Talk:Standard German phonology

Phonology vs. Orthography

I'm wondering how the german ö can be the same as the french oeu. If I'm not mistaken the sounds are completely different between the 2 letter combinations.

They are as much the same as any two sounds of different languages can be said to be the same. Both French and German have two different pronunciations of that sound: [ø] (as in jeûne [ʒøn] "abstination" or Höhle ['hø:lə] "cave") and [œ] (as in jeune [ʒœn] "young" or Hölle ['hœlə] "hell").

I'm thinking that this might more properly go in the German language article under an 'Orthography' heading. What do you think? -- Djinn112 21:04, Feb 13, 2004 (UTC)

To Djinn112: I started this as a separate German_pronunciation article in the style of the Esperanto_pronunciation article. However, you have more Wikipedia exeprience than me, so if you think it should go as a subsection of German language, then you may go ahead and move it. -- Anon.

I made an orthography section in German language and put a link to here under it. I noticed that the grammar section in that article outsources like that, so I suppose this all is fine as is. -- Djinn112 01:01, Feb 14, 2004 (UTC)

Phoneme /r/

I think that I've heard some uvular r's in German. Are they standard and, if so, how are they used? -- Djinn112 22:18, Feb 13, 2004 (UTC)

German 'r' is an uvular r when initial, and pronounced as a vowel, [ɐ], otherwise. Reading the article makes me suspect it's based on the Bühnenaussprache, a 19th-century pronunciation standard that is no longer being followed (I hope). Today's Standard German is quite different and I'm not even going to try describing it in SAMPA. What's wrong with IPA, anyway? Prumpf 22:56, 20 Jul 2004 (UTC)

This is wrong. The most common form in (Northern) Germany is a uvular fricative or approximant. A uvular trill is also occasionally used, even more frequently on TV. These are also described in Duden's "Aussprachewörterbuch" (4. ed., 2000, p. 53f.) as the most frequent forms (well, not the approximant). I would not accept an alveolar trill as standard pronunciation, but Bavarians will certainly object.

Not only Bavarians - Austrians too... As you said, the uvular r is the most common form in Northern Germany. In fact, in Austria the alveolar trill is considered to be standard. IMO the "free allophones" solution is the best you can offer. Kubusj 19:41, 12 Feb 2006 (UTC)

I am an Austrian and never use the alveolar trill. Until recently I didn't even know how to articulate it. I'd say it's acceptable in standard Austrian German – some TV and radio journalists use it –, but it's a distinctive feature of a few dialects, not including mine.

David Marjanović | david.marjanovic_at_gmx.at | 17:23 CET | 2006/11/11

Out of interest: Where do you come from exactly? I've heared uvular r from speakers from Graz, Linz, and especially from Viennese people (besides from people from Tyrol and Vorarlberg). But my family uses an alveolar trill exclusively. Qubux 00:36, 30 April 2007 (UTC)

Apparently Duden now considers [ɐ] in syllable rhymes standard (according to my phonetics lecturer - I haven't read it myself, though this symbol is used as such in my Oxford-Duden bilingual dictionary). Hairy Dude 02:34, 4 February 2006 (UTC)

There is a wide range of regional allophony in the pronunciation of standard German /r/. In most varieties of standard German spoken in Germany, the allophonic [ɐ] may be common, but this article is not Germany phonology, but German phonology. So I think the current article text does not need any change: [r], [ʁ] and [ʀ] are free allophones of each other. [r] is used only in Southern varieties. In the syllable coda, the allophone [ɐ] is used in many varieties, except in the South-West. ― j. 'mach' wust | ✑ 11:53, 6 February 2006 (UTC)

I do not think they should be called "allophones". This would imply that /r/ is pronounced one way in some contexts and another in other contexts. But I don't know of anyone who uses more than one rhotic consonant (at the very least while speaking the same lect). Most people who use one rhotic can't even pronounce the others. They are regional variants, I'd say, not allophones.

David Marjanović | david.marjanovic_at_gmx.at | 17:26 CET | 2006/11/11

I should say it is even more complicated. I know of few regions where the oldest generation of rural speakers consistently use uvular r, whereas there are, to my knowledge, no towns of any significant size in Germany where a majority of young speakers have alveolar r. Alveolar r is receding rapidly and has been so for some generations. Outside Bavaria, speakers under 50 seldom have it, but it is still a majority feature in many parts of rural Bavaria and it can be heard from many middle-aged speakers in the rural parts of South and Central Hessen, the Palatinate and along the North Sea Coast. --129.35.231.2 14:04, 24 July 2006 (UTC)

Baltic Sea Coast too. The only non-Bavarian I've ever met with an alveolar r was from East Prussia. User:Angr 17:37, 24 July 2006 (UTC)

Phonology vs. Orthography 2

It feels like it could turn into a Phonology of German article. It would need a fair amount of work of course. Secretlondon 22:24, Feb 13, 2004 (UTC)

As a native german speaker, I can tell you that a lot of this article is simply FALSE. For example, Bund and bunt are pronounced differently whereas the article tells that both are pronounced [t] at the end. Alas, I don't know SAMPA and so I can't correct the article... But there is really a lot of work to do. 195.14.206.167 13:36, 16 Jul 2004 (UTC)

The pronunciation described certainly is a rather odd one, to say the least. However, while that might not be true in your dialect, Bund and bunt are pronounced the same in Standard German. Prumpf 22:56, 20 Jul 2004 (UTC)

No, they aren´t. The "t" is spoken a little harder, than the "d". And you missed out some specialities like "Oer-Erkenschwick" (a City) the "e" of "Oer" ist quiet but the "O" is spoken longer. In many names for cities or places there is an "e" or "i" to show that the letter ahead of it is to be spoken longer. Kiss me, I´m German. 195.37.188.210 11:44, 10 Aug 2004 (UTC)

Please cite a source for the bund/bunt thing. I've never seen a description of Standard German that distinguishes final "voiced" plosives from their voiceless counterparts. I suppose we should add a note about proper names, but then those confuse even Germans.

Prumpf 17:47, 10 Aug 2004 (UTC)

"Bund" and "bunt" are pronounced the same in standard colloquial German, in spite of the fact that many native German speakers are misled by the spelling and by their schoolteachers to believe there is a difference. If asked to make it, they produce "Bund" with an unspirated final [t] and "bunt" with an aspirated final [t^h]. In ordinary speech when they're not thinking about it, though, people pronounce the two words the same. Certainly this page should not be concerned with the peculiarities of proper names. For example, is there any German word besides "(Bad) Oeynhausen" in which [ø:] is spelled "oey"? I don't think so. --Angr 07:09, 11 Jan 2005 (UTC)

Interesting. I consistently distinguish Bund and bunt – lenis vs fortis (both voiceless, both unaspirated). But this only seems to happen after /n/. Rad and Rat are homophones for me (both end in a lenis).

