlap. So using both transcription systems is not really an option unless we clarify each and every time whether we mean the semi-narrow or narrow transcription. That is the reason for which we should use the semi-narrow transcription in almost every case, excluding the situation in which we want to explain the actual phonetic representation of a certain symbol (e.g. [ɛ] = [e]). Mr KEBAB (talk) 18:25, 5 November 2016 (UTC)

We need to use the transcription used by the best sources, also when we disagree with it. We should stick to phonemes and allophones, not narrow phonetic transcription unless the sources use it to illustrate some important aspect of the phonology. I doubt very much that Basbøll uses narrow transcription except when he is transcribing specific utterances frm specific speakers as data. Your personal disagreements with Grønnum and Basbæll are irrelevant they are the sources we need to follow. We can however discuss whether to use Dania or not or whether to retranscribe Dania to straight IPA (following the conventions used by Grønnum and Basbøll). ·maunus · snunɐɯ· 19:55, 5 November 2016 (UTC)

Let me clarify, Kebab. When I say we shouldn't make notational parsing of two different types of phonetic transcriptions, I mean we shouldn't be using two different symbols for the same phone. Whatever level of precision we agree on, we should not have overlapping symbols.

Maunus, I'm not sure I understand your distinction between allophones and narrow phonetic transcription. I think we might generally agree about the degree of precision, but I don't think we need to be wedded to how precise the sources are. For our purposes here, we should be sticking to broader transcriptions unless we're illustrating an important part of phonology or phonetics. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 21:30, 5 November 2016 (UTC)

In Danish a lot of allophony is quite systematic for the most part and describing it is necessary in order to describe the way that phonemes result in standard Danish pronunciation. "standard Danish pronunciation" is of course something abstract and fairly variable for which reason one does not need descriptions of standard danish pronunciation to have a high degree of detail - just enough to capture the types of pronunciation that are within the scope variation that can be called "standard Danish". This in my view is not narrow transcription - narrow transcription is what one uses to describe specific instances of pronunciation in actual utterances as accurately as possible - narrow transcription is furthermore distinguished by not having gone through any phonological analysis but being simply a graphic representation of the acoustic signal as perceived by the transcriber (i.e. it is subjective and pre-phonological). There may be occasions where narrow transcriptions is warranted in this article, but I cannot really see when. And certainly there is no need whatsoever for implementing a convention for distinguishing between narrow and broad transcription - that will only serve to confuse the average reader. Here even several editors who are well versed in phonology had no idea what the double brackets were supposed to mean - the reader that we are trying to teach about Danish phonology will have even less of a chance to understand it. To me it is a general problem for all of our phonology articles that they tend to delve deeply into phonetics giving copious examples (usually based on OR) without ever giving adequate rerview of the actual phonology of the language - often ignoring stuff like phonotactics, syllable structure etc.·maunus · snunɐɯ· 07:02, 6 November 2016 (UTC)

Our article on phonetic transcription describes narrow transcription as that which "encodes more information about the phonetic variations of the specific allophones in the utterance" than broad transcriptions. The meaning you provide for "narrow transcription" is a much more restrictive one, making it basically inaccurate. But, given that meaning you've provided, I don't disagree with the point you are making. Just keep in mind that broad and narrow exist in a spectrum.

The kind of precision you refer to above as "narrow transcription" can have a place in this article (just as it has it some of our other phonology pages, even if there is a problem of OR), but I think that place is pretty restrictive. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 06:23, 7 November 2016 (UTC)

Lengthy responses were added just when I was about to reply, so I delayed my response until today. I generally agree with Aeusoes.

Let me just reply to Your personal disagreements with Grønnum and Basbæll are irrelevant they are the sources we need to follow. - Yes, that's obvious, and I wasn't suggesting not following them (that'd be unreasonable and nobody would listen anyway). I was just trying to illustrate my point.

I propose converting much of the Vowels section into a table, so that the narrow transcription of vowels can be easily accessed, but won't take so much space. Who's with me? Mr KEBAB (talk) 13:48, 7 November 2016 (UTC)

I think that is a good idea. And if we are using narrow transcriptions, I suggest that instead of double brackets we make a note at the top of the table or the subsection that sues them that states "the following transcriptions are narrow and based on x source".·maunus · snunɐɯ· 15:24, 7 November 2016 (UTC)

OK, will do it tonight. Mr KEBAB (talk) 15:47, 7 November 2016 (UTC)

Sorry, it took me longer than I hoped because of a table bug I couldn't fix (or was it just my Firefox? Anyway, it's fixed now). I'll post the table within a few hours, when I'm back home. Mr KEBAB (talk) 13:54, 9 November 2016 (UTC)

/r/ vs /ʁ/

Every /r/ in this article needs to be replaced with /ʁ/. The Danish "r" is a Voiced uvular fricative. Emil Kastberg (talk) 22:22, 28 March 2008 (UTC)

in phonemic script /r/ is used for the phoneme that is phonetically /ʁ/.·Maunus·ƛ· 14:54, 16 July 2008 (UTC)

why? Mechanic1c (talk) 23:16, 12 November 2016 (UTC)

For typing ease I suppose, and because the phonetic specificity is irrelevant in phonemic transcription - also because it is often not pronounced that as a voiced uvular fricative (that pronunciation is only in the beginning of syllables).·maunus · snunɐɯ· 23:29, 12 November 2016 (UTC)

fjende pronunciation by syllable

@Maunus:, according to Wiktionary, this word is pronunced as /ˈfjenə/, however, in a previous question concerning stød, you said hænder was /ˈhɛnˀ.ʌ/. For a word with also a missing d and a missing stød sound, is /ˈfje.nə/ the correct way to render this word? – AWESOME meeos ！ * (chōmtī hao /t͡ɕoːm˩˧.tiː˩˧ haw˦˥/) 20:56, 28 January 2017 (UTC)

There is no stød in fjende (except I believe in some Jutish dialects), so yes that is a potentially correct way to divide it into syllables. Another is /ˈfjen.ə/. I don't know a reason to consider one more correct than the other.·maunus · snunɐɯ· 21:08, 28 January 2017 (UTC)

@Maunus: Well I should consider the first one, /ˈfje.nə/, 'cause it seems easier to pronunce – AWESOME meeos ！ * (chōmtī hao /t͡ɕoːm˩˧.tiː˩˧ haw˦˥/) 21:19, 28 January 2017 (UTC)

I think that maybe if one thinks etymologically the /ˈfjen.ə/ may be preferable - since for example fjendtlig (hostile) would have to be /ˈfjent.li/. But I don't think it matters much.·maunus · snunɐɯ· 21:37, 28 January 2017 (UTC)

Unstressed words with at least two syllables

Hi @Maunus: Can you give me a list of words in Danish that are at least two syllables, which are unstressed or only with secondary stress? – AWESOME meeos ！ * (chōmtī hao /t͡ɕoːm˩˧.tiː˩˧ haw˦˥/) 03:13, 1 March 2017 (UTC)

I'm not sure how the latter would work. Correct me if I'm wrong, but secondary stress without primary stress would be primary stress (or simply 'stress'), at least when we're speaking about words in isolation.

One of the words you're looking for is eller 'or' which has a weak form [ɛlɐ]. Then again, when you pronounce it in isolation, you'd stress the first syllable ([ˈɛlɐ]). This is called a 'strong form'. Compare English was (weak [wəz], strong [ˈwʌz ~ ˈwɒz]). Mr KEBAB (talk) 04:07, 1 March 2017 (UTC)

I can't think of other two syllable words that have stresssless forms apart from eller, and I don't know where to look for others. I can ask my phonologist friends.·maunus · snunɐɯ· 07:17, 1 March 2017 (UTC)

There's also måtte [mʌd̥ə], from the JIPA transcription. Again, in isolation it would be [ˈmʌd̥ə], with the first syllable being the stressed one. Mr KEBAB (talk) 14:35, 1 March 2017 (UTC)

Ah, then all the auxiliary verbs would be like that I guess.·maunus · snunɐɯ· 14:58, 1 March 2017 (UTC)

• My phonologist friends say that all Danish polysyllables have stress when standing alone or used as "words", but in flowing speech all polysyllables can be unstressed. So there is no way to make a list. One of them noted that in the Danpass corpus dictionary there are also occurrences of eller with stress[11], though most are unstressed.·maunus · snunɐɯ· 07:07, 2 March 2017 (UTC)

aspiration when?

Article says:

/p, t, k/ are aspirated (and, in the case of /t/, also strongly affricated)[8] voiceless lenis in syllable onset: [b̥ʰ, d̥ˢʰ, ɡ̊ʰ] (hereafter transcribed as [pʰ, tˢ, kʰ] for simplicity). Aspiration is lost in syllable coda.[9]

where [9] is Grønnum (2005:303–305).
This seems to be easily contradicted by stil, spare, skære, where /t p k/ becomes unaspirated [d̥, b̥, ɡ̊] in syllable onset; it seems more reasonable to say that these sounds are aspirated in syllable-initial position and unaspirated everywhere else. I do not have p. 304 of the above work, but is it possible that someone misinterpreted it?__Gamren (talk) 11:27, 22 February 2017 (UTC)

Since the article provides no information on phonotactics we can't know if this is supposed to mean that the clusters in stil, spare, skære are phonemically interpreted as /sd̥, sb̥, sɡ̊/ and /p t k/ do not occur as second consonants in those clusters or if it is as you suggest an oversight not to mention that those consonants are unaspirated.·maunus · snunɐɯ· 13:14, 22 February 2017 (UTC)

The /p t k/ vs. /b d g/ contrast is neutralized in that context, so feel free to interpret it either way. --Schwa dk (talk) 14:00, 2 March 2017 (UTC)

Well, the choice of interpretation suggests a phonotactic analysis, so we should choose which we want to use. Do you know what is the more traditional analysis?·maunus · snunɐɯ· 14:09, 2 March 2017 (UTC)

Grønnum uses /st sp sk/. Basbøll uses /sb sd sg/. I think Grønnums approach might be considered more traditional. --Schwa dk (talk) 22:17, 2 March 2017 (UTC)

And Basbøll's closer to phonetic reality, I guess? ;) Mr KEBAB (talk) 23:40, 2 March 2017 (UTC)

Well, the stops are voiceless after /s/, so /p t k/ is closer to the narrow phonetic transcription [p t k]/[b̥ d̥ ɡ̊]. --Schwa dk (talk) 17:06, 3 March 2017 (UTC)

Correct. /p, t, k/ are aspirated in syllable onset, unaspirated elsewhere, as Grønnum says. --Schwa dk (talk) 14:00, 2 March 2017 (UTC)

Can /b, d, ɡ/ ever be voiced? If not, why do we transcribe the stops in such a strange manner, when only aspiration is phonemic? Mr KEBAB (talk) 17:29, 2 March 2017 (UTC)

Yes, they are very commonly voiced. /d/ > [ð̞], and /g/ > [ɪ̯~ʊ̯] in coda, as per the article. Those allophones are definitely voiced. Also, in spontaneous speech they are commonly voiced in voiced surroundings in onset position as well. /p t k/ are not aspirated except in initial position, so the /p t k/ vs. /b d g/ contrast is not just a matter of aspiration, but perhaps more a matter of fortis vs. lenis, like in many other languages. --Schwa dk (talk) 22:17, 2 March 2017 (UTC)

Oh, I forgot about the approximant allophones. My bad, although I wasn't aware that they're that common. However, the article (citing Grønnum) says that the stops are all lenis and so does Basbøll I guess (I don't have access to it now). Perhaps it is just a (underlying) voiceless-voiced distinction after all, as in northern standard German, Luxembourgish, Oslo Norwegian, etc.. However, with that being the case, I think that we should remove diacritics from the stop symbols (here, on Help:IPA for Danish and on pages that link to it). I think it's a too narrow a transcription for our purposes, and the diacriticless transcription is just as valid (and it is used, e.g. by Den Danske Ordbog). It is also a bit inconsistent to narrowly transcribe the stops, as the vowels are in broad transcription, and some of them are massively different from the corresponding CV's (notably [ɛ], which is close-mid). Mr KEBAB (talk) 23:31, 2 March 2017 (UTC)

Also, could you provide a citation about the voicing of the stop allophones? It's worth including in the article. Mr KEBAB (talk) 23:37, 2 March 2017 (UTC)

It's something Danish phoneticians are aware of, but I'm not sure anything has been published yet. --Schwa dk (talk) 17:17, 3 March 2017 (UTC)

Examples

These were added to the Sounds-section at Danish language, but they didn't seem quite at home there. Also, I don't think the phonemic analysis is all that correct.

• /'ilʔ/ <ild> "fire"
• /'i:lə/ <ile> "run"
• /'ilə/ <ilde> "bad/ill"
• /'et/ <et> "one"
• /'me:nə/ <mene> "mean/signify"
• /'ɛlə/ <ælde> "oldage"
• /'kɛ:lə/ <kæle> "cuddle"
• /'alə/ <alle> "everyone/all"
• /'ha:lə/ <hale> "tail"
• /'bɑkə/ <bakke> "hill"
• /'ɑ:wə/ <arve> "inherit"
• /'yðʌst/ <yderst> "outermost"
• /'y:ðə/ <yde> "yield/contribute"
• /'øst/ <øst> "east"
• /'ø:sə/ <øse> "pour"
• /'lœnʔ/ <løn> "wages/pay"
• /'ʁœ:bə/ <røbe> "disclose,show"
• /'lɶjʔ/ <løg> "onion"
• /'dɶje/ <døje> "endure/suffer hardship"
• /'huʔs/ <hus> "house"
• /'hu:lə/ <hule> "cave"
• /'foto/ <foto> "photograph"
• /'ro:sə/ <rose> "rose"
• /'rɔst/ <rust> "rust"
• /'rɔ:bə/ <råbe> "shout"
• /'kɒlʔ/ <kold> "cold"
• /'sɒ:wə/ <sove> "sleep"
• /'sɒwʌ/ <sover> "sleep (present tense)"
• /'ro:sə/ <rose> "rose"

Peter ^Isotalo 16:44, 17 August 2005 (UTC)

I added them, and while you may be right that they didn't belong there, I think there should be examples, exactly because it shows the phonemic analysis lying behind it. The phonemic analyses are based on Hans Basbøll, Jacques Durands article about Basbølls analysis of the Danish vowel system and Nina Grønnums book on Phonology. They are from my pronunciation which is close to rigsdansk. Would you care to tell me which parts you find incorrect? [[Magnuspharao]]

The glottal stop is not a phoneme in Danish. Also, if you want to include examples, you need to structure them properly (see Swedish phonology for example). And there needs to be more suitable minimal pairs. I can barely find any in this list that would seem obvious to someone who doesn't speak Danish (or a Mainland Scandinavian language). Preferably a set that illustrates more than two different phonemes, and if possible, nearly all. It's hard to find one, but you can usually come close. In Swedish, for example, there's /rVt/ that can be used to show the constrast between almost all but one of the 8 short vowel phonemes. Only */ryt/ is not an actual word.

And as for the article lacking phonology info, well... it's what where here to fix. I would do it, but I have trouble getting hold of Basbøll's The Phonology of Danish.

--Peter ^Isotalo 03:43, 19 August 2005 (UTC)

It is true that the transcription is sort of half phonemic half phonetic. Gloittal stop is the standard way of representing stød (the IPA variant adapted to danish is called Dania and uses it). Length is also not phonemic. While neither stød nor length are phonemes in danish they are prosodies that are used in distinguishing minimal pairs, and as so I think that it is natural to represent them in some way also in the phonemic transcription. [[Magnuspharao 06:12, 19 August 2005 (UTC)]]

Here are some minimal- and near minimal pairs that I've found in order to illustrate vowel qualities. If you like them you can put them in where they might be of use, if you don't then I will continue to work with it. Some of them are my own some of them are from Nina Grønnums "Fonetik og Fonologi" (akademisk forlag 1998)If you want me to I can record soundfiles of these examples, I just don't know which format to use or how to upload.

mile [mi:lə] - milde[milə]

mele [me:lə] - Mikkel [mekəl]

mæle [mɛ:lə] - melde[mɛlə]

male [mæ:lə] - malle [mælə]

arne [ɶ:ne] - bakke [bɶkə]

orne [ɒ:nə]/årle [ɒ:lə]

mule [mu:lə] - fulde [fulə]

mole [mo:lə] - mos [moʔs]

måle [mɔ:lə] - hulle [hɔlə]

smøle [smø:lə] - mølle [mølə]

syle [sy:lə] - myldre [mylRə]

si [siʔ]- sy [syʔ]

se [seʔ] - sø [søʔ]

sag [sæʔ]

zar [sɶʔ]

sug! [suʔ]

so [soʔ]

så (se pst.)[sɔʔ] - så (adv.)[sʌ]

These are hard to find good minimal pairs for because of their limited distribution:

døje [dɶjə]- løg [lɶjʔ] dølge [døljə]

kost (food)- [kʌst]kost "broom" [kɔst]

række[Rɐkə] - rakke [rɶkə]

skøn [skœnʔ] - skød [skøð]

[[Magnuspharao 12:05, 19 August 2005 (UTC)]]

Unfortunately, one cannot use half-phonemic transcriptions. If it's it not in brackets, it has to be 100% phonemic or it's not useful. Also, Dania is not applicable to Wikipedia since it is only used for Danish and has a lot of symbols that are either unique or are similar to IPA but actually represents different sounds. IPA is really the only way to go since there is no other widely recognized standard for phonetic transcription. Representing stød as a glottal stop is also not useful since it's more complex than that, so I wouldn't recommend it. Recording pronunciation files is a good idea and the info on how to do it and what format to use (.ogg only) can be found here. Make sure you upload it to Commons to make it available to all other Wikipedias. Finding minimal pairs can't be that hard, though. Just take your time and consult proper literature (written with IPA) before recording. Here I can recommend Basbøll's The Phonology of Danish, since it is sure to be have been written with IPA and has a reputation of being very authoritive.

Peter ^Isotalo 15:36, 19 August 2005 (UTC)

There is no IPA symbol for stød. Basbøll and Grønnum use a superscript Glottal stop sign. I don't know how to make the superscript so a glottal stop sign will have to do. I just presented a bunch of minimal pairs taken from Grønnums book and no they weren't that hard to find. But it's not possible to find the lax vowels in the same positions as the tense vowels or all long vowels in the same positions as vowels with stød etc. because of distributional limitations. I am using "proper literature" and, by necessity, it most often uses Dania and not IPA because IPA is not suitable for some danish phonological features such as stød and placement of some vowels (heavy use of diacritica would be the result, and you'd have to invent a symbol for stød anyway). Thanks for the comments, I am new to wikipedia and need to learn how things are done here. But maybe you could be a little more open minded, when it comes to my ways of presenting data? Maybe I'm not just another hack who knows nothing about the subject I'm writing about.[[Magnuspharao 20:26, 19 August 2005 (UTC)]]

There are special Unicode characters for superscripted IPA, but I usually just use the wikicode . To give an example, mål would be [mɔ:^ʔl]. You don't need to use Dania here, since IPA covers it all, including stød. And this isn't just my opinion, but a generally accepted guideline of Wikipedia; all phonetic transcriptions are to be in IPA with no exceptions. If you don't believe me, just ask our fellow wikiphoneticians at Wikipedia:WikiProject Phonetics. I'm going to help to improve this article quite a bit during the coming month, because I just got a hold of The Phonoloy of Danish, but I'm going to make sure Danish language gets a proper summary first.

Peter ^Isotalo 19:28, 1 September 2005 (UTC)

• /'i:lə/ <ilde> "run"

Should be <ile>. Ilde is an almost obsolete word for bad, and the d marks the i as short. /'ilə/, if I understand the notation. --Klausok 08:04, 22 January 2007 (UTC)

I have corrected the above : /'ilə/ <ilde> "run" to: /'ilə/ <ilde> "bad/ill" Examples: Jeg er ilde/jeg har det ilde = I feel a litle sick/not quite well Det er ilde set = it is frown upon Det er ildelugtende = it smells bad Lerura (talk) 00:05, 1 June 2017 (UTC)

Non-syllabic vowels

It's existence is mentioned, but not explained. What are some words that uses these? Are they allophones? Can you predict them from the spelling? A quick browser search yields nothing but Wikipedia copy-cats. Does these non-syllable vowels even exist? — Preceding unsigned comment added by 69.206.181.191 (talk) 19:42, 1 August 2018‎ (UTC)

[ɐ̯] is an allophone of /r/ in the syllable coda. It can even appear before vowels within the same word, as in næring [ˈnɛːɐ̯eŋ] ('food'). [ɪ̯] appears as an allophone of /ɡ/ after front vowels and maybe sometimes /j/, but I'm not sure about the latter. [ʊ̯] appears as an allophone of /ɡ/ after back vowels and sometimes it's also an allophone of /b/.

In some transcription systems [ɪ̯, ʊ̯] are transcribed with ⟨j, w⟩. The latter isn't wrong but it's rather broad, as these sounds are rather open (about close-mid). Also, when you transcribe [ɪ̯] with ⟨j⟩ then you can't show the difference between [ɪ̯] and [j], which is one of closeness ([j] can even become a fricative [ʝ] when it follows /l/ in a word-final position).

I hope this makes sense (and that I got everything right, I'm not a native speaker of Danish). Kbb2 (ex. Mr KEBAB) (talk) 23:29, 4 August 2018 (UTC)

no mention of /w/ or the word semivowel

Help:IPA/Danish says to come here for a more thorough look but even basic info is missing here. --Espoo (talk) 22:40, 29 January 2019 (UTC)

That sound is transcribed as ⟨ʊ̯⟩ here. It might be a good idea to be consistent. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 23:54, 29 January 2019 (UTC)

Vowel length

How come that this page does not explain when to pronounce a vowel as long or as short? It only say that those vowels exist. Unlike this German page there it does explain the vowel length. FanNihongo (talk) 02:33, 22 May 2019 (UTC)

@FanNihongo: This isn't Danish orthography but Danish phonology. You should ask your question on the talk page of the former. Kbb2 (ex. Mr KEBAB) (talk) 04:32, 22 May 2019 (UTC)

Ok, thanks for your answer. FanNihongo (talk) 06:48, 22 May 2019 (UTC)

The vowels are confused

The vowels were all placed in square brackets, regardless of whether they were phonemes, allophones or transcription symbols. I've tried to fix it up, but e.g. the table of allophones (which I didn't touch) seems hopelessly confused, transcribing allophones with the letter of the phoneme with the closest elsewhere pronunciation rather than actually transcribing their phonetic value. And while we have 15 phonemic vowel qualities in the list at the top of the section, and in the formant graph, we have only 10 in the table at bottom of the section. So are some of the vowels at top just allophones, or is this a difference in analysis? Because if they're allophones, it would be gibberish to present them the way that we do ([a] is sometimes pronounced [b] and sometimes [c] ...) We either need angle brackets for transcription conventions, double square brackets to indicate narrow transcription, or some other way to indicate what we're talking about. — kwami (talk) 08:28, 8 February 2020 (UTC)

See above. The phonemes should be written /i y u e ø o e̞ ø̞ ə o̞ ɛ a/ /i y u e̝ ø o e œ ə ɔ ɛ a/ and perhaps /ɐ/ (this isn't an actual phoneme, AFAICS). What we currently transcribe with ⟨ɶ ʌ ɒ a⟩ (actually ⟦œ ɒ ɔ æ⟧) aren't phonemes either. [ɶ ɒ] (meaning ⟦œ ɔ⟧) appear mostly in /r/-contexts and [ʌ a] (meaning ⟦ɒ æ⟧) are the short versions of what Basbøll writes /ɔ a/ (these should be written with ⟨o̞ ɛ⟩) (the latter should be written with ⟨ɛ⟩). Kbb2 (ex. Mr KEBAB) (talk) 09:43, 9 February 2020 (UTC)