Talk:Icelandic language/Archive 1

From Wikipedia, the free encyclopedia

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.

Contents 1 Icelandic language speakers 2 Orthographic reforms 3 Official regulator 4 typo? 5 Former capitalisation of commoun nouns? 6 Possible copyright violation 7 Regulation 8 Official regulator again 9 Vocabulary 10 Translation from Icelandic 11 Phonology 11.1 Minimal pairs 12 Addition 13 Removal 14 Glottal stop 15 Velarized, alveolar, lateral approximants 16 Preaspiration, again 17 High Icelandic 18 Spoken in 19 <v> 20 Pronunciation of g 21 More on g

Icelandic language speakers

If you speak icelandic and come across here, please add icelandic words to the comparison table at Germanic languages. Thanks! — Sverdrup 21:04, 10 Mar 2004 (UTC)

Orthographic reforms

A couple of years back I noticed that the spelling in Teach Yourself Icelandic at my then-local library was different to the Icelandic I was finding online. I can't remember the details any more but I seem to recally the old spelling using "z" more. Can somebody shed some light on this? — Hippietrail 02:45, 12 Nov 2004 (UTC)

The "Z" was abolished from the language in 1973 and replaced with an "S", the old rules regarding use of the "Z" where quite complicated and it was the exactly same sound as "S". It's only allowed today to use "Z" in personal and family names. Regardless of that, many people who learned to spell before 1973 still refuse to give up the "Z" and it is still widely used in Iceland's largest newspaper (Morgunblaðið). A respectable secondary school in Reykjavík, Verzlunarskóli Íslands still uses the "Z" in it's name although it goes against the rules. --Biekko 12:23, 12 Nov 2004 (UTC)

And some contributors still use the "z" because of age or where they were educated. The letter z is not a part of the current Icelandic alphabet, but it will only become extinct, when the last writer, who uses it, dies. Io 20:34, 9 August 2005 (UTC)

Official regulator

According to the article the Icelandic Language Institute is the official regulator of the Icelandic language, what is the source of this claim exactly? I'm almost 100% certain that it's incorrect. --Ævar Arnfjörð Bjarmason 02:02, 2005 Feb 9 (UTC)

Look here- http://www.randburg.com/is/lang-instit/

typo?

it's now:

 [..] some of which may be regarded as tenses, other as aspects to varying degrees.

shouldn't that be:

 [..] some of which may be regarded as tenses, others as aspects to varying degrees.

(or is that other than?)

It should have been "others" Io 20:34, 9 August 2005 (UTC)

Former capitalisation of commoun nouns?

I've either read or heard that several Germanic languages besides German used to capitalise all common nouns. In Danish this practice was abolished in a spelling reform in the late 1940s. Can anybody tell me if this was ever practiced in Icelandic and if so when was it abolished? — Hippietrail 16:47, 15 Apr 2005 (UTC)

It was practised in the 19. century, but not consistently. It was abolished in a spelling reform of the 19-twenties. For the precise date, please fill in. Io 20:34, 9 August 2005 (UTC)

Possible copyright violation

The 00:46, 26 Mar 2005 (diff) edit by 194.144.21.61 seems to be a copy-and-paste job from [1]. Notice that the sigur rós homepage has every letter in lower case, even the word "english," just like the anonymous's contribution. Sigur Rós's homepage's pronunciation guide in turn looks like a lot of other pages [2] [3] [4]. --Tokek 02:37, 22 Apr 2005 (UTC)

Reverted to pre-copyvio version--Duk 03:55, 2 May 2005 (UTC)

Regulation

Shouldn't something be mentioned about the fact, that there is an officially stated aim against loanwords in Icelandic, and how calques or old words are being given new meanings to avoid this. (For instance "simi" (seam) for telephone.) Also, other examples of lexicon.

There is in fact, an officially stated stance against foreignisms, but there is no regulatory body with any real power. All in all, the language develops according to the wishes of the speakers. As for examples of the lexicon, look up in any dictionoray. Cheers Io 20:57, 26 November 2005 (UTC)

Official regulator again

Why is Íslenzk málstöð counted as an official regulator? It is (as far I know) only an advisory body. I'd suggest that this field be left empty.Io 19:56, 9 August 2005 (UTC)

Vocabulary

I think it would be interesting to see some more examples of Icelandic vocabulary, and how it, unlike most other European languages, uses calques and alternative meanings to avoid borrowings, such as tölva (a portmanteau of tala(number, related to english tell and tale) and völva) and simi, for telephone (seam). These are examples I know of, by now.

Hmmm, this example from the Swedish Wikipedia article is quite good, if someone is willing to translate it later, note that most of the words seem chosen to sound amusing to Scandinavian speakers, smjör is related to smear, and I think there is a Scots word related to leik(?)

Puristisk språkvård

Grammatikens många böjningsformer är en bidragande orsak till den konservativa isländska språkvården. Precis som svensktalande ogärna säger radior och babyer är för särpräglade utländska ord helt enkelt omöjliga att använda på isländska. Men orsaken är också språkpolitisk och psykologisk: Islänningarna föredrar egna ord framför inlånade. Den isländska språknämnden föreslår, i samråd med fackexperter, och islänningar accepterar i regel nyskapelserna. Under hela 1900-talet har man därför uppfunnit många ord för nya företeelser på grundval av det urnordiska ordstamförrådet. En viss språklig isolering uppvägs av att termer och begrepp blir självförklarande och genomskinliga, ofta också för andra nordbor:

   bibliotek = bókasafn (boksamling)
   kontor = skrifstofa (skrivstuga)
   teater = leikhús (spelhus)
   pass = vegabréf (vägbrev)
   barometer = loftvog (luftvåg)
   margarin = smjörlíki (något smörliknande)
   bli opererad = vera skorinn upp (bli uppskuren)
   demonstrera = halda kröfugöngu (hålla kravgång)
   www = veraldarvefurinn (världsväven)

http://www.intersolinc.com/newsletters/newsletter_49.htm

Icelandic language policy

The Icelandic Language Council, provides advice and guidance about language usage and neologisms. Icelandic language policy is the preservation of the language and creating words for new technology.

Icelanders have been quite successful at cleverly creating words for new technology ‘recycling’ the resources of their own language. Neologisms are formed by combining old words in a new way. Examples:

Computer = ‘tölva’ created by joining of ‘tölur’ (numbers) and the ending of ‘völva’ (prophetess)

Meteorology = ‘veðurfræði’, the combination of ‘veður’ (weather) and ‘fræði’ (science)

Garage = ‘bílskúr’ was created by joining ‘bíll’ (car) and ‘skúr’ (shed).

AIDS = ‘alnæmi’ comes from adding the prefix ‘al-‘ (all) to ‘næmi’ (openness to).

Jet = ‘þota’ from the past participial form of the verb ‘þjóta’ (speed)

Television = ‘sjónvarp’ created by combining “sight + sending”

Radio = ‘útvarp” the combination of “out + sending”

Mobile telephone = sími (line)

Notwithstanding all of the efforts to keep the language pure, some loan words have been introduced into the language. However, loanwords are typically adapted quickly to the Icelandic pronunciation, writing and inflection. Examples:

‘Borrowed’ from Danish:

Precisely = ‘Akkúrat’ Sober = ‘Edrú’ To see, understand = ‘Fatta’

Borrowed’ from English:

The occupation of Iceland by the U.S. and Great Britain during WWII, and increasingly close relations with English-speaking nations, have contributed to the predominantly English influence on Icelandic during the 20th century. Examples of loanwords from British English during WWII include:

Jeep = ‘Jeppi’ Bus = ‘Rúta’ Party = ‘Partí’ Business = ‘Bissness’

Most ‘ borrowed’ words colloquial, words of Icelandic origin are preferred in the written language.

Conversely, English has the word ‘geyser’ for hot spring (from the Icelandic ‘geyser’) and many English words derive from Old Norse, such as ‘mire’ from ‘mýrr’; ‘fjord’ from ‘fjörýr’ and ‘reindeer’ from ‘hreindýr’.

http://www.nat.is/travelguideeng/icelandic_language.htm

Vocabulary innovations. In the late eighteenth century, language purism started to gain noticeable ground in Iceland and since the early nineteenth century, language purism has been the linguistic policy in the country. Instead of adopting foreign words for new concepts, new words (neologisms) are coined or old words revived and given a new meaning. As examples may be mentioned simi for telephone, tolva for computer, thota for jet, hljodfrar for supersonic and geimfar for spacecraft. The Icelandic language committee is an advisory institution which is to "guide government agencies and the general public in matters of language on a scholarly basis."

And the Icelandic word for cinema (bió), where does it come from? Ciacchi 16:18, 27 July 2006 (UTC)

Danish: biograf. Stefán Ingi 15:18, 2 August 2006 (UTC)

From the early biograph movie projector, (from Greek bio (life) and graphein (depict), same roots as in biography.) apparently found only in Swedish and Danish. (Dutch has bioscoop, I'd guess from a competing brand of projectors.) More commonly, words for cinema/movie theater is based on the cinematographe (English "Cinema", German "Kino" etc.) 惑乱 分からん 12:51, 10 September 2006 (UTC)

Swedish has biograf, and the very common colloquial abbreviation bio [ˈbiːʊ] (pronunced as if it came from an Old Norse "bíó"). For example, the largest cinema in my hometown Östersund is called Biostaden, lit. 'The Cinema Town'.

Jens Persson (213.67.64.22 23:54, 3 February 2007 (UTC))

Translation from Icelandic

If anyone knows Icelandic, could you please help in translating the following:

"Þeir Brynjólfur og Ársæll tóku strax til starfa á þeim vettvangi sem um var að ræða, Félagi ungra kommúnista og Jafnaðarmannafélaginu og Alþýðuflokknum gegnum það. Snemma árs 1924 var stofnað Samband ungra kommúnista, en þá höfðu verið stofnuð félög ungra kommúnista á örfáum stöðum utan Reykjavíkur."

"Á Alþýðusambandsþingi haustið 1924 var Félagi ungra kommúnista í Reykjavík meinuð innganga í sambandið og útgáfa Rauða fánans fordæmd.".

? --Soman 13:59, 26 September 2005 (UTC)

It would be something like "Brynjólfur and Ársæll immediately went to work in the forums that existed, The Society of young communists and the Socialist Society and the People's party through that. Early in the year 1924 the Confederation of young communists was founded but at that time societies of young communists had been founded (only) in a handful of places outside of Reykjavík".

I must warn though that I am no expert on how these Societies are usually refered to in English, literally "Jafnaðarmannafélaginu" means society for men in support of equality or something like that. Edinborgarstefan

Phonology

There have been some changes made recently to the Phonology section. Many of the points raised were good but I feel others need discussion.

Firstly, I cannot see how it is possible to argue that c is an allophone of k because there exist many minimal pairs. E.g. "gjóla" and "góla" and similarly for the aspirated versions, e.g. "kjól" and "kól". Also, can "g" or "k" ever follow "j" in writing?

Secondly, I am curious about the claim that "hnífur" is increasingly being pronounced as "nífur". Has there been a study of this or is there some anecdotal evidence?

Thirdly, although I of course think the statement "Icelandic does not have any dialect differences that can cause misunderstanding." is true I also feel it sounds quite general and would allow for much greater differences than there actually are. Maybe a better wording could be found.

Edinborgarstefan 18:44, 30 October 2005 (UTC)

You have just had your first glimpse into the wacky world of generative phonology where the poor innocent Prague school phoneme gives way to esoteric theories about "underlying representation". - Haukur Þorgeirsson 19:18, 30 October 2005 (UTC)

Can I try and summarise the situation: The paragraph on c and k is clearly very inaccurate. The only way of making it accurate is to change is to read something along the lines "The sound c is a phonemic in Icelandic but there are other types of phonological analysis available where phonems are not basic units and where c is not a basic unit." It strikes me that a comment along these lines could more or less be put in every language article and that it doesn't say very much. In conclusion I think the paragraph should be removed. Stefán Ingi (talk) 00:10, 11 November 2005 (UTC)

You can indeed use generative phonology to say almost anything you want and that does, indeed, make its statements close to meaningless in many cases. But I'm hopelessly biased against it and it would be nice to bring in someone else to get a second opinion. Try User:Angr maybe. - Haukur Þorgeirsson 00:45, 11 November 2005 (UTC)

Hi, I just got two requests to add my opinion here. I don't know that much about Icelandic, but I can say this much for now: (1) the phoneme is still alive and well in generative phonology, (2) if there are minimal pairs for c~k and c^h~k^h, then they're separate phonemes, even if the environments in which they can contrast are limited (e.g. even if only the palatals and not the velars can appear before front vowels, if they both can appear before back vowels, they're separate phonemes), (3) the phonology sections discusses "letters" rather than "sounds" too much, there should be a separate section on orthography, (4) rather than saying e ø o are present only as the first members of diphthongs, remove them from the monophthong chart and add a diphthong chart, (5) does Icelandic really have ɮ? --Angr/_{tɔk tə mi} 07:19, 11 November 2005 (UTC)

Thanks for that, Angr, it's appreciated. Our phonology article says that generative phonology is a system where a phonological representation (surface syntactic form) is a structure whose phonetic part is a sequence of units which have characteristic features. Although there are no phonemes in generative phonology, these units are often loosely referred to as phonemes, nonetheless. This is very much my experience as well - a GP text may use the term "phoneme" but in some other sense than classical phonology does. Other than that I agree with everything you say and ɮ in Icelandic is as much news to me as you. - Haukur Þorgeirsson 09:13, 11 November 2005 (UTC)

Thank you very much for the help. I have changed the consonant section, I hope what I have written is clear, but there is clearly more work to be done. Stefán Ingi (talk) 09:29, 11 November 2005 (UTC)

Uh-oh, I'm right in the middle of editing the consonant section myself. I sense an edit conflict in my future. As for what it says about phonemes in generative phonology, that sounds like anti-generativist POV that needs to be cleaned up PDQ. --Angr/_{tɔk tə mi} 09:48, 11 November 2005 (UTC)

Sorry, I won't touch the article more today Stefán Ingi (talk) 10:04, 11 November 2005 (UTC)

Actually, I'd be grateful if you would touch it, namely to add glosses for the Icelandic words I've used as examples. I found them in the grammar book I have, but it doesn't provide glosses and I don't know Icelandic myself. While we're about this, it would be nice if someone would move Icelandic alphabet to Icelandic orthography and then add information about how the letters are pronounced individually and in combination with other letters. Don't worry about any more edit conflicts, I'm done with Icelandic for today. Maybe I'll work more on it over the weekend. --Angr/_{tɔk tə mi} 10:09, 11 November 2005 (UTC)

(Reduced indent) This was great, the table looks very good and the text reads clearly now. I changed the minimal pairs slightly for fun, added glosses, changed eða "or" to iða "vortex" because I am not convinced the ð is pronounced in eða. Also I don't really think "það" is pronounced with a voiced ð in the end, maybe Haukur can comment on that. Finally, I added ɲ̊ the unvoiced palatal nasal to the table, I am sure it is used in bankinn. That got me thinking, is banginn vs. bankinn [pauɲcin] vs. [pauɲ̊cin] so the voicing is phonemic or would people rather like so say that the c is preaspirated? Stefán Ingi (talk) 11:05, 11 November 2005 (UTC)

The second fricative in "það" is often not voiced, especially when pronounced in isolation. Some linguists like to mark Icelandic consonants in these positions as "half-voiced". The voicing of the nasal is phonemic in bankinn/banginn. If you read through Eiríkur Rögnvaldsson he'll tell you some tall tale about how the "underlying" truth here exactly confirms to the orthography and that the "underlying" difference lies in the plosive. In fact he goes to great lengths to obscure the very interesting fact that the voicing of nasals is phonemic in Icelandic whereas the voicing of plosives is not, the orthography notwithstanding. But I'm hopelessly biased against this type of analysis which is why it's very nice to have Angr here to provide some balance. - Haukur Þorgeirsson 11:16, 11 November 2005 (UTC)

If Angr agrees with the non-POV part of this comment then we definitely need to remove the sentence "Of the voiceless nasals, only [n̥] is phonemic," Stefán Ingi (talk) 11:34, 11 November 2005 (UTC)

How about "Of the voiceless nasals, only [n̥] occurs in word-initial position"? AFAIK that is true and it avoids the POV complications of a phonemic analysis. --Angr/_{tɔk tə mi} 11:46, 11 November 2005 (UTC)

That is indeed a true statement and one which I don't think anyone would dispute. But I still don't think we should shirk away from a phonemic analysis. I think even a generative phonologist, when pressed, would admit that "on the surface level" there is a phonemic distinction between ɲ and ɲ̊. But the article should certainly make clear that there are alternative interpretations possible for how to derive this surface level from underlying forms and phonological processes or morphophonemic principles etc. - Haukur Þorgeirsson 12:08, 11 November 2005 (UTC)

Is the problem with það the fact that it's a function word rather than a lexical word? If so, maybe orð "word" is a better example of word-final ð. On the other hand, if ð is always partially devoiced word-finally, then we should just mention that. --Angr/_{tɔk tə mi} 11:53, 11 November 2005 (UTC)

Yes, it's a very common function word and that makes it a somewhat troublesome example - especially as regards the length of the vowel. I suggest "bað" ([paːð]) (e. "bath") instead. Yes, we should mention the partial word-final devoicing. Personally I doubt there's anything partial about it - when I say "bað" in isolation all I hear is [paːθ]. But if the word is followed by a voiced sound (such as a vowel) it comes out voiced. If I say "ég fór í bað í morgun" (I had a bath this morning) then it comes out as [paːð]. - Haukur Þorgeirsson 12:08, 11 November 2005 (UTC)

Just so you know where I'm coming from I once had a TA at a linguistics course at the University of Iceland who told us, the tender young souls he was entrusted with, that /i/ and /y/ were in fact still two different phonemes in Modern Icelandic - the contrast between them was simply completely neutralized. If the word "phoneme" is to have any meaning beyond the word "letter" then this is a very weird theory indeed. I've since been very skeptical of elaborate theories about underlying representations - most of them seem to me to be attempts to explain diachronic changes synchronically. - Haukur Þorgeirsson 12:20, 11 November 2005 (UTC)

It's true there are some phonological analyses that are excessively abstract, positing phonemes that never surface or that never surface as distinct from other phonemes. But it's incorrect to assert that all generative phonology does that. Nevertheless there are often good structural reasons for assuming a somewhat abstract underlying representation. If the voiceless nasals m̥ ɲ̊ ŋ̊ are phonemes, why are they prohibited from word-initial position when n̥ is allowed there? If /c/ and /k/ are different phonemes, why are the clusters /pj/ and /tj/ allowed, but */cj/ and */kj/ aren't? Positing underlying /pj tj kj/ but saying that /kj/ surfaces as [c] brings symmetry to the system. I'm not saying the abstract analysis is therefore necessarily correct, I'm just saying these aren't trivial questions and the more abstract analysis is not absurd. --Angr/_{tɔk tə mi} 14:14, 11 November 2005 (UTC)

I understand the ideas and I think they should be fairly represented (and I'm glad you're here to do so), I just personally disagree with them. If this beautiful symmetric underlