Talk:Swedish phonology

From Wikipedia, the free encyclopedia

Peer review Swedish phonology has had a peer review by Wikipedia editors which is now archived. It may contain ideas you can use to improve this article.
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.
Archive
Archives
About archives • Edit this box

Contents

[edit] Too few vowels

Standard Swedish actually have at least 19 vowels: 16 pure vowels, two half-vowels and one diphthong. I have made a list of them in alphabetical order.

1. A (long): pronounced as “a” in “father”.

2. A (short): pronounced as “u” in “under”.

3. au: pronounced as “ow” in “owl”.

4. E (long): pronounced as the first “e” in “here”.

5. E (short): pronounced as “e” in “red”.

6. I (long): pronounced as “ea” in “read”.

7. I (short): pronounced as “i” in “wit”.

8. O (long): pronounced as “o” in “open”.

9. O (short): pronounced as “u” in “put”.

10. U (long): the close central rounded vowel pronounced long.

11. U (short): the close central rounded vowel pronounced short.

12. Y (long): pronounced as “ue” in French “rue” (“street”).

13. Y (short): pronounced as “u” in French “Gustave”.

14. Å (long): pronounced as “a” in “call”.

15. Å (short): pronounced as “o” in “corn”.

16. Ä (long): pronounced as “a” in “dad”.

17. Ä (short): pronounced as “a” in “hat”.

18. Ö (long): pronounced as “i” in “girl”.

19. Ö (short): pronounced as “a” in “about”.

Please tell me if any of my descriptions of the sounds are wrong!

2007-02-17 Lena Synnerholm, Märsta, Sweden.

Standard Swedish does not have phonemic diphtongs. In the southern dialects long vowels are very marked diphtongs, but they are limited primarily to Skåne and the surrounding areas. The dialects around Stockholm also have diphthongized long vowels, but like with the souther diphtongs, they are not used in the rest of the country, and they are not as noticeable as the southern ones. But these two are separate dialect areas and don't really coincide. The instance of the use of an /au/-diphthong is limited to a few fairly recent loanwords, like aula, paus and foul ("auditorium", "pause" and "foul" respectively), so saying that Standard Swedish has diphthongs isn't really accurate.
So that brings the total down to 18, and short "e" and "ä" coincide in pretty much all dialects (even if they can be pronounced differently from one region to another). Very few people in Swedish actually distinguish män "men" men "but" or värk "pain" and verk "work, act".
As for the number of phonemes in Standard Swedish, it's 17 for pretty much all of Sweden. To my knowledge, there's not much dispute about this in the linguistic community as far as I know, even if it's common to claim that there's still a 9-vowel system with long and short pairs being used. This description is historically accurate and it's clearly visible in the spelling of many words, but it no longer accurately describes how modern Swedish is actually spoken. And the existence of two semi-vowels in Swedish is definitely new to me. It's not in any of the major phonologies.
If you're still unsure about this, I recommend checking the references provided at the bottom of the article.
Peter Isotalo 18:32, 17 February 2007 (UTC)

":So that brings the total down to 18, and short "e" and "ä" coincide in pretty much all dialects "

Depends what you mean with "pretty much all dialects". They are separate in most of Småland (I myself pronounce "rett" and "rätt" differently) and from what I can tell much of Norrland as well. In Svenska Språknämndens "Svenska skrivregler" from 2000 they transcribe short e as "e" and simply notes that the vowel has merged with ɛ in most of eastern Sweden. /Viktor —Preceding unsigned comment added by 83.254.249.46 (talk) 23:19, 28 April 2008 (UTC)

Yeah, the analysis seems based on the ortography rather than the spoken language. The distinction between "pure vowel" and "half-wovel" in particular, looks strange. The article Labial-palatal approximant mentions "Swedish: yla [yɥla] "howl" although that might be secondary. 惑乱 分からん 02:54, 20 February 2007 (UTC)
That [yɥ] is a not uncommon realization of /y:/. There are analogous realizations for all the long high vowels.
Peter Isotalo wrote:
"Very few people in Swedish actually distinguish män "men" men "but" or värk "pain" and verk "work, act"."
Modern Swedish orthography isn't really 100% relevant on this. The short e vs short ä in western dialects is based on what the vowel quality and consonant quantity was in Old Swedish. The words verk and värk have the same vowel from an etymological point of view (OSwe verk [vɛrk] for both). In any case, the r will complicate things in your example. The same goes for män and men, they had the same vowel in Old Swedish (menn [mɛnː] vs men [mɛn]), so one wouldn't expect any difference here either. A more relevant minimal pair is e.g. ett vs ätt, where the former comes from Old Swedish étt [eːtː] and the latter from Old Swedish ǽtt [æːtː].
So, the e vs ä only exists where the spelling respects the etymology - if Old Swedish had no difefrence in spelling, then modern Swedish dialects can't be expected to either.
Jens Persson (217.211.10.14 (talk) 13:42, 29 March 2008 (UTC))

I accept your dialect argument about the diphthongs. You are probably also right about the number of phonemes. (I don't make any difference between “verk” and “värk” ether...) About the “half-vowels”, I actually did not know the English term. It was not in my dictionary ether so I made a literal translation of the Swedish term (“halvvokal”). It stands for a group of language sounds that can be used as vowels yet have some consonant properties. Except for the close central rounded vowel I only know about one such sound: the one written with the first letter in the Hebrew alphabet. I hope I have explained it clear enough for you to understand!

2007-02-22 Lena Synnerholm, Märsta, Sweden.

A close central rounded vowel (which is considerably more front in the Swedish spoken around Stockolm) is not a semi-vowel per se. In fact, no vowels are. What you're referring isn't really a separate sound, but a phonological process which occurs in many languages. For example Mandarin and in the diphthongs of German. However, this occurs only in a few loan words in Swedish. I can't speak for minor dialects, but I'm pretty sure it's fairly marginal.
Peter Isotalo 15:43, 22 February 2007 (UTC)

Either I have misunderstood the concept or it is not called “half-vowel” (or “semi-vowel”) in English.

2007-02-24 Lena Synnerholm, Märsta, Sweden.

I think you have misunderstood(?) The Swedish Wikipedia article on halvvokal seems to be about the same concept as semi-vowel. I think the scope of the article is about "standard, native" phonology, not about exceptions in loanwords and dialects. (Dialects could be mentioned, though, but they would require their own sections.) 惑乱 分からん 22:45, 25 February 2007 (UTC)
I think the comment about aleph in the Hebrew alphabet refers to the glottal stop. It isn't mentioned because it's not considered a separate phoneme in Swedish, i.e. there are no words with change in meaning depending on whether they're pronounced with or without a glottal stop. 惑乱 分からん 22:52, 25 February 2007 (UTC)

[edit] Question concering long I

Hi, can anyone tell me exactly how this (long i) is articulated? I see that it is like "ee" in meet but in Swedish it sounds like it's nasal or something, so could anyone explain it to me? --Sergiusz Szczebrzeszyński |talk to me||what i've done||e| 22:16, 6 April 2007 (UTC)

It's a diphthong in Central Swedish that leans towards a schwa. Same as all the long vowels. It's not pronounced the same way all over Sweden, though. Some dialects have different diphthongs and other don't have diphthongs at all.
Peter Isotalo 13:20, 20 April 2007 (UTC)
I think a nasal pronunciation might be dialectal. I know, for instance, that Swedish has a dialectal variation known as "Viby-i" or "Lidingö-i", where a long i-vowel often gets some kind of buzzing quality. 惑乱 分からん * \)/ (\ (< \) (2 /) /)/ * 15:32, 28 April 2007 (UTC)
Thanks guys :D --Sergiusz Szczebrzeszyński |talk to me||what i've done||e| 00:39, 9 May 2007 (UTC)

[edit] The short e

Shouldn't there be an "e" listed among the short vowels as well, although it's comparatively rare? I'd guess most speakers pronounce "ett" (the cardinal) and "ätt" differently. Or is that non-standard?Kallerdis 13:48, 15 May 2007 (UTC)

Different pronunciations for ett and ätt seems to be very rare in modern Swedish. The vowel might be pronounced differently from one dialect to another, but they're still homonyms, not a minimal pair. I'm sure there are dialects that differentiate, but if that's the case, I would like to know which those dialects are.
Peter Isotalo 15:10, 15 May 2007 (UTC)
At least I would distinguish them -- maybe it comes from my native dialect (Western "östgötska"). Wouldn't you distinguish the cardinals ett or en from the determiners (which rhyme with ätt and än)?Kallerdis 14:36, 6 June 2007 (UTC)
I certainly wouldn't. According to Claes Garlén (Svenskans fonologi, Studentliteratur 1988) the merger of short e and short ä is characteristic of the dialects of Finland and eastern Sweden, so presumably the best place to look for dialects maintaining the distinction would be Västsverige. Orcoteuthis 19:13, 15 July 2007 (UTC)
Kallerdis, if you're sure about this, you should record a sample for us as it would make a very nice illustration for the article. Preferably in the context of a sentence; a suggestion is to use a sentence like jag sa XXX igen, a common phrase use by many phoneticians when recording samples for surveys. I've always been curious about which dialects that actually still have both short and long /e/.
Peter Isotalo 20:26, 15 July 2007 (UTC)
I don't make a huge difference, but there certainly is a difference between how I say "ätt" and "ett". (SE Småland here, btw). Heh, I remember from school when dealing with spelling, how there were a section in our textbook for different sounds which were at risk of being misspelled - skj/stj/sk/sch/..., k/kj/tj, o/å and so on. There was also a section about when to use "ä" and when to use "e" which we skipped due to "only people in Stockholm confuses those sounds anyway" :D Of course that's exaggerated, but e vs. ä never seemed to really confuse us, so I think there got to be something which told us when to use which, and I have difficulty seeing what else than pronunciation would accomplish that. \Mike(z) 21:19, 9 September 2007 (UTC)
Well, ok, there are exceptions. verk and värk, I pronounce the same. men and män: it feels like I pronounce them virtually the same, but not exactly. Lacking recording equipment, it's hard for me to tell how much one would hear of that difference, though. However, ätt and ett are definitely distinct. \M
The issue is also whether other speakers can hear the difference or not.
As an immigrant to Sweden, where I've lived for well over 40 years by now, one of which I studied linguistics, I can say that I rarely, or maybe very rarely, hear the difference, but more often one can see the difference in the speaker's face. The lips produce a more rounded figure for short /ä/ and a more smiling position for short /e/.
This does, however, not apply to vowels followed by /r/.
Without any doubt the article ought to be corrected on this point.
My personal experience is from Uppsala, Lund, Ljungby, Norrköping, Jönköping and Motala.
/Henrik Schmidt —Preceding unsigned comment added by 80.251.192.4 (talk) 04:02, 4 December 2007 (UTC)

[edit] b, d, g as approximants?

In the first table under "Consonants", /b d g/ are shown as having approximant realizations. That's news to at least this native speaker, and if true really ought be expanded upon in the sections about allophony. Orcoteuthis 19:32, 15 July 2007 (UTC)

Being a native speaker rarely means that one is full aware of the exact phonetic gestures of one's own language, especially for adults who are heavily influenced by orthography. It's actually almost easier to notice these things as a non-native. In general, though, approximation of plosives is a very common feature of many languages and in Swedish occurs colloquial speech. In this case it's taken straight out of Engstrand's Fonetikens grunder.
Peter Isotalo 20:22, 15 July 2007 (UTC)
It should still be expanded on in the article's prose. — Ƶ§œš¹ [aɪm ˈfɻɛ̃ⁿdˡi] 21:04, 7 April 2008 (UTC)

[edit] The short "i".

What is rendering the Swedish short “i” as [ı] in IPA based on? This is the same sound as the short “i” in English "hit" or German "mit" and if you ask me these sounds are in no way similar to the Swedish sound. To me the Swedish short “i” is simply the same (in quality) as the long “i” i.e it should be [i]. This is not the first time I’ve seen this and I’m curious to know where it comes from, I feel it is an obvious error and yet it keeps on being repeated in various forums. Nothingbutmeat 16:28, 30 July 2007 (UTC)

It's slightly, but not considerably different. For one thing, it's slightly lower. If you want sources, then check Garlén or Engstrand. The latter, btw, is the professor of phonetics at Stockholm University. If you think he's mistaken on this point, you can always email him and I'm sure he'll explain himself.
Peter Isotalo 17:57, 30 July 2007 (UTC)

[edit] Audio samples

The man doing the pronunciation samples has an almost extreme Stockholm dialect. One should know that unlike many other languages, the Swedish that is spoken in the capital is distinct from what is considered Standard or High Swedish. As such, I think that either should this be stressed or there be made available other samples. 213.112.137.175 17:01, 14 August 2007 (UTC)

All pronunciations below are made by a Swedish male, age 25, in a variety of Central Standard Swedish spoken in the greater Stockholm region. I think that's clear enough. And if you think that's an extreme Stockholm dialect, you can't possibly know much about how people in Stockholm speak. I should also point out that what most Swedes refer to as rikssvenska (one term for Standard Swedish) is actually just the more neutralized Stockholm dialects, rather than some bogus "neutral" Swedish. Linguists tend to prefer the term standardsvenska and include all major regional dialects in the concept of Standard Swedish.
Peter Isotalo 17:52, 14 August 2007 (UTC)
You live in Stockholm yourself, no wonder that you don't notice. As for what Standard Swedish is, try a pre-war Swedish film, for example Hets. Since the teacher speaks Standard Swedish and some of the students, on the other hand, speak with a Stockholm accent, there is ample contrast to be heard between them. That Standard Swedish is merely a neutral form of today's Stockholm dialect is nonsense.
And no, the "warning" is not clear enough, as, firstly, it contains the phrase "Central Standard Swedish" and presents the samples as done in a mere variety of it, while we are in fact speaking of a dialect exclusive to Stockholm, and secondly because foreign readers very easily will assume that the dialect spoken in the capital is the neutral or prestige dialect. Therefore I am reverting your edit until additional samples have been submitted to which the current ones can be compared. 213.112.137.175 15:20, 15 August 2007 (UTC)
There is no dialect neutral Swedish. In fact, there are no neutral prestige dialects whatsoever; they are all influenced by one form of geographical vicinity or another. What most people, especially people without any knowledge of linguistics, perceive as rikssvenska is actually just a rather standard Central Swedish regional dialect. This is rather easily proved by simply trying to find dialects outside of this area that actually sound the same. However, it is extremely common for people who live Stockholm to believe that their own speech is the epithome of Standard Swedish as long as they don't sound like the stereotypical 45-year-old cab driver from Södermalm.
Please read the article Standard Swedish to get your bearings on the topic. If you don't trust that, check out the article rikssvenska in Nationalencyklopedin. If you want to insist on these type of disclaimers, you better back up your claims with sources, since this article is rather well-referenced.
Peter Isotalo 16:28, 15 August 2007 (UTC)


This article in its current stand has several good points, but its main failure is that it seems to be written by people who overestimate the importance and/or dominance of the Stockholm-Uppsala varieties, but at the same time some paragraphs correctly point out that this language, despite its small size, has several localized high status varieties. Particularly compared to nearby languages such as French, British English and European Castillian, this is a peculiar feature, although not totally different from the situation for German and Dutch?

Much of this article's content, and particularly the audio samples, whould fit better into an article on "Central Standard Swedish phonology".

/Henrik Schmidt —Preceding unsigned comment added by 80.251.192.4 (talk) 04:19, 4 December 2007 (UTC)

I recommend that this issue is addressed by expanding the information and recording more samples.
Peter Isotalo 13:33, 11 December 2007 (UTC)
I completely agree with Henrik Schmidt here. The phonology describes Central Standard Swedish (the dialect spoken by "educated" people in the Mälardalen region), and the audio files represent some middle-class variety of Standard Stockholmish. Living in Jämtland, I can accept the Central Standard Swedish phonology description, but the awkward audio samples seem out-of-place in a serious article on Swedish phonology. Maybe they belong to an article on middle-class white Standard Stockhomish?
Jens Persson (217.211.10.14 (talk) 13:15, 29 March 2008 (UTC))

Just a small note here on an audio sample that I believe need re-recording. The "västkustskt" recording is cut off at the beginning, and sounds more like "ästkustskt". 83.255.186.163 (talk) 11:38, 27 December 2007 (UTC)

Ugh... I can't believe they still haven't fixed this problem. The audio file is fine, but your player for some reason chooses to truncate the beginning of it. Try using Quicktime or Winamp instead.
Peter Isotalo 11:49, 27 December 2007 (UTC)

[edit] Plosives

Aren't the plosives unaspirated in Finland Swedish? JdeJ 20:10, 23 September 2007 (UTC)

I don't think so, but many people with Finnish as mother tongue probably have this artefact when speaking Swedish.
Jens Persson (217.211.10.14 (talk) 13:17, 29 March 2008 (UTC))