From Wikipedia, the free encyclopedia

This article is part of WikiProject Phonetics, an attempt to build a comprehensive and detailed guide to phonetics and descriptive phonology on Wikipedia. If you would like to participate, you can edit the article attached to this page, or visit the project page, where you can join the project and see a list of open tasks.

I would like to avoid original research here as much as possible, but there is an interesting phenomenon I think should be noted. In German language classes in the US (and presumably other English speaking countries as well), Standard German pronunciation is taught. Therefore ich-Laut is [ç]. I have observed (as have my German teachers and a few Germans I have spoken to about the matter) that many English speakers, particularly those with little or no formal training in German, pronounce ich-Laut as [ʃ]. When I have noticed it myself and asked the speakers where they learned German, they have said that they learned a few German phrases from Germans from the south of Germany. Unfortunately, this has been only on two occasions and both were before I was aware of the distinct pronunication in the Rhineland dialect, so I didn't ask for more specific location. I went ahead and qualified the statement about English speakers having difficulty discerning [ɕ] from [ʃ] or [ç] with this example from German, but it could probably use some peer review. Also, if anyone knows of a good reference about acquisition of German as a second language by English speakers or studies of German with an English/American accent, this would be very helpful. CyborgTosser (Only half the battle) 13:13, 30 May 2005 (UTC)

Almost all the learners of German I have ever met (from the UK or France) pronounce the ich-Laut as /ʃ/. If challenged, they try to blame it on regional German, but the real reason is obviously that /ç/ is hard to pronounce, and /ʃ/ sounds vaguely similar; so they just substitute it. — Chameleon 15:12, 30 May 2005 (UTC)

It all sounds like unverfiable speculation to me and probably confuses more than it helps. I seriously doubt the vast majority of native English speakers ever manage produce an acceptable [ç] or [ɕ]. I think we should simply skip the "In English" section when there are neither phonemes nor naturally occuring allophones of a particular sound. We'll probably just wind up with people thinking that it's pronounced [ʃ] and never be the wiser.

Time to get bold...

Peter ^Isotalo 22:59, 25 July 2005 (UTC)

I have no problem with [ç] or [ɕ], and yet the ich I heard in Dresden sounded closer to English [ʃ] than to either of those (though I must admit that I didn't listen terribly carefully). East Germans, in both Dresden and Berlin, have told me that this is a regional pronunciation. I don't know that it truly was [ʃ], and of course it's a long way from the Rhineland, but maybe that's what some of these people are talking about. kwami 00:31, 2005 July 26 (UTC)

I was born in Saarland (South West Germany) and so I never knew what was the difference between ch (in Standart German [ç]) and sch (in Standart German [ʃ]), until I moved to Northern Germany, as in Saarland both were pronounced equally. By the way, many Germans from Saarland and Palatinate still confuse it, when they try to speak Standart German; the most popular example might be former chancellor Helmut Kohl, who pronounces every sch as [ç], thinking this would be Standart German. 84.132.225.231 18:18, 15 August 2005 (UTC)

Contents 1 Mandarin 1.1 Accuracy 1.2 Russian: щ sound 1.3 Dutch?

[edit] Mandarin

I've tried to fix the representation of Pinyin x. This dorso-palatal monstrosity is far from a simple [ɕ]. It is a coarticulation of two fricatives, one that varies (between different idio- or perhaps dialects, I don't know) between [ç] and the midpoint between [ç] and [x], and another that varies between [ɕ] and perhaps [sʲ]. In addition, it has a [j]-like offglide, like many dorso-palatal consonants (for example Italian gn and gl). So I suspect the most correct IPA representation could be [çɕj͡] with the tie bar extending over all three.

I have also forced Internet Explorer to use Arial Unicode MS to display the characters that would otherwise be shown as squares (both in the article and this page). The trick is to write

{ { I P A |   } }

around the symbols (or around the [] or //).

Apart from that I have added the American example. It goes without saying that I'd appreciate it if a native speaker elaborated that!

Addition 15:43: I have added Polish, the poster child for [ɕ]. I will try to add Belorussian next. Does someone know whether Italian sci is [ɕ] in at least some dialects?

David Marjanović david.marjanovic_at_gmx.at 15:23 CET-summertime 2005/9/10

Where are you getting this? First of all, doubly articulated fricatives are not known to occur in any language. They have been claimed several times, but always turn out to be something else when investigated.

I don't know how else to explain how it sounds. Do you have Skype?

(And for that matter I doubt that the sounds in question occur in any other language... x seems to exist in one Swedish dialect as a realization of /ɧ/, but here I'm judging merely from written descriptions in Wikipedia.)

Secondly, Ladefoged made palatograms and several other measurements of [ɕ] from three Mandarin speakers.

Maybe they didn't come from the same region as the speakers on the tapes of the Chinese textbooks I've used, and as at least some of the Chinese I've heard speak in Vienna's subway, and on TV... or they simply weren't speaking pure Pǔtōnghuà. "Mandarin" is a rather loose term.

Third, the occlusion in Mandarin q is coronal, not dorsal, even in the third speaker.

Difficult to explain how this could sound like what I hear.

That is, it should be transcribed as a [t] with some diacritic. Ladefoged has [t̠]. [c] would mean that the tip/blade of the tongue was not used.

Indeed not. The tip of the tongue touches the lower alveoli in x, q and j. This is even mentioned in writing in one of my textbooks (otherwise they're all not very exact on phonetics).

Fourth, the tie bar needs to go between the segments, not after them.

Sorry, I made it so that it looks convincing in Arial Unicode MS, which is known to have a bug there.

Also, the [j] should be outside the tie bar; it is physically impossible to pronounce it within (unlike a double fricative, which is merely very difficult). kwami 18:57, 2005 September 10 (UTC)

In addition to being simultaneous [ç] and [ɕ], it has a [j]-like offglide like French gn, which I would write [ɲj] with a tie-bar (and feel bad about it because the j probably needs some sophisticated diacritic). So the transcription of x should contain two tie-bars, one above the other and spanning over both the first tie-bar and the j.

P.S. If an American pronounced estimate with a [ɕ], I would think they had a speach defect. It would sound like a lithp. kwami 19:11, 2005 September 10 (UTC)

It's not a true [ɕ], it's somewhere between that and [ʃ]. You know how Bush says "misundreshtmate", don't you?

David 00:18 CET-summertime 2005/9/27

[edit] Accuracy

I have reason to believe that the sound that is pronounced in the ogg file (sound sample) approaches a Voiceless postalveolar fricative consonant, rather than a Voiceless alveolo-palatal fricative. It would be greatly appreciated if a phonetician would look over this article.Danielsavoiu 11:32, 15 July 2006 (UTC)

[edit] Russian: щ sound

The current article reads: In Russian, щ represents /ɕɕ/ or /ɕtɕ/ which I think is very vague and inexact. AFAICS modern Russian does not anymore use the /ɕtɕ/ variant at all - please prove me wrong: /ɕɕ/ everywhere - with one exception: to the vegetable soup, Germans say Borschtsch (Russian Борщ), using the older transliteration. -andy 80.129.116.55 00:20, 9 March 2007 (UTC)

[edit] Dutch?

What about Dutch, folks? They do not pronounce an "ordinary" s either! It resembles the /ɕ/ a lot! Look at the pronunciation of this town: 's-Hertogenbosch. The Dutchman who pronunces this in the audio sample does NOT pronunce the s as normal s in English, but something in-between! Or doesn't he? -andy 80.129.116.55 00:31, 9 March 2007 (UTC)