Help talk:IPA/Standard German/Archive 1
![]() | This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Archive 1 | Archive 2 |
Corrections
This goes mainly to Angr: I'm not too happy with all of your reverts/changes, I have a few remarks/questions:
- Why are tie marks on affricates unnecessary? Without them, they aren't affricates but simple combinations of two sounds. There's an audible difference between [pf] and [p͡f].
- I'm not discussing any allophone status, I mentioned that many (if not most) speakers of German pronounce /x/ as [χ], I'd even argue that [χ] is the standard variant in German. Why do you think that discussing the allophone status of /ʌ/ (as [ɔ]) is worth being included on the page, then? Furthermore, the statement "Often replaced by [eː]" is actually the same thing.
- Where's the difference between marginal vowels and non-native ones anyway? I'd opt for deletion of [ɔː] and [ʌ] (see my previous point).
- The standard pronunciation of "aktuell" is [ˌʔaktuˈʔɛl], there's no such thing as [u̯] in this word or in German in general. As for "Studie", I'm not so sure, but I doubt the existence of [i̯] there as well. The Aussprache-Duden is not particularly precise or thoroughly, unfortunately...
- If nonsyllabic marks are necessary (I agree with you there), affricate tie marks are as well, because it's more or less the same sort of sign (indicating that two characters belong to one phoneme, i.e. diphthongs or affricates), see my first point on this.
- I'd vote for exclusion of foreign sounds like [θ] or [ɹ]. They're too arbitrary, one would have to include other foreign phonemes like Spanish [ɲ], Russian [r], French [ɥ] etc. as well.
— N-true (talk) 00:07, 8 August 2008 (UTC)
- Please keep in mind that this is supposed to be a pronunciation guide in various non-linguistic articles for the general readers. The difference between [pf] and [p͡f] or the presence of [ʔ] before a vowel are completely trivial issues in this context. The point is to instruct people who don't know German how a German word is a pronounced, not to dabble in detailed phonetics.
- Peter Isotalo 12:32, 20 April 2009 (UTC)
Suggestion
You know what would be really neat? a pronouncer thingy that would say the word properly with all the weird pronunciation marks. —Preceding unsigned comment added by Ggb667 (talk • contribs) 14:33, 6 October 2009 (UTC)
Adding English Approximations for Vowels
The consonants section has english approximations, but the vowels section does not. I would like to volunteer. Is there a specific reason the vowels section hasn't got English approximations?
Cantorman (talk) 19:19, 6 October 2009 (UTC)
allophones of /r/
I've noticed a lot of pages use [ʀ] or even [r] in their transcriptions of German words (e.g. Saarbrücken). Indeed German_phonology#Consonants notes that "[r], [ʁ] and [ʀ] are in free variation with one another. [r] is used mainly in Southern varieties." Would it be appropriate to add a footnote to this effect, under the entry for [ʁ]? Otherwise, people are going to see these symbols and then follow the link to a page which fails to mention them as phones of German. Lfh (talk) 11:15, 18 October 2009 (UTC)
- That would be good, but it would also be good to correct the uses of [ʀ] and [r] to [ʁ] as we encounter them. +Angr 06:14, 19 October 2009 (UTC)
- Including placenames in regions where [ʀ] or [r] predominate? Lfh (talk) 10:01, 19 October 2009 (UTC)
- I'd say so. We use a standardized transcription for English that ignores regional variation, so we should do the same for German, and for the same reason: even people from the North pronounce place names from the South, and vice versa. Both [ʁ]-users and [r]-users have the same initial consonant in both Rügen and Rosenheim, so it doesn't make sense for us to transcribe them differently. +Angr 18:56, 19 October 2009 (UTC)
- OK. That makes good sense for names within Germany. It was Bern that got me wondering, because we transcribe [ˈbɛrn] and that is also the value of the sound recording. I considered changing it to show the vocalisation of /r/ but realised this might be considered inappropriate for a Swiss name, and I'm still not sure. Lfh (talk) 10:23, 20 October 2009 (UTC)
- And I just noticed that even the main German language article was describing [ʀ] as the most common rhotic, which I've amended. Lfh (talk) 13:17, 20 October 2009 (UTC)
- I think for Bern we can put /ˈbɛʁn/; after all, not all German speakers vocalize r after short vowels, and the article already gives the local Bernese pronunciation separately. +Angr 13:27, 20 October 2009 (UTC)
- Even though no Swiss person would ever say /ˈbɛʁn/ neither when speaking their own language nor when speaking German.157.157.101.224 (talk) 11:43, 8 January 2010 (UTC)
- I think for Bern we can put /ˈbɛʁn/; after all, not all German speakers vocalize r after short vowels, and the article already gives the local Bernese pronunciation separately. +Angr 13:27, 20 October 2009 (UTC)