David Marjanović | david.marjanovic_at_gmx.at | 17:30 CET | 2006/11/11

I would certainly prefer if this article didn't concern itself with orthography at all (as with all phonologies, to be honest). An article like this should concern itself mainly with the phonetis of German, allophones, phonotactics and such. Orthography should be explained in German language or maybe even in German orthography. Peter Isotalo 17:34, May 7, 2005 (UTC)

Agreed. But unfortunately, German orthography redirects to German pronunciation. Someone needs to start that as a new page. BTW, should it be called "orthography" or "spelling"? Parallel articles are English spelling and Dutch orthography. (The technical difference is: orthography = spelling + punctuation.) I suppose whoever makes the effort to start the article gets to decide. There is, of course, already an article on the German spelling reform of 1996. --Doric Loon 19:27, 7 May 2005 (UTC)

This seems to be the place to ask/request something: Can someone add something about the 19th-century (or whenever it was) spelling reform, the one that eliminated -th for -t in words like Thal? I came to this article trying to find out about the change from -th to -t, and whether there was any real-life pronunciation difference at the time, and whether such differences still exist in dialects of today, and such questions. These questions seem to me to straddle the boundaries between orthography/spelling and phonology, as well as including a bit of history. Where should one put such topics? I don't know the answer. Dveej 13:48, 16 May 2006 (UTC)

There was no sound change corresponding to that spelling change. I think these issues are discussed at German orthography. I do know that at least until the unification of the German Empire in 1871, and maybe for some years after it, the different German states had different standards, so while you might encounter späth and Thal in some places, you'd encounter spät and Tal in others. Angr (t • c) 14:04, 16 May 2006 (UTC)

There simply was no standard orthography of German before 1901 (!). Without exception, th in native words was purely cosmetic (maybe to make it look like Greek), and elsewhere it was purely etymologic (Greek again).

David Marjanović | david.marjanovic_at_gmx.at | 17:31 CET | 2006/11/11

Phonology & tables

This article could use a reorganization into tables, similarly to other articles in the Language phonology category, and maybe a moving to German Phonology.

Would anyone object to moving this article to German phonology? Peter Isotalo 17:20, May 7, 2005 (UTC)

It would certainly sound more serious and academic: "pronunciation" sounds like a section in a phrase-book for tourists. --Doric Loon 19:27, 7 May 2005 (UTC)

OK, I moved it as requested, hopefully there aren't any objections to it now after the fact. Everyking 10:52, 8 May 2005 (UTC)

As far as I can tell, there's been consensus for this kind of naming for some time now at Wikipedia:WikiProject Phonetics. And it is rather logical, when you think of it. Peter Isotalo 10:57, May 8, 2005 (UTC)

Colloquial German phonology

It might be useful to include a section on colloquial German (which obviously varies by region). Maybe info on various phonological processes that occur only colloquially. -- jonsafari 19 May 2005

ʔ

The minimal pair [ʔaɪs raɪs] is not very illustrative as long as it's not opposed to the corresponding pair of Southern varieties [aɪs raɪs]. J. 'mach' wust • tskʃpræːx 00:58, 20 May 2005 (UTC)

"Tau-chen" or "Täu-chen"?

The diminutive ending usually triggers umlaut. This should be mentioned. -- j. 'mach' wust ^{ˈtʰɔ̝ːk͡x} 14:09, 26 May 2005 (UTC)

Velar vs. uvular

It makes no sense to mention the uvular allophone of /x/ if the uvular allophones of /k/, /g/, /ŋ/ are not mentioned. -- j. 'mach' wust ^{ˈtʰɔ̝ːk͡x} 29 June 2005 22:23 (UTC)

Do you have uvular allophones of these in Switzerland? And are they considered standard? ~:-| David Marjanović 10:20, 29 April 2007 (UTC)

Yes, Allemannic speakers have these allophones (and to a lesser extent also Tyrolian speakers). But I don't think they are considered standard, at least in Austria. Switzerland may be another case, though (and Vorarlberians would surly object, too). Qubux 00:42, 30 April 2007 (UTC)

Varieties of Standard German

The main problem of this article is the problem of the varieties. Most of this article describes the standard pronunciation of Germany, vowels and consonants. But then other varieties are included. This is misleading. You would have to include the differences of the varieties in the entire article; vowels, consonants, ich-Laut and ach-Laut and so on. At least make clear every time, if only the German of Germany is tackled.

But then, there's another problem. Other varieties (Austrian, Swiss) are not codified nor widely accepted. I believe, this article should AT LEAST make clear, what the standard of German pronunciation in Germany is. It's the only widely accepted, well described and codified form of German standard pronunciation today. This is the point to start and this is what is most useful to students of the language.

There is no codified standard pronunciation, not even in Germany. The article describes the pronunciation shared by all varieties as well as important differences. If you think more differences should be included, go ahead and do so. From my Swiss point of view, I think the article is acceptable. -- j. 'mach' wust ^{ˈtʰɔ̝ːk͡x} 3 July 2005 22:51 (UTC)

One should make a clear distinction between codified standard(s) and spoken standard(s). Confusing students of the languages with the peculiarities of spoken standards of "northern varieties" and "southern varieties" doesn't seem to be right for me. There is a de facto codified standard pronunciation for the teaching of German as a foreign language in Germany. This should be the core of the article probably even from the Swiss point of view. BTW, I do not volunteer for the job.

So by codified you mean 'put on record' rather than 'officially sanctioned' or even 'prescriptive'. This article is not intended to be a teaching aid for German. If I remember correctly, there's a wikibook with that purpose. -- j. 'mach' wust ^{ˈtʰɔ̝ːk͡x} 5 July 2005 07:35 (UTC)

It's not just a teaching aid for German, but its purpose is to serve the reader and you have to be open about who might be using it. A lot of people looking up this article will be people wanting to learn German. That's OK. Indeed, I suspect that people NOT wanting to learn German will not be too interested in this kind of detail anyway. So yes, we should think (among other things) about what is helpful for students of the language. --Doric Loon 5 July 2005 16:23 (UTC)

/ch/-Laut-Section

This section is much to long. After all, this is only a phonological process. There are many others not yet tackled.

Front and back a

The article mentions that /a/ and /ɑ/ are free allophones. But there are quite a few minimal pairs. For example:

• wann /ʋɑn/ "when"
• wahn /ʋan/ "delusion"

and

• bann /bɑn/ "spell"
• bahn /ban/ "road"

In my opinion they are really different phonemes. Anyone comment? −Woodstone 08:05, August 3, 2005 (UTC)

I agree with Woodstone. --Doric Loon 08:25, 3 August 2005 (UTC)

Sorry, I think I reversed the symbols, now corrected. −Woodstone 11:12, August 3, 2005 (UTC)

I haven't seen any analysis yet that doesn't mark the length. Sources? However, many analysis don't mark any difference in vowel quality but use /a/ vs. /aː/, for instance A. Linke/M. Nussbaumer/P. R. Portmann (1996): Studienbuch Linguistik, Tübingen:Niemeyer, p. 428.

There seems to be no consensus on which of the two a-phonemes should be the back vowel. In the phonology course at Bern university, I have been teached it's /ɑ/ vs. /aː/ (similar to Woodstone's correction), whereas I've seen /a/ vs. /ɑː/ in other places (similar to Woodstone's first version). Therefore, I think the distinction of two different vowel qualities among the two a-phonemes is somewhat arbitrary. -- j. 'mach' wust | ‽ 11:28, 3 August 2005 (UTC)

The Bantam New College German and English Dictionary lists the German pronunciation entirely as quality, not length. The back a is the long one (Wahn) and the front a the short one (wann). This is supported by Comrie The World's Major Languages p. 121, which asserts that the length distinction is "more accurately described as a difference of tense versus lax articulation". The reason is that "There are perceptible differences between tense /i:/ in /di:ne:/ Diner and lax /i/ in /difu:s/ diffus, ... and yet both i vowels are technically short." I don't know if a similar distinction exists for /a/, and if so, what the quality difference is in unstressed syllables. Benwing 06:51, 6 August 2005 (UTC)

The IPA handbook has only /a/ (hatten) and /aː/ (baten) for German. However for Dutch it has back /ɑ/ (bad) and front /aː/ (zaad). "Zaad" sounds similar to "Wahn" and "bad" to "wann" so the Dutch choice is just the reverse of the Bantam dictionary cited above by Benwing. In Dutch the opposition between front and back a is very phonemic (but usually accompanied by a slight lengthening of the front a). Listening to the sound files at the IPA site, I would say the Durch assignment is correct. This leaves me to believe that the choices for German (Bantam) and French /ɑ/ (pâte) and /aː/(patte/) is wrong. So I'm still utterly confused. Who can help me?

In German, the difference between Stadt and Staat is both quantitative and qualitative, but the length distinction is more obvious. We should definately transcribe in a way which indicates both aspects. This applies to all Germanic languages including English, though in English length is less obvious. But even there, a trascription which covers both aspects can't be wrong. --Doric Loon 13:42, 6 August 2005 (UTC)

The answer above misses my point. I agree that indicating both aspects is of value. But the real question is which is the "front" and "back" variant of a.

• German "Staat", "Wahn", "baten", Dutch "zaad" and French "pâte" have a similar a sound
• German "Stadt", "wann", "hatten", Dutch "bad" and French "patte" have a similar a sound

As I hear them (comparing to IPA sound file) the former is the long front a /aː/, the latter the short back a /ɑ/. This corresponds to the IPA handbook usage in Dutch, but conflicts with the Bantam dictionary for German (as quoted above) and several sources for French usage.

Dutch a and aa vary between [ɑ, aː] and [a, ɑː] depending on the generation of the speaker, even just within Amsterdam. The French distinction has collapsed for most speakers, but where maintained it is definitely patte [a], pâte [ɑ]. The former is like Spanish casa, the latter like English father. (I don't believe there's much difference in length any more, but if there were, it would be [a, ɑː].) Ladefoged has [a, ɑː] for German in an exercise in his introductory A Course in Phonetics. kwami 11:29, 2005 August 7 (UTC)

Kwamikagami is right, the French distinction is not much present any more, but when it was, the sound in pâte was much farther back than that in patte, and, I´m pretty sure, rounded; hence it is actually more like the sound in British ´not´, but longer. The â sound was long, as is the sound in ´faute´, as is the sound in jeûne "fast (n.)", as are all nasal vowels, and as are all vowels before /r/, /v/, /vr/, /z/, and ezh (as in ´plage´). (Hint: assuming the actual value for French /r/, what do all five of these conditions have in common?)

The real question is, though, are the two a´s distinguishable in non-stressed syllables? German native speakers, is there a difference between the first a in Kanone vs. Kannibale? Is there a difference in the unstressed first a´s in kanonisieren vs. Kannelüre, or Kanone vs. kanadisch? What about between any of the previous six? (I.e. are they all alike, or some different?) If all six are alike, is there a difference in the first vowel between Diner and diffus? Benwing 12:16, 7 August 2005 (UTC)

Kanone vs Kannibale --> There is a difference for some speakers, for some there isn't. As a native speaker of German, I think the article is correct in saying that /a/ and /ɑ/ are free allophones. According to Siebs, it should be /a/ and /ɑ:/, but this distinction is now obsolete. For present-day German, there is no standardised distribution I am aware of. In fact /a/, /a:/, /ɑ/, /ɑ:/ can occur in all possible combinations depending on the speaker's native dialect.

It's not becoming much clearer. There seems to be wide disagreement on what the symbols stand for. On the question of unstressed a, I would say that the difference collapses both in German and Dutch (to [ɑ]). In the Amsterdam dialect all a's are close to [ɑ]. In French, to widen the question: would it be correct to say that "accepter" has [ɑ] and "aligner" has [a]?

No, they should both be [a]. In Amsterdam there is great speaker variation, so no, I don't think all aes are close to [ɑ]. Given the variety of German dialects, is everyone talking about the same one? kwami

I, as a native speaker of both German and Dutch experience German 'ah' (as in Hahn) and Dutch 'aa' (as in haan) as [a], and German and Dutch 'a' (as in 'kann' and 'kan') as [α]. People who thus pronounce these characters will call the first a 'long a', the second a 'short a'. [oliver lenz]

Anthony Fox used /ɑ/ and /a/ for long and short a's in his book. It's the opposite of the above. He used in the following way:

/i/ : /ɪ/ = bieten : bitten

/e/ : /ɛ/ = beten : Betten

/u/ : /ʊ/ = spuken : spucken

/o/ : /ɔ/ = Ofen : offen

/ɑ/ : /a/ = Staat : Stadt (neutralized when not stressed)

The last line doesn't seem correct. Is it Siebs's pronunciation? - TAKASUGI Shinji 03:31, 7 April 2006 (UTC)

french version

hi. i saw the french version here: fr:Prononciation_allemande. maybe everyone has already seen this, but if not, maybe its useful. anyway, has nice pictures. peace – ishwar  (speak) 01:19, 2005 August 6 (UTC)

Merge current orthography article

The article German orthography is not about German orthography at all. (See Orthography.) It discusses phonology as related to or discernable from the spelling of a word. It might therefore make a good addition to this article.

There is no phonological information in the orthography article whatsoever. It merely describes the pronunciations of certain letters or letter combinations. If you're not satisified with the other article, fix it or redirect it to German language. Merging it with this article will only make people add more irrelevant orthography information here. I'm taking the merge sign down for now.

Peter ^Isotalo 20:42, 29 August 2005 (UTC)

Affricates missing ‼

The affricates [ʦ], [ʧ] and most importantly, [pf] MUST be mentioned. This article won't be complete without them. By the way, [ʤ] (often written as 'dsch') could be included as well, but it exists in loanwords only. --Pipifax 19:40, 8 September 2005 (UTC)

They've been already listed. By the way, some linguists think the German [ʧ] is not a single phoneme but a sequence of /t/ and /ʃ/ because it doesn't appear in a consonant cluster while the true phonemes /pf/ and /ʦ/ do like Pfropf /pfropf/ and zwei /ʦvaɪ/, and treat the /tʃ/ in /dɔʏtʃ/ (Deutsch) just like the /pʃ/ in /hʏpʃ/ (hübsch). What do you guys think? - TAKASUGI Shinji 04:53, 7 April 2006 (UTC)

Phonologically you can treat it that way - but phonetically, it's an affricate, not a simple stop followed by a fricative. Well, in this case we should also mention other non-phonological, phonetical affricates (like <bsch>, <x>, <ps> and so on... Qubux 17:24, 15 May 2007 (UTC)

Äch, Öch, Üch

Are Äch, Öch, and Üch pronounced /ɛx/, /œx/, /yx/; or /ɛç/, /œç/, /yç/?

The information in the article is correct: The allophone [x] occurs after back vowels and [a] (for instance in Bach [bax] 'brook'), the allophone [ç] after front vowels. Since /ɛː ɛ øː œ yː ʏ/ are all front vowels, the following ch is pronounced as [ç] in the standard. -- j. 'mach' wust | ‽ 22:39, 14 September 2005 (UTC)

Given the disagreement earlier about whether /a/ is [a] or [ɑ] and /aː/ as [aː] or [ɑː], does /x/ vary based on the quality of the vowel, or are /a/ and /aː/ always considered back for the purposes of this consonant? —Felix the Cassowary (ɑe hɪː jɐ) 06:31, 21 September 2005 (UTC)

It can only be followed by [x] (unless it is followed by the notorious diminutive suffix "no-umlaut-[çən]" which is not sensitive to the previous sound), so by this criterion, it is always to be considered back. -- j. 'mach' wust | ‽ 09:00, 21 September 2005 (UTC)

vɪçtɪç vs. vɪçtɪk

"Another common merger is the merger of /g/ at the end of a syllable with /ç/. In the case of the ending -ig, this pronunciation is prescribed by the Siebs standard, for instance wichtig [ˈvɪçtɪç]. The merger is found in Northern German. It occurs neither in Southern standard German nor in Southern German dialects."

It does occur in Franconian, which I do consider a Southern German dialect. If noone objects, I will change the article accordingly. Does anyone know whether it also happens in other southern german dialects? --Schuetzm 18:43, 20 September 2005 (UTC)

It does not in Alemannic German, nor does it in Austro-Bavarian. I thought Franconian was Middle German, so that's why I worded it that way, and I'm sorry if I've misrepresented your dialect. -- j. 'mach' wust | ‽ 19:49, 20 September 2005 (UTC)

Hmm, maybe the problem is the confusion of "Southern" vs. "Upper German". If there is a commonly accepted definition of "Southern German" that excludes Franconian, than you are of course right. It is definitely a part of "Upper German" (see the map at [1], area #25). --Schuetzm 14:40, 25 September 2005 (UTC)

I think there is no such term, so the best wording I see is Austro-Bavarian German and Alemannic German, and I hope I'm not mistaken again! ;) -- j. 'mach' wust | ‽ 17:30, 25 September 2005 (UTC)

G without a vowel following (Krieg, Tag) turns into "ch" in many parts of Germany (nearly everywhere except Southern Bavaria and Baden-Württemberg). So the two examples can be pronounced as either "Kriek"/"Tak" or "Kriech"/"Tach". In Standard pronounciation there is sort of an equalizing solution: g without a vowel following is usually pronounced as "k", but as "ch" in -ig.

x vs. χ

As far as I know as a native speaker, the 'ch' sound in 'ach', 'bach, 'lachen', etc is actualy a vl uvular fricative χ, rather than the velar fricative x. What gives? [oliver lenz]

Oliver is right. It's uvular, not velar (it's in the IPA Handbook). This article could really use some improvement with the help of a proper German phonology.

Peter ^Isotalo 08:20, 4 March 2006 (UTC)

Lemme revise that somewhat. According to Klaus Kohler in the IPA handbook, it's [x] in buch and hoch, but [χ] in Bach and doch.

Peter ^Isotalo 08:31, 4 March 2006 (UTC)

According to the research I can find (I'll add this to the article with its references), it's uvular after [a(:)], velar after [u:] and [o:], and freely varies between uvular and velar after [ʊ] and [ɔ]. Angr/talk 08:34, 4 March 2006 (UTC)

I don't understand why everybody is so keen about the uvular allophones of the velar fricative. What about the other uvular allophones [q ɢ ɴ]? I'm not happy at all about the revision that has mixed up the allophony of /ç/ and /x/ with the allophony of [χ]. I believe these allophonies are of a very different nature: The /ç - x/ is a well-known and undiscussable feature of any standard pronunciation. It is perceivable to native speakers and some even consider it to be a phonemic distinction. A pronunciation that does not respect it will be noticed not to conform to the standard. The [χ] allophony, however, is a variable feature of certain varieties. No untrained native speaker perceives it, and even the trained phoneticians argue about the extent to which it occurs (that's why I believe it varies if not regionally at least idiomatically). A pronunciation that does not respect the [χ] allophony will not be noticed at all. ― j. 'mach' wust | ✑ 09:24, 4 March 2006 (UTC)

Other uvular allophones of what? I didn't know German had two uvular stops as well as a nasal.

Peter ^Isotalo 00:20, 5 March 2006 (UTC)

German does not have any uvular phonemes at all. Velar phonems happen to have uvular allophones. ― j. 'mach' wust | ✑ 19:38, 8 March 2006 (UTC)

But even as allophones, does German have any uvular sounds at all besides [χ] and [ʀ ~ ʁ]? AFAIK, /k/ does not have an allophone *[q], nor does /g/ have an allophone *[ɢ], nor does /ŋ/ have an allophone [ɴ]. Angr/talk 19:46, 8 March 2006 (UTC)

Well, now it'd be up to me to find references... I'm afraid I won't find any. However, I still don't believe that only the velar fricative has an uvular allophone, but the other velar sounds not. I think the main reason that only the fricative allophone is talked about is that its IPA sign, [χ], is familiar to the transcription of German (it's sometimes used as a neutral sign that signifies [x] as well as [ç], but without uvular meaning, and in older transcription systems, it was used for IPA [x]), whereas the signs [q ɢ ɴ] are utterly awkward. I also think that the difference between [x] and [χ] is easier to perceive for phoneticists than the other differences between velar and uvular consonants. But again, I don't have references.

However, references that don't mention any uvular allophones at all are easy to find, for example the Aussprache-Duden, to which I'd give more credit than to the IPA handbook because the description of German pronunciation is its primary aim. So I'd think we'd better up not to mention any uvular allophonies, or at least not to mix them up with the [ç - x] allophony. ― j. 'mach' wust | ✑ 09:43, 10 March 2006 (UTC)

I think I trust trained phoneticians and phonologists more than lexicographers when it comes to this question. I have no doubt that if /k, g, ŋ/ did have allophones [q, ɢ, ɴ], the IPA handbook and Wiese's book and other German phonlogists and phoneticians would have mentioned them. I don't think it's helpful at all to encyclopedia-building to remove verifiable, sourced information. Angr/talk 10:12, 10 March 2006 (UTC)

I just think the current text obscures the basic ç-x allophony. And I think the mention of the [χ] should be labled "according to some", since not all sources agree on it. The Aussprache-Duden is certainly also a work by trained phoneticians and phonologists, and its introduction is one of the most thorough analysis of German pronunciation. ― j. 'mach' wust | ✑ 11:34, 10 March 2006 (UTC)

Pronouncing schr-

I find that it is hard (or at least very unwieldy) to pronounce the schr- cluster with a trilled Austro-Bavarian R. Is there any Austro-Bavarian workaround to this (like, sound changes)? Does the R "soften", is the SCH just pronunced as /s/, what?

Just curious.

I'd say that in initial consonant clusters with /r/, the /r/ is realized rather as a tapped [ɾ] than as a trilled [r] (if it is pronounced alveolarly at all), so the pronunciation is not [ʃr], but rather [ʃɾ]. At least that's how it is in Alemannic German, but I guess it's not different in Austro-Bavarian (nor in many other languages that have such clusters, such as Spanish or Italian). ― j. 'mach' wust | ✑ 09:23, 12 March 2006 (UTC)

I don't think either Spanish or Italian has either a [ʃr] cluster or a [ʃɾ] cluster. Angr/talk 12:17, 12 March 2006 (UTC)

Certainly not! I was thinking of the general consonant + r clusters. I'm sorry for my mistakable wording! ― j. 'mach' wust | ✑ 15:40, 12 March 2006 (UTC)

No, in Modern Spanish in fact we don't even have the sound [ʃ] any longer (except in Andalusian, where it has reappeared as deaffricated 'ch', unrelated to Mediaeval Spanish 'x'), and AFAIK in Italian [ʃ] is always followed by a vowel (scia, sce, sci, scio, sciu) . 213.37.6.65 00:06, 24 April 2007 (UTC)

Both Spanish and Italian pronounce (in the main dialect) the cluster "tr" as [tr] with a trilled r. So what are you trying to say? (Of course in rapid speech, it may sometimes degenerate into a tap.) −Woodstone 16:18, 12 March 2006 (UTC)

Then I was mistaken. I haven't been much in Italy, but when I was learning Spanish in Argentina, everbody laughed at me as long as I could only pronounce [tr] or [dr] etc. and not yet [tɾ] or [dɾ] etc. They seemingly did not only perceive it as an unusual pronunciation, but simply as plain wrong. Anyway, I don't believe much in main dialects, but rather in pluricentric languages. And again, I think phonology is a science much influenced by its own traditions, so it's good always to stay sceptical. ― j. 'mach' wust | ✑ 19:24, 12 March 2006 (UTC)

No way. The r in the clusters pr-, br-, fr-, tr-, dr-, cr-, gr- in Spanish is always pronounced with a tap, never with a trill, except in the verbs subrayar and subrogar (because in these cases the cluster is not syllable initial br-, but [b.r] with a syllable break in between, since morphologically these words are derivatives from rayar and rogar with the prefix sub-). Pronouncing a trill in the cluster tr in Spanish sounds "funny" to say the least, and is definitely non-standard. 213.37.6.65 00:06, 24 April 2007 (UTC)

Speakers of Spanish find it difficult to pronunce the cluster /sr/, e.g. in 'las rosas'. Even in dialects where /s/ is pronounced in all positions, this pre-rhotic /r/ is nearly always dropped. --129.35.231.16 13:43, 24 July 2006 (UTC)

I lived for some time near Bavaria and just tried out a typical Bavarian phrase "das schreibst du mir auf" [dœs ˈʃræːbstmɐ ʔaːf] and it seems that a tiny [ɘ] is inserted between sch and r. Needless to say that the problem does not occur with standard [ʃʀ]. Btw, exaggerated stage pronunciation is often parodied as [ʃɑˈraɪːbən]. (as you can hear from old records, [r] was stage standard in the 1920s) -- megA 10:38, 25 October 2006 (UTC)

No, what is dropped in the cluster [sr] in Spanish is the 's', never the 'r': las rosas [larːosas] or [larosas], never [lasːosas] (las sosas) nor [lasosas] (las osas). Final s is dropped or aspirated in all cases in many dialects, anyway. 213.37.6.65 00:06, 24 April 2007 (UTC)

Siebs = Bühnenaussprache?

My impression is that Bühnenaussprache is a more widely recognized term than "Siebs standard". Any objection to my substituting it?--Chris 19:21, 30 May 2006 (UTC)

I'd rather mention Siebs, since this explains that we're speaking of a very specific pronunciation guide. How about "Siebs' stage pronunciation"? ― j. 'mach' wust | ⚖ 21:02, 30 May 2006 (UTC)

Diphthongs

Could the 'io' in words like "Reklamation" and "Konfession" be considered a diphthong? To me, the 'ion' ending sounds like it is pronounced in one syllable. Redtitan 01:55, 17 July 2006 (UTC)

I think the usual explanation is that the /i/ in such words is nonsyllabic. It could be considered a post-consonantal allophone of /j/. User:Angr 05:59, 17 July 2006 (UTC)

It is not considered a diphthong because of its peripheral state: It occurs only in specific endings of loan words. ― j. 'mach' wust | ⚖ 19:37, 17 July 2006 (UTC)

Quick uninformed personal impression: -ion is -[ˈjoːn] (one syllable) in Germany and -[iˈoːn] (two syllables) in Austria. No idea about Switzerland. (Note that most Germans would feel compelled to dissolve the vowel cluster of the "Austrian" version by inserting [ʔ] – in other words, they couldn't even pronounce it.)

David Marjanović | david.marjanovic_at_gmx.at | 17:42 CET | 2006/11/11

They could. —The preceding unsigned comment was added by 89.52.129.239 (talk) 11:37, 1 March 2007 (UTC).

Really? "It" is "-[iˈoːn]". I'm not talking about Bavarians or people who have learnt French, but about those people who insert glottal stops into Luise and Naomi and Beamter and Asteroid. Does "Kon-fes-si-ohn" sound artificial to you? (That's how I pronounce it.) If so, if you tried to pronounce it that way nevertheless, would you put a glottal stop in front of the last syllable? David Marjanović 22:25, 1 May 2007 (UTC)

Varieties to be mentioned

The "ch" as Ich-Laut is often pronounced as [ɕ], which is a [ç] tending towards a [ʃ]. This is frequent in and around Berlin, in Hesse and in Cologne area. The speakers who use [ɕ] would also use [ʃ] for "ch" in the beginning of a word. So there are three possibilites to pronounce "China": [ki:na] [çi:na] [ʃi:na]. Aditionally, "pf" is pronounced in all Northern and Central Germany as "f". In colloquial language but also on TV or radio a huge lot of speakers would pronounced "Pferd" as [fɛət]. All this information is proven. You can check it and then please add it, because your article tends a bit to stress on Southern German varieties and specials. —The preceding unsigned comment was added by 89.52.129.239 (talk) 11:45, 1 March 2007 (UTC).

I don't think we should mention the [ɕ] because it's AFAIK nowhere considered standard. However, the [f] instead of [pf] phenomenon is so common that we probably should mention it. David Marjanović 10:27, 29 April 2007 (UTC)

The next few edits...

Hi everyone. I should start by mentioning that I'm not a trained linguist; I have learned the IPA and everything else I know about linguistics on my own, by reading (largely Wikipedia, though also some primary literature). I am a native speaker of the, um, to be precise, the younger city-dialect of Linz (Austria, 200 km west of Vienna) and of Austrian Standard German (as spoken on TV, most notably by newscasters). If you let me, I'll make many changes to the article, mostly because in many places it describes regional variation as allophonic or free variation (and then misses some regional variation), but because I don't have comprehensive knowledge of Standard German in Germany and am way underinformed on Standard German in Switzerland, I should probably discuss what I want to do here first.

I'd like to start with the ich-Laut/ach-Laut affair. That section currently states:

The diminutive suffix -chen is always pronounced with an ich-Laut [-çən]. Usually, this ending triggers umlaut (compare for instance Hund 'dog' to Hündchen ‘little dog’), so theoretically, it could only occur after front vowels. However, in some comparatively recent coinings, there is no longer an umlaut, for instance in the word Frauchen [ˈfra͡ʊçən] ‘female dog master’ (a diminutive of Frau ‘woman’), so that a back vowel is followed by [ç], even though normally it would be followed by a [x], as in rauchen [ˈra͡ʊxən] ‘to smoke’.

I wonder if this one word is actually derived from the version with the nickname suffix -/ɪ/ preceded by an /l/ that could be epenthetic or derived from the Bavarian-Austrian diminutive suffix -/l/~-/ɐl/. The nickname suffix does not trigger umlaut. I have no evidence, but the most parsimonious hypothesis I can come up with that this word was borrowed and the nonstandard/babyish suffix replaced, and then people forgot to add the umlaut. Apart from this, the umlaut triggering by -chen is still productive; I've come across words like Progrämmchen "neat little computer program", and they sound natural to me. I haven't come across any word in -chen other than Frauchen that doesn't have umlaut when it could.

There is even a minimal pair for [ç] and [x] due to this effect: [kuːçən] Kuhchen ‘little cow’ vs. [kuːxən] Kuchen ‘cake’.

WTF. This is bewildering. I don't even know where to begin.

• I've never found anyone saying or writing Kuhchen. It sounds wrong. I find it outright difficult to articulate. If anything, it's Kühchen with [yː].
• The example is contrived. Nobody I've encountered would say Kühchen. It's Kalb "calf" or Kälbchen.
• The practice of writing [ən] instead of [m̩], [n̩], [ŋ̩], which AFAIK comes from the "broad phonetic transcription" used by the Duden, may be justified phonologically – at least in the northern varieties that have such a thing as a reduced vowel –, but phonetically it is a hypercorrectivism. Syllabic morpheme-final nasals are standard (with variation within the standard: up north they can even occur behind nasals or fuse to them; down south [ɛn] is used behind nasals). Perhaps due to the absence of the palatal nasal ([ɲ]) in German, however, syllabic nasals do not occur behind [ç]. So even if anyone said *Kuhchen, it would stay distinct from Kuchen: [ˈkuːçɛn] (or -[ən] if you must) vs [ˈkuːxŋ̩].

I'm not aware of another possible minimal pair between [x] and [ç], so I submit there is none.

An allophonic distribution of [ç] after front vowels and [x] after other vowels is a common one, and can be heard also in Scots, in the pronunciation of light. However, it is by no means inevitable: Dutch, many Southern German dialects, as well as Yiddish, which comes from one of them, retain [x] in all positions.

• Yiddish is famous for its uvular fricative (presumably derived from something Alemannic, or from Hebrew, or both).
• "Southern German dialects"? The Alemannic ones have, like Yiddish, [χ] in all positions with no allophony I can perceive. The Bavarian-Austrian ones have the "usual" allophony range from [x] to [ç].
• At least some varieties of Dutch also have [χ] (this is e. g. what happens to Flemish word-initial g). I don't know how widespread this is, however.

In short… there are languages that have [x] but not [ç] (the Slavic ones, Mandarin…), but are you sure there is a West Germanic one?

According to certain analysis, the German ach-Laut is further differentiated into two allophones, [x] and [χ]. Some say that [x] occurs after /uː oː/ (for instance in Buch [buːx] ‘book’) and [χ] after /ʊ ɔ a aː a͡ʊ/ (for instance in Bach [baχ] ‘brook’), others say that [x] occurs after /uː oː ʊ ɔ a͡ʊ/ and [χ] after /a aː/.

This is not a question of analysis; all those analyses are correct as far as I can tell, having heard people who talk like that. It's regional variation. It would just be nice if someone could find out which parts of central and/or northern Germany use [χ] and in which conditions. David Marjanović 11:29, 29 April 2007 (UTC)

Forgot something:

(for instance in Furcht [fʊrçt] ‘fear’)

That (or rather, [fʊ̞ːçt]) is a northern thing. Austria at least: [fʊɐ̯xt], [dʊɐ̯x] and so on. After all, there is no consonant in there (anymore). David Marjanović

Vowels

Another difficult one.

Short [i y u e ø o] occur only in unstressed syllables of loanwords, for instance in Psychometrie [psyçomeˈtriː] 'psychometry'.

Are there really people who pronounce that with [y]? I've never heard any psycho- word with [y], always with [ʏ]. The o oscillates freely between [o] and [ɔ] and probably ends up somewhere in between most of the time – this is the least stressed syllable in the word, so the distinction is least stable here. I agree on the [e], however.

The schwa [ə] occurs

There is much variation right here.

• In Austria, it does not occur. When it can, it disappears and leaves a syllabic nasal consonant or /l/ behind (see above), and when it can't disappear, it is [ɛ]. That's right: Hände has twice the exact same vowel. There is no schwa. [ə] is considered a paralinguistic effect, not speaking.
• Up north, reduced vowels are common; however, what results is at least not always [ə]. I haven't listened closely to enough Germans to tell, but those I have listened to produce exotic vowels that must be [ɘ] and [ɵ] (the latter is more common in my "sample"). I don't know if anyone uses an actual [ə].

The long open-mid front unrounded vowel [ɛː] is merged with the close-mid front unrounded vowel [eː] in many varieties of standard German, so that, for example, Ähre [ɛːrə] 'ear' (of wheat, etc.) and Ehre [eːrə] 'honour' are homophones for many speakers.

• We should make clear that when they merge, the result is [eː]. So maybe "is merged into" instead of "is merged with" would be better.
• Where except in Austria does the merger happen?
• The Duden's "broad phonetic transcription" notwithstanding, all speakers I have heard who do not merge have a wide-open [æː] that is not terribly similar to the {{[ɛː]}} that occurs in, say, French or in my dialect (where vowel length is apparently not phonemic, but a wide range of phonetic lengths occurs).
• This is marginal and trivial, but the merging varieties retain [æː] (usually not quite that open) as the name of the letter Ä. Maybe this should be mentioned, too.
• One thing about the examples. There are places in Germany where [ɐ] does not occur behind long vowels, making the examples correct. Elsewhere, however, this is unthinkable (making learning French, which has words like libération with [eʀ], even more difficult).

The open vowels [a] and [aː] are free allophones together with [ɑ] and [ɑː], respectively.

This, too, is regionally different. In Austria /a/ and /aː/ have very little or no difference in quality, both being front [a]. Elsewhere, some people have [a] vs [äː], while others have [ä] vs [aː]. (The two dots are supposed to be the IPA "centralized" diacritic. By comparison with Mandarin – central – and English – back –, I don't think the back vowel [ɑ(ː)] occurs in any variety of Standard German, though, again, my sample size is small.)

I think we should also add a footnote about /ɪ/. It is [ɪ] (the English sound) in some parts of Germany, but there it tends to display at least allophony all the way to what must be [ʉ] (in front of at least /m/, /l/, and /r/). On the other hand, in Austria it is very close to [i], so I'd suggest explaining that variant as [ɪ̽] with the "mid-centralized" diacritic. David Marjanović 14:19, 29 April 2007 (UTC)

Diphthongs

The German diphthongs are /a͡ɪ a͡ʊ ɔ͡ʏ/, for instance in Ei /a͡ɪ/ 'egg', Sau /za͡ʊ/ 'sow', neu /nɔ͡ʏ/ 'new'. Occasionally, these are transcribed as /a͡e a͡o ɔ͡ø/. Instead of the transcription /ɔ͡ʏ/, the transcription /ɔ͡ɪ/ is used as well.

As far as I know, all this is regional variation (the "occasional" variants may be limited to Siebs, however). Importantly, we should mention that in Austria ei is [ɛ͡ɪ] (and eu is [/ɔ͡ɪ/]).

Homepage [ˈhɔʊ̯mˌpʰɛɪ̯d͡ʒ]

Well, describing the first diphthong as /ɔʊ̯/ may make sense in English, but less so in German where the sound it begins with is much closer to /o/ than to /ɔ/. Besides, you'll be hard-pressed to find someone who keeps that d͡ʒ] voiced at all, let alone word-finally. David Marjanović 14:19, 29 April 2007 (UTC)

Consonants

In the northern varieties, [ʔ] occurs before word stems with initial vowel. It is often not considered a phoneme, but an optional boundary mark of word stems.

I'm not sure. It is inserted into each and every vowel cluster – most of which have a morpheme boundary running through them – and used utterance-initially, but not always word-initially. As an example, I recently watched a German TV documentary on asteroids and comets = Asteroiden und Kometen = [ˌʔastɐʀoˈʔˑiːdn̩ʊndkoˈmeːtn̩]. I'd have dropped the second [ʔ], which makes me hear *Astero – Iden, and likely the first one, too (because the occurrence was in the middle of a sentence), but note there was no [ʔ] in front of und. I think I've read somewhere that some analyze [ʔ] as occurring in front of stressed vowel-initial syllables "in German".

[d͡ʒ] and [ʒ] occur only in words of foreign origin. In certain varieties, they are replaced by [t͡ʃ] and [ʃ] altogether.

Is this really regional variation, or does this depend more on how well people can produce these sounds and bother doing so while speaking German? – Whether the intermediate realization, voiceless lenes, occurs may be a regional phenomenon.

[r], [ʁ] and [ʀ] are free allophones of each other.

As mentioned way above, I object to calling them allophones because, AFAIK, nobody uses two or three of these sounds (at least while speaking Standard German), let alone with rules governing when to use which. That said, I haven't found anyone yet who says [ʁ] instead of [ʀ]. (The fricative does occur in French, but even there few people use it, AFAIK – I live in Paris and haven't found anyone here who uses it.)

Englisch /ɛnɡlɪʃ/ [ɛŋlɪʃ]

The /g/ is not deleted in Austria (at least), probably because it's voiceless there. Interestingly, someone introduced this variation into the next example:

Ganges /ɡanɡəs/ [ɡaŋəs] ~ /ɡanɡɛs/ [ɡaŋɡɛs]

[g̥aŋˑɛs] in Austria, identical to the native word Ganges, genitive of Gang "gait".

[ʋ] is occasionally considered to be an allophone of [v], especially in Southern varieties of German.

It's more restricted than that. I never encountered that sound before the BAWAG /ˈbaːvag/ scandal last year, which some (not all) newscasters pronounced using an odd, vaguely [ɔ]-like consonant. I suppose the open vowels are to blame; apart from that, I've only met with [v] in German, the same as in English or French or Russian, no matter in what environment. – To my surprise, however, I once came across a Linguistlist post claiming large parts of Germany use [β], explaining why someone else thought a German friend of his always seemed to exchange /w/ and /v/ when speaking English. Is there any truth in this?

The voiceless stops /t/, /p/, /k/ are aspirated except when preceded by a sibilant.

And except word-finally, and except in Austria.

We should also mention that /t/ and /d/ are apical in northern Germany (like in English) but laminal elsewhere (like in French, Spanish, Russian, Mandarin…).

The obstruents /b d ɡ z ʒ/ are voiceless [b̥ d̥ ɡ̊ z̥ ʒ̊] in the Southern varieties.

Well, yes, but… original research warning… I deny the existence of /z/ in the first place. Instead, what's IMHO going on is a length contrast. Here goes:

• In Austria at least, fortes behind short vowels (usually orthographically doubled) are a bit longer than fortes behind long vowels. Just a bit – not comparable to the long consonants of Swiss German, Italian, or Finnish; I'll use the IPA half-long diacritic –, but still, the difference is there. In the Bavarian-Austrian dialects (not in the standard language!), this difference alone can be phonemic: vowel length is not phonemic, and /ɔ/ has merged into /o/, yet Ofen [ˈofm̩] "oven" and offen [ˈofˑm̩] "open" stay distinct.
• This is the difference between s and ss (new orthography). That's why the difference does not exist word-initially; the voicing varieties voice every word-initial s, often extending this to English words that are supposed to start with [s].
• Unlike the other consonants, /s/ can occur in both lengths behind both long vowels, producing minimal pairs like reisen "travel" vs reißen "rip". Despite the orthography, this contrast is neutralized word-finally: Eis and Fleiß rhyme, even though eisig and fleißig don't; this does not happen to most fortis-lenis contrasts in Austria.
• I lied. There is another consonant that can occur in both lengths behind long vowels: /x/. This fact is completely ignored by the orthography (maybe just because chch would look too clumsy – schsch doesn't exist either). I haven't found a minimal pair (justifying the orthography), but I've found several pairs that don't rhyme. Unfortunately I have no idea how widespread this phenomenon is outside of Austria, but importantly, the same distinction exists in my dialect with overlapping but not quite identical sets of words, making it unlikely that the distinction in the standard language was invented in Austria. Is anything published on this?

Short /x/:
weich "soft"
weichen "retreat"
tauchen "dive"
Tauchvogel "diving bird"
reichen "reach"
Longer /xˑ/:
Teich "pond"
Eichen "oaks"
Leichen "corpses"
Buch "book"
Buchen "beeches"
suchen "look for/seek/search" (dialect: short /x/)
Schlauch "hose"
Rauch "smoke (noun)"
rauchen "smoke (verb)"
Zeichen "sign (noun)"
zeichnen "draw"
reich "rich"
Reich "empire"

• On the matter of /x/, there are dialects in northern Germany that possess a [ɣ], though no variety of Standard German seems to have that. Does anyone know if this sound corresponds to the length distinction I have found? David Marjanović 14:19, 29 April 2007 (UTC)

Fortis & lenis

The aspiration is […] weakest in the syllable coda (such as in Saat [zɑːtʰ])."

Are there really people who aspirate word-finally? More importantly, are there any who aspirate word-final /t/? Is that a Swiss thing? All speakers I can remember merge word-final /t/ into /d/ – it becomes a lenis, unless it's part of a consonant cluster (including /rt/).

The nature of the phonetic difference between the voiceless lenis consonants and the similarly voiceless fortis consonants is controversial. It is generally described as a difference in articulatory force, and occasionally as a difference in articulatory length; for the most part, it is assumed that one of these characteristics implies the other.

In my limited experience that's yet another regional difference. The Swiss "fortes" are simply longer than lenes (much longer, as in Italian or Finnish). The Austrian ones are not longer (at least not consistently, and only marginally longer if at all). It's not the Korean thing either (described as stiff voice, sounds very weird). I gather it must be the loudness of the release, produced by increased pulmonic air pressure. Unfortunately I haven't been able to find any publication about Austrian consonants, though Google is all I have, so if you know of one, please help me out!

In most varieties of German, the opposition between fortis and lenis is neutralized in the syllable coda, due to terminal devoicing (Auslautverhärtung). A few southern varieties of German, such as Swiss German, present an exception to this.

Yes, but not because they don't devoice. Their lenes are already voiceless. The length contrast stays.

As far as I can tell, there's no length contrast in Austria (see above). What happens? Instead of /d/ becoming /t/ word-finally, /t/ (the short version, the one behind long vowels) becomes /d/ except in consonant clusters. This happens in Germany, too, as far as I've paid attention. Word-final /k/ and /p/ are rare, but they don't become lenes: Rad "wheel" and Rat "council/counsel" are homophones, ending in the consonant that is in the middle of Räder "wheels" but not that of Räte "councils"; Hub "lift (noun)" and hup "honk (imperative)" are not homophones; and if I take Tag "day" and substitute the /g/ by /k/, I hear myself speaking Russian (так "so, thus, this way").

However there are southern varieties which differentiate between a fortis /f/ […] and a lenis /f/

That's a Swiss (or Alemannic?) thing, and it can happen because this contrast is a length contrast. It would be difficult or impossible otherwise, and does not occur in Austrian Standard German nor AFAIK any Bavarian-Austrian dialects. David Marjanović 17:52, 29 April 2007 (UTC)

"Kuhchen"

Hi, I'm from Germany and my english isn't so good, so please excuse me :) In the "Ich-Laut and ach-Laut"-Category you wrote "There is even a minimal pair for [ç] and [x] due to this effect: [kuːçən] Kuhchen ‘little cow’ vs. [kuːxən] Kuchen ‘cake’." This is false! A little cow is "Kalb" or "Kälbchen", at most "Kuhlein" (noone would say that), but never "Kuhchen". Sorry again for the bad english, I hope you understand =) —The preceding unsigned comment was added by 88.68.126.60 (talk) 18:41, 1 May 2007 (UTC).

Splitting "German Phonology" like "English Phonology"?

Since some people seem to be a bit emotional if it comes to "pronounciation" and "standard" we could split this article like it's English counterpart does: A general intoduction with links to Recieved Pronounciation, American and Australian Standard... Objections? Qubux 17:30, 15 May 2007 (UTC)

Vowel Chart

It would be great if someone could find a vowel chart of German vowels so that one could compare German vowel postiton with the Cardinals.

geröntgt pronunciation

what the heck is the proper pronunciation of geröntgt(past participle of röntgen, to x-ray)? It seems impossible to say without losing a sound or adding one. Seems like since it was based on a name, it was sort of shoehorned into the German morphology system.--Shadowdrak 08:11, 30 May 2007 (UTC)

[gəˈʁœntkt] is the standard pronunciation. You think that's bad, try some of the tongue-twisters at Consonant cluster. —Angr 18:17, 30 May 2007 (UTC)

Phonology vs. Phonetics

The organization of the consonant section seems pretty confusing to me, as it stands. This is supposed to be an article on German phonology, not phonetics, but the consonant chart here doesn't simply list phonemes: it lists tons of allophones and putative allophones. Especially with cases like [ʋ] this seems backwards to me. Right now it's listed on the chart with the note "[ʋ] is occasionally considered to be an allophone of [v], especially in Southern varieties of German." To me, this seems to be saying that [ʋ] is a particularly common phone in German that some people argue is an allophone of /v/. If what I know of German is right, though, [ʋ] is just an occasional allophone of /v/. (Similar comments apply to listing [χ].)

Wouldn't it be easier to list only the phonemes /v/ and /x/ and include notes saying that they have [ʋ] and [χ] as allophones? Masily box 15:05, 9 June 2007 (UTC)