- I'd say so. We use a standardized transcription for English that ignores regional variation, so we should do the same for German, and for the same reason: even people from the North pronounce place names from the South, and vice versa. Both [ʁ]-users and [r]-users have the same initial consonant in both Rügen and Rosenheim, so it doesn't make sense for us to transcribe them differently. +Angr 18:56, 19 October 2009 (UTC)
- Including placenames in regions where [ʀ] or [r] predominate? Lfh (talk) 10:01, 19 October 2009 (UTC)
- The same is happening with French - a lot of [ʀ] and [r] being used - again Bern is an example, in fact. Lfh (talk) 13:02, 14 November 2009 (UTC)
A quick question
I know that this may not be the most appropriate place to ask this question, but I don't speak German, and I don't know where else to ask. What is the correct pronunciation for the letter "ß"? —Preceding unsigned comment added by Mossman fmcb (talk • contribs) 06:17, 19 December 2009 (UTC)
- Have you checked out ß? — Ƶ§œš¹ [aɪm ˈfɻɛ̃ⁿdˡi] 09:44, 19 December 2009 (UTC)
- Quick answer: it's pronounced /s/, like the s in the English word sit. +Angr 11:10, 19 December 2009 (UTC)
- I'm sorry. I just now realized that. It's a ligature, but I thought it had some kind of different pronunciation. My mistake. —Preceding unsigned comment added by Mossman fmcb (talk • contribs) 05:43, 2 January 2010 (UTC)
- Quick answer: it's pronounced /s/, like the s in the English word sit. +Angr 11:10, 19 December 2009 (UTC)
English [a]
Do we think it's a good idea to give an English approx. for [a]? "Bra", or "cat", or "cut"? What about [ɔʏ], which is not a million miles from [ɔɪ]? Lfh (talk) 10:24, 7 January 2010 (UTC)
- Usually, the vowel of father is used as a close approximation to [a]. Yes, the vowel of boy should be a good enough approximation to [ɔʏ]. — Ƶ§œš¹ [aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi] 19:50, 7 January 2010 (UTC)
R symbol
I see that the pronunciation key table has an upside-down R in it that correlate to the French R. However, I came upon this page from clicking the phonetic spelling of the author's name ([ˈhɛʀman ˈhɛsə]) at the top of the Herman Hesse article, and in that phonetic transcription, there is a miniature capital "R" (ʀ) that does not correlate to any of the symbols on the table. I assume that it's pronounced the same as an English R in the same context, however, even the IPA characters that are pronounced the same as in English are shown in the table, so why not "ʀ"?--71.104.225.218 (talk) 06:02, 23 February 2010 (UTC)
- There's a note on the page that explains that there's some variation in Wikipedia articles on the rhotic (or r-like sound) of German. In the past few days, an editor has been incorporating links to pages specific to sounds in these pronunciation guides and I suspect that this particular guide will soon follow.
- The more common pronunciation of the German rhotic is a uvular trill (small capital r) or a voiced uvular fricative (upside down capital R). I imagine we use the upside down one because of how linguists tend to transcribe this sound but maybe others can weigh in on the matter. — Ƶ§œš¹ [aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi] 09:21, 23 February 2010 (UTC)
- (edit conflict) The upside-down capital R, [ʁ], is the voiced uvular fricative, which is the pronunciation of "r" generally taught to learners of modern German, in my experience. The small capital R, [ʀ], is the uvular trill, a different sound that is generally found in more old-fashioned accents. Neither of them is the English "r" - follow these links to here sound samples.
- In theory we're always supposed to use [ʁ] on Wikipedia, but [ʀ] is so common that I added a footnote about it to clear up any confusion, but it displays in small text and you seem to have missed it. Some people also write [r], which technically represents the alveolar trill (as heard in Italian). How any given speaker of German actually pronounces "r" depends on where they come from.
- I don't know whether we should (a) try to remove all the [ʀ]'s or (b) just make it clearer that more than one symbol is in use. Lfh (talk) 09:24, 23 February 2010 (UTC)
- I think we should do both (a) and (b). We need to do (b) because the task of doing (a) will probably never be done. +Angr 11:16, 23 February 2010 (UTC)
- OK. Tbh I've been in two minds until now about whether to leave or change [ʀ]'s in articles. Lfh (talk) 14:43, 23 February 2010 (UTC)
- Back on this topic - I note that Uvular trill does not include German in its (short) list of occurences, even though in the main text it is heavily implied that [ʀ] is still in use. Can anyone provide a decent source so we can add German to that article? Lfh (talk) 20:11, 10 April 2010 (UTC)
- OK. Tbh I've been in two minds until now about whether to leave or change [ʀ]'s in articles. Lfh (talk) 14:43, 23 February 2010 (UTC)
- I think we should do both (a) and (b). We need to do (b) because the task of doing (a) will probably never be done. +Angr 11:16, 23 February 2010 (UTC)
Vocalisation of coda /r/
Would this be a good time to ask about when we should transcribe vocalised post-vocalic /r/'s (as [ɐ̯], see e.g. Berlin) and when we should leave them as [ʁ]? Lfh (talk) 14:43, 23 February 2010 (UTC)
- In my opinion we should use the actual pronunciation and use [ɐ̯] whenever the word is pronounced that way in standard German. That's why I changed the pronunciation of Hermann Hesse some hours ago. An /r/ is practically always vocalized in coda position (after vowels at the end of a syllable). Pronunciations with [ʁ] are possible, but non-standard (dialectal or hypercorrect). — N-true (talk) 16:21, 23 February 2010 (UTC)
- Duden's pronunciation guide uses [ɐ̯] after long vowels and [r] (i.e. [ʁ]) after short vowels. While many people do use [ɐ̯] after both kinds of vowels, it's neither dialectal nor hypercorrect to use [ʁ] after short vowels. I think most newsreaders, for example, do (hence the parody of newsreaders pronouncing Sport as "Spocht"). +Angr 17:17, 23 February 2010 (UTC)
- Collins seems to do the same thing, except without the [ ̯ ]. So, am I to understand that Kirsch should be [ˈkɪʁʃ], Pferd [ˈpfeːɐ̯t] and Donner [ˈdɔnɐ]? Lfh (talk) 18:01, 23 February 2010 (UTC)
- Do we need the semivowel diacritic? — Ƶ§œš¹ [aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi] 01:54, 24 February 2010 (UTC)
- Of course, 'cause it's phonemical in German and because otherwise you don't see how many syllables there are. There are minimal pairs like sehr [zeːɐ̯] (very) vs. Seher [zeːɐ] (seer). I don't see a reason to arbitrarilly leave out important distinctions. — N-true (talk) 02:04, 24 February 2010 (UTC)
- Here's a test case: Fürth. It's been given as [fyrt]. Should this be
/fy:rt/ -> [fy:ɐ̯t], or
/fʏrt/ -> [fʏʁt]? Lfh (talk) 15:22, 30 March 2010 (UTC)- I don't know if it's a long ü or a short one. I've asked at de:Diskussion:Fürth#Aussprache. +Angr 15:50, 30 March 2010 (UTC)
- I've always heard "Fürth" with a short ü, and the Duden gives short ü as well, so it should be [fʏɐ̯t]. I despise that artificial "newsreader German" [fʏʁt], because it sounds very stilted. BTW, I always thought the Spocht parody was a parody on people from western Germany, like the Eifel, who always pronounce /r/ before /t/ as [χ], could be both, though. — N-true (talk) 21:44, 30 March 2010 (UTC)
- I don't know if it's a long ü or a short one. I've asked at de:Diskussion:Fürth#Aussprache. +Angr 15:50, 30 March 2010 (UTC)
- Here's a test case: Fürth. It's been given as [fyrt]. Should this be
- Of course, 'cause it's phonemical in German and because otherwise you don't see how many syllables there are. There are minimal pairs like sehr [zeːɐ̯] (very) vs. Seher [zeːɐ] (seer). I don't see a reason to arbitrarilly leave out important distinctions. — N-true (talk) 02:04, 24 February 2010 (UTC)
- Do we need the semivowel diacritic? — Ƶ§œš¹ [aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi] 01:54, 24 February 2010 (UTC)
- Collins seems to do the same thing, except without the [ ̯ ]. So, am I to understand that Kirsch should be [ˈkɪʁʃ], Pferd [ˈpfeːɐ̯t] and Donner [ˈdɔnɐ]? Lfh (talk) 18:01, 23 February 2010 (UTC)
- Duden's pronunciation guide uses [ɐ̯] after long vowels and [r] (i.e. [ʁ]) after short vowels. While many people do use [ɐ̯] after both kinds of vowels, it's neither dialectal nor hypercorrect to use [ʁ] after short vowels. I think most newsreaders, for example, do (hence the parody of newsreaders pronouncing Sport as "Spocht"). +Angr 17:17, 23 February 2010 (UTC)
examples for ö
I'm not sure that "mull" and "but", for [øː] and [œ], work well for most Anglophones. In England we often approximate both those German vowels with /ɜː/, the non-rhotic vowel of "nurse". This doesn't work so well for most North Americans, because they always have /r/ after that vowel (AFAIK); and I know that in some US accents the short u of "but" is like /ɜː/. But it isn't for most of us. WP:IPA for French and WP:IPA for Hungarian both use "nurse" words for the [ø] vowel, and I think we should do the same here, but make it clear that there is no /r/ sound. Lfh (talk) 09:38, 3 April 2010 (UTC)
- I agree that the approximations are rough, but better than nothing as was before. I added the first ones that came to mind. Feel free to improve. −Woodstone (talk) 14:06, 3 April 2010 (UTC)
- I am a native German-speaker, and I thought your examples were just fine. A quotation from Arnold Schwarzenegger (former Austrian German) comes to mind: "... if they don't have the guts, I call them girlie men." At the 2004 Republican National Convention, he implored the audience not to be pessimistic about the economy, saying "Don't be economic girlie men."
- BTW, it was I who added Hafen to the [ɑ:] sound, just to show that other words besides those with the trailing h such as in Ahnung can have the long /a/ in German. However, I don't dispute that for the purpose of an example, Bahn will do just fine by itself. Skol fir (talk) 20:42, 3 April 2010 (UTC)
I think personally they are very weak example to . But is certainly more like German 'hat' than anything with ü- I think actually that many English words have have an 'u' similar to ü such as 'surprise' 'murder', 'curling' which are surely much more similar to ü than 'but'. I agree that 'ö' is quite hard to describe but I think they are basically like above words just with a further opened sound than ü.82.83.99.169 (talk) 14:55, 3 April 2010 (UTC)
- The best example I can think of for the [yː] sound in German is chop suey, where the "u" is actually pronounced very close to the German [ü]. I think this "-oo-" sound (as in loot) is just more rounded in the German U with umlaut. The shorter [ʏ] represented by the German word füllt could be approximated with the English word murmur, again just slightly more rounded in the German version. That combination [ur] was already suggested above. It's not easy being [ü]. Skol fir (talk) 21:27, 3 April 2010 (UTC)
- I've changed the examples to hurt and hurl and removed the note about removing the r. Since rhotic dialects pronounce the vowel of nurse monophthongally (that is, there is no part of the vowel that doesn't have the r), there's no way native speakers will be able to break it up. — Ƶ§œš¹ [aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi] 23:53, 3 April 2010 (UTC)
- As a native German-speaker, I have to disagree with your examples for the [ø:] and [œ] sounds, as what you have given more closely resembles the [ʏ] sound. Therefore, I have reinserted the word girl for the [ø:] sound as I know that it is closer to the desired sound for that vowel. I employed a well-known city of Göttingen with its Wikilink for the [œ] sound. The famous city of Lübeck stands in well for the [yː] sound, while fur is the closest approximation for the [ʏ] sound. Before you had inserted hurt and hurl—which for me are the same sound anyway—I had already commented on how the [ur] in murmur was best suited for the [ʏ] sound, not the [ø:] or the [œ]. I am confident that these changes will hold up to scrutiny, considering my background as a native German-speaker. Also, please note that Lfh had originally inserted girl on his own accord.
- Skol fir (talk) 02:59, 4 April 2010 (UTC)
- I found a very useful site for the pronunciation of German words, from the Universities of Portsmouth/Exeter, which not only is beautifully designed, but also extremely informative. Here is the address:
- A Guide to German Pronunciation. It explains when to use long or short vowels, and gives clear audio examples of the different sounds of consonants and vowels. It also describes the speech process that goes into making the correct sound.
- Skol fir (talk) 04:15, 4 April 2010 (UTC)
- As a native English speaker, I can tell you that murmur, fur, girl, hurl, and hurt all have the same vowel. The distinction between hurt and hurl that I was thinking when I made my recent edit is that the vowel of the latter is slightly longer.
- Using Lübeck for y is tautologically unhelpful. The closest equivalent to that vowel is the vowel of cute or few. — Ƶ§œš¹ [aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi] 08:04, 4 April 2010 (UTC)
- This discussion just supports my long-held opinion that the "Closest English equivalent" columns on these pages are useless or even harmful. There is no English equivalent to the ü and ö sounds, and putting in things that might sound vaguely like them in certain dialects just makes the page more confusing rather than less. +Angr 08:55, 6 April 2010 (UTC)
- They're not always useless. There's nothing wrong with telling people that [j] is an English y, not an English j, and that [dʒ] is an English j, not an English d followed by a bit of secret code. But I agree that they are sometimes a bad idea - that's why I didn't include the ü and ö sounds when I started the Vowel Equivalents column. We just need to be judicious and decide when to include English equivalents and when not. Perhaps the solution is to write something like "No close equivalent" whenever there isn't one, so that both first-time visitors and regular contributors know why the box is being kept empty. Lfh (talk) 16:28, 6 April 2010 (UTC)
- That is unsatisfactory for some people. Couldn't we say "somewhat like" for the sounds that don't have an English equivalent? — Ƶ§œš¹ [aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi] 23:49, 6 April 2010 (UTC)
- I guess so. On some pages, we do already. On others, we give examples from the most familiar foreign languages, mostly German, French and Spanish. (Obviously it would be tautological to do so on their own pages.) I think the important thing is to distinguish between sounds that have a close equivalent, e.g. [ɑ], [i], and sounds that don't, e.g. [ʏ, œ], and treat them differently in a consistent way, at least within a given page - be it "no close equivalent", or "somewhat like x", or "like French x". Perhaps now everyone should start going through these pages, and make these judgements where necessary. Of course many of them seem good already, e.g. Danish. Lfh (talk) 10:28, 7 April 2010 (UTC)
- That is unsatisfactory for some people. Couldn't we say "somewhat like" for the sounds that don't have an English equivalent? — Ƶ§œš¹ [aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi] 23:49, 6 April 2010 (UTC)
- They're not always useless. There's nothing wrong with telling people that [j] is an English y, not an English j, and that [dʒ] is an English j, not an English d followed by a bit of secret code. But I agree that they are sometimes a bad idea - that's why I didn't include the ü and ö sounds when I started the Vowel Equivalents column. We just need to be judicious and decide when to include English equivalents and when not. Perhaps the solution is to write something like "No close equivalent" whenever there isn't one, so that both first-time visitors and regular contributors know why the box is being kept empty. Lfh (talk) 16:28, 6 April 2010 (UTC)
- This discussion just supports my long-held opinion that the "Closest English equivalent" columns on these pages are useless or even harmful. There is no English equivalent to the ü and ö sounds, and putting in things that might sound vaguely like them in certain dialects just makes the page more confusing rather than less. +Angr 08:55, 6 April 2010 (UTC)
Just to add a bit of levity to this serious discussion, I don't know if anyone caught my reference above to Arnold Schwarzenegger speaking at a Republican Convention in 2004. He used the phrase "Don't be economic Girlie Men!" When "Aahnowld" says "girlie," I feel it is almost right on the money for the [ø:] sound in German, represented by the long [ö] in a word like "schön." You can listen to it here at YouTube and see what you think. After doing my own research, I found out that Ƶ§œš¹ was right on about the sounds in English for ir,er and ur being identical. Therefore, it is inconsequential which one is used (if any) to approximate the [ø:] and [œ] sounds. I suppose we may never find any reasonable English equivalent for the [yː] sound, or the [ʏ] sound.
I hope that we can all agree at some point that non-German speakers might appreciate a bit of assistance in pronouncing these strange sounds to their ears, as Ƶ§œš¹ pointed out. That is why I suggested some link to A Guide to German Pronunciation, which I thought would be extremely helpful to people who want to get it right. My mother tongue is German, and I don't need a guide because it is instinctual for me, but I do appreciate that someone from another language group might want some help. Skol fir (talk) 18:22, 7 April 2010 (UTC)
- Then this is the wrong page for two reasons: First, it's not an article but a Wikipedia:-prefixed page. Second, it's called IPA for German, not German phonology. In that article there already are two links to pronunciation websites, one of them helping you to pronounce the lovely German (nordic) names "Konstantin" or "Katharina" (i.e. I encourage you to replace that link with your suggestion).
- Some general remarks: Why is [j] in yard just an approximation of [j] in ja? Wouldn't it be better to distinguish between correspondency and approximation? I assume that it's mostly the vowels that can only be approximated. In addition, the replacements of non-native nasals are too arbitrary, since there are many other common ways of replacing them (e.g. [ɔː] for all of them, or by saying Parfüm [pafyːm] in stead of Parfum). Also, I never heard someone say [woːkmæn], it's rather [vɔːkmɛ(ː)n], so I doubt that notice about replacing with [oː]. --Zahnradzacken (talk) 09:26, 8 April 2010 (UTC)
- Zahnradzacken, I like your idea of noting correspondency where appropriate, but wouldn't that need a separate column from the approximations? I actually think that if people really want to find the correct pronunciation, they would not go to an IPA table because that is primarily for the alphabet—the pronunciation is just an extra, optional feature. It might help to say at the head of the column "Closest English Equivalent" and that would cover approximations and true equivalents. A reference to phonology is already given in the intro: "See German phonology for a more thorough look at the sounds of German." The Wikipedia:-prefaced table format is by nature short and sweet, with varying degrees of sweetness. :)
- I would gladly place my link to German Pronunciation at the appropriate site, in addition to the one(s) already there, as they also serve a purpose.
- Skol fir (talk) 01:48, 9 April 2010 (UTC)
convert?
The Black Forest Clinic has the pron. of its German name given in English, which is a bit odd. Maybe s.o. can convert? — kwami (talk) 06:00, 7 May 2010 (UTC)
- Done +Angr 06:38, 7 May 2010 (UTC)
Consistency
While looking at the preceding question and correction of it, I noticed a strange inconsistency in the IPA-de guide. Here is an extract from the table:
IPA | Long | IPA | Short |
---|---|---|---|
aː | Bahn | a | Dach |
eː | Beet | ɛ | hätte |
iː | viel | ɪ | bist |
oː | Boot | ɔ | Post |
uː | Hut | ʊ | Putz |
Except for a, all of these denote both a change in length and in quality. In fact, the difference for a is in practice even greater than for i and u. So we should modify the table to have /aː/ versus /ɑ/ (and, incidentally, reverse the lines). −Woodstone (talk) 07:05, 7 May 2010 (UTC)
- Um, [citation needed]? According to both what I have read in descriptions of German phonology and to what I hear from native German speakers every day, there is virtually no qualitative difference between /aː/ and /a/, and if there is, it is the short vowel that is a bit closer to [a] and the long one that's a bit closer to [ɑː]. The system presented on this page is closely based on the one used by Duden, who do show a qualitative difference for the nonlow vowels but not for the low ones. +Angr 08:33, 7 May 2010 (UTC)
- I will not press this further, but when I hear spoken German, there is hardly any difference in quality for the i and u pairs (always front), but fully evident difference for o, e and a, where without doubt the short ones are all back. That is also the way I learnt it at school. My private (unfounded) theory is that the confusion between front and back a is caused by the fact that the difference is not phonemic in English and consequently, many English people aren't able to hear any difference. In my language they are fully separate phonemes, which biases my interpretation. −Woodstone (talk) 09:03, 7 May 2010 (UTC)
- I hear a definite difference between /i:/ and /I/ in German, although the difference between /u:/ and /U/ is not as great. But the qualitative difference between /a:/ and /a/ is really very small, and as I say, this is what descriptions of German written by Germans (like Richard Wiese, not to mention Duden) say as well. +Angr 09:17, 7 May 2010 (UTC)
- I will not press this further, but when I hear spoken German, there is hardly any difference in quality for the i and u pairs (always front), but fully evident difference for o, e and a, where without doubt the short ones are all back. That is also the way I learnt it at school. My private (unfounded) theory is that the confusion between front and back a is caused by the fact that the difference is not phonemic in English and consequently, many English people aren't able to hear any difference. In my language they are fully separate phonemes, which biases my interpretation. −Woodstone (talk) 09:03, 7 May 2010 (UTC)
AWB maintenance
There are a thousand transclusions of this, so I'm doing basic maintenance with AWB. Really basic: I'm only flagging IPA symbols which do not occur in the native part of our table, such as g, χ, ʀ; oeuiy without length and length without aoeuiyɛ; diphthongs like aj, au, ɔɪ; and non-syllabic marks on anything but iuɐ. I'll leave more exacting corrections to others, but let me know if there's something basic that I didn't think of. (One thing I purposefully didn't bother with was Vɐ without a non-syllabicity sign, as at Arnold Schwarzenegger, as we don't bother with one for other falling diphthongs.) I haven't touched length in words derived from Romance, incl. Gestapo, as I don't know the difference. I haven't added stress to stressed monosyllables. Maybe next time around.
Question: Is Haydn meant to be [ˈhaɪdn̩] with a syllabicity mark? If so, we should add that to our key. But why should Niedersachsen have a syllabic n when Sudeten has a schwa?
Also, do we want ɐ or ʁ after short vowels, or do we allow it to be random? — kwami (talk) 07:08, 10 May 2010 (UTC)
- Haydn, Niedersachsen and Sudeten should be consistent with each other. German doesn't distinguish syllabic n from schwa + n any more than (rhotic) English does. I'd vote for schwa + n since it's typographically easier, though the orthography may lead people to balk at the transcription [ˈhaɪdən] (just as many English speakers balk at the transcription [ˈrɪðəm]). I'm in favor of ʁ after short vowels because it's what Duden does and because some people really do pronounce it that way; but I realize that a lot of editors are in favor of ɐ after short vowels because it more accurately represents the pronunciation of most Germans. +Angr 12:55, 10 May 2010 (UTC)
- I'll live the latter alone then unless there's some consensus here. We can always go back. About a quarter of the articles need cleanup as it is, and that's enough for me for now.
- I'm finding a lot of [s] for [z], though I'm not searching for it. — kwami (talk) 17:46, 10 May 2010 (UTC)
- A lot of the [s]'s that you've changed to [z] were correct as [s]. German has final obstruent devoicing at the syllable level, and it doesn't have regressive voicing assimilation, so [z] never occurs in the syllable coda, not even when followed by a voiced sound. So Regierungsbezirk, Regensburg, and Köslin all have [s], not [z]. +Angr 21:39, 10 May 2010 (UTC)
- I'm finding a lot of [s] for [z], though I'm not searching for it. — kwami (talk) 17:46, 10 May 2010 (UTC)
- Ah, okay. Still to [z] between vowels (before a vowel-initial word)? Or is there always a glottal stop there?
- Also a lot of entries lack secondary stress marking. Is there agreement on what should and shouldn't be marked 2ary stress on WP? Or is the symbol just here to explain the cases in which it's used? — kwami (talk) 22:09, 10 May 2010 (UTC)
- No enchaînement in German either; vowel-initial words are always glottal stop-initial. Duden marks secondary stress sparingly: it marks it on ˈBahnhofˌstraße, but not on ˈBahnhof alone. Maybe it has to be a compound in which the two stressed syllables are not adjacent. I wouldn't worry about adding it, or removing it except where it's obviously wrong. +Angr 22:54, 10 May 2010 (UTC)
stressed and unstressed vowels
The phonology article notes that i:, y:, etc. are not found in unstressed positions, but this key lists them as "stressed and unstressed". It would be useful to distinguish which can actually be both. If /u:/ can only be stressed, and /u/ is only found in loans, does that mean that Luzern is mistranscribed? But I see that we have unstressed /o:/ in this very key. Is the phonology article misworded? — kwami (talk) 19:24, 10 May 2010 (UTC)
- The long vowels certainly occur in syllables that don't bear the primary stress, like durchziehen and Durchzug, both of which are stressed on the first syllable, but depending on your analysis, you may believe those syllables bear secondary stress. Luzern was indeed mistranscribed, and I've corrected it (the first vowel is short but tense). It may be considered a loanword; at least, it patterns phonologically like one. If it patterned like a native word , we would expect *[ˈluːtsɐn]. +Angr 20:37, 10 May 2010 (UTC)
- Yes, as I understood it, that is one reason for positing 2ary stress in German. But what of words where 2ary stress is marked, but long vowels still occur in unstressed position? Should they be lax?
- I'm going to need help with short tense vowels. Many of them are probably like Luzern, but it seems clear that many more are errors for either lax or for long tense vowels. Often I can't tell which. I'll post them here when I'm through running through the list. (About 150 to go.) Meanwhile, it would probably be a good idea to review my IPA-de edits (all of my edits since yesterday tagged as AWB), since my German's absolutely pathetic and I can't distinguish these vowels. At the least, this will enable you to check likely mismatches with our conventions without having to trawl through all 1003 articles with IPA-de transclusions. — kwami (talk) 20:47, 10 May 2010 (UTC)
- The usual place for finding short tense vowels in German is in open pretonic syllables like Luzérn, Methán, vitál, Ökonóm, and Psychologíe. I think native words allow only [ə] in open pretonic syllables, and then only in the prefixes be- and ge-. Short tense vowels can also occur in word-final position, as in Kuli, Uhu, and Pankow (and these examples really do make me doubt that they occur only in foreign words). You ask, "But what of words where 2ary stress is marked, but long vowels still occur in unstressed position? Should they be lax?" More likely, they should be short and tense, but I'd have to see some examples to know for sure. I think the German-speaking editors who have transcribed these words are more likely to get the quality right and the quantity wrong than vice versa. +Angr 21:26, 10 May 2010 (UTC)
- For some articles, though, it appears that /eiou/ are being used for lax vowels, and in others for long tense vowels, as we often see in English. In Bönen it appeared to be both at once, at least going by the spelling. One article transcribed the ow in Güstrow as long tense, and another as short tense, so I've been "fixing" places in -ow to long; but your comment on Pankow makes me think that the original article was wrong. Anyway, that's why I think it's a good idea for s.o. to review my edits. Even if I need to be reverted, at least we've confirmed that these potentially ambiguous transcriptions are correct. (I don't think all were done by native speakers.)
- Are der and die at least /der/ and /di/? I'd thought they were /dEr/ and /di:/. — kwami (talk) 22:00, 10 May 2010 (UTC)
- The unstressed -ow should be short tense, and I've corrected Güstrow. Whoever did Bönen was using "e" for [ə], and I've corrected that too. Der and die are actually [deːɐ̯] and [diː] (and while we're at it, des, dem and den have long [e:] as well). Leave it to German to have superheavy syllables in the definite article... +Angr 22:59, 10 May 2010 (UTC)
- Shouldn't that be [dɛs], [deːm], [deːn]? −Woodstone (talk) 03:35, 11 May 2010 (UTC)
- Yes, you're right. Des is indeed [dɛs]. +Angr 20:36, 11 May 2010 (UTC)
- Shouldn't that be [dɛs], [deːm], [deːn]? −Woodstone (talk) 03:35, 11 May 2010 (UTC)
- Yeah, I remember those as exceptions now. Thanks. — kwami (talk) 23:53, 10 May 2010 (UTC)
- The unstressed -ow should be short tense, and I've corrected Güstrow. Whoever did Bönen was using "e" for [ə], and I've corrected that too. Der and die are actually [deːɐ̯] and [diː] (and while we're at it, des, dem and den have long [e:] as well). Leave it to German to have superheavy syllables in the definite article... +Angr 22:59, 10 May 2010 (UTC)
- The usual place for finding short tense vowels in German is in open pretonic syllables like Luzérn, Methán, vitál, Ökonóm, and Psychologíe. I think native words allow only [ə] in open pretonic syllables, and then only in the prefixes be- and ge-. Short tense vowels can also occur in word-final position, as in Kuli, Uhu, and Pankow (and these examples really do make me doubt that they occur only in foreign words). You ask, "But what of words where 2ary stress is marked, but long vowels still occur in unstressed position? Should they be lax?" More likely, they should be short and tense, but I'd have to see some examples to know for sure. I think the German-speaking editors who have transcribed these words are more likely to get the quality right and the quantity wrong than vice versa. +Angr 21:26, 10 May 2010 (UTC)
questionable transcriptions
Okay, here are 63 articles I flagged w AWB but did not change.
- Illegal: Stressed short tense vowels in Wolfgang Amadeus Mozart ([ama'deus]), Christian Democratic Union (Germany) (also has odd syll. in [un.'jo:n]), Theodor Fontane, Karl Lueger, Toblerone, Altona, Hamburg (also no schwa), Gebhard Leberecht von Blücher (long [y] or short?), Ernst Toch, Migros, Adrian Sutil (also [j], [u]), Clueso, Montabaur (also full first [a]), Bucerius Law School ([bu'tseRius]), Zeno.org, Difficile lectu (Mozart) (in G. pron of Latin 'mihi')
- Probably some errors: Final short tense vowels in Willy Brandt, Willi Stoph, Gregor Gysi, Wizo, Benno Ohnesorg, St. Pauli, Möhlin, Röstigraben (in Roesti), Bülow (also in 'von'), Teterow, Großvenediger (in 've' and 'di'), Hagenow, Puhdys, August Wilhelmj, Lüchow, Achim (but long in Achim (name)), Mirow (also final voiced [v]), Jerichow, Guildo Horn, Torgelow, Altentreptow, Dassow, Grabow, Malchow, Anajo, Currywurst (the [i]),
Other dubities:
Alfred Hrdlicka (syllabic R, long unstressed [e:]) Friedrich Hölderlin (nonsyllabic 2nd [i] in 'Christian'?), Johanna Spyri (no schwas), Dirk Nowitzki (odd), Germany women's national football team (odd stress pattern), Sachertorte ([a:] or [a]?), Aktiengesellschaft and .ag (non-syllabic [i]?), Elfriede Jelinek (no schwas, short [i]), Breitspurbahn (long [u:]?), Price index (illegal diphthong), Albert Speer, Jr. (letter ash; per Albert Speer it's ˈʃpɛɐ]) Amerika, Saxony (no schwas), Johanna (no schwa), Akif Pirinçci (no stress), Diemelsee (odd syll. [u:.plant] for 'Upland'), Ziesar & Ziesar (Amt) (non-syllabic [i]?), Nazism (odd stress 'Nationalsozialistische'), Cäcilie (Richard Strauss) (short or non-syll. 2nd [i:]?), Reiner Knizia (non-syll 2nd [i]?), Osnabrück, Osnabrück (district) (no schwa), Ernst Chladni (final [I]), Grossglockner (long [o]?), Achern (changed [əʁ] to [ɐ]) Jasmin Ouschan (novel diphthong)
- Where's the stress? Werner Herzog, Zyklon B, Bad Wörishofen, Burgebrach (auto change may've been wrong),
I'll go through my changes now for things I think should be dbl checked. — kwami (talk) 01:14, 11 May 2010 (UTC)
- Added final vowel length I prob'ly shouldn't have: Otto Dix, Alt-Treptow
- Added vowel length to illegal stressed short tense vowels, which possible should be lax instead, or vice versa: Albrecht Dürer, Major (Germany), Uwe Muegge (made lax), Schrammelmusik, Gerhard Schröder (made lax),
- Other: Johann Gottlieb Fichte & Gottfried Leibniz (made -ie- long), Möbius strip & August Ferdinand Möbius (made 'i' non-syll; plus made 'oe' long in latter), Hans-Jürgen Papier (the -ier), Alois Glück (made 'a' long), Machern (is the 'a' long or short?), Schwedt (vowel)
If anyone else wants to take a look, it looks as though Angr covered my edits through about 15:25 May 10, with 'Gymnasium'. — kwami (talk) 20:34, 11 May 2010 (UTC)
- That sounds about right. I was up until late last night doing that, and I don't want to repeat that again tonight. I'm going to bed now. Sometimes, a good night's sleep is more important than the immediate accuracy of German phonetic transcription on Wikipedia! +Angr 21:02, 11 May 2010 (UTC)
- Ah yes, sleep. What a quaint concept! — kwami (talk) 00:37, 12 May 2010 (UTC)
- Angr, I'm going through them again, but only to add stress to monosyllables; unless either the Das in Das Boot is stressed, or the Boot isn't, I doubt there's anything that requires review.
- I'm leaving things like Gelsenkirchen, assuming that they really are 2ary-1ary in stress. — kwami (talk) 08:03, 12 May 2010 (UTC)
- Ah, changed Rothenburg ob der Tauber to match der above, but want to verify it isn't lax in this name. — kwami (talk) 09:12, 12 May 2010 (UTC)
- I think Gelsenkirchen is right. German place names often have a non-compound stress pattern even when they look like compounds. (Another example is Salzgitter, which is stressed Sàlzgítter as a place name; if it were an actual compound meaning "salt grating" it would be Sálzgìtter.). I can't say I've ever heard anyone pronounce "Rothenburg ob der Tauber", but the transcription of der is no doubt right, at least in careful speech. (In quick speech it may get reduced to [dɐ], but that doesn't need to be indicated here.) +Angr 11:04, 12 May 2010 (UTC)
- Alright, good. I'll ignore such patterns then.
- When you get the chance, I'd appreciate it if you'd comment on the lists above; if the final /oeui/ words are okay, I can take them off my to do list. Etc. — kwami (talk) 20:21, 12 May 2010 (UTC)
- I think Gelsenkirchen is right. German place names often have a non-compound stress pattern even when they look like compounds. (Another example is Salzgitter, which is stressed Sàlzgítter as a place name; if it were an actual compound meaning "salt grating" it would be Sálzgìtter.). I can't say I've ever heard anyone pronounce "Rothenburg ob der Tauber", but the transcription of der is no doubt right, at least in careful speech. (In quick speech it may get reduced to [dɐ], but that doesn't need to be indicated here.) +Angr 11:04, 12 May 2010 (UTC)
ç pronunciation
ich (German) = hue (English)??? Is this correct? --MK (talk) 00:52, 10 June 2010 (UTC)
- Depends on the dialect. In Dresden it's more like shoe. Also, few English speakers actually have a [ç]; hue is about as close as we're likely to get. — kwami (talk) 01:31, 10 June 2010 (UTC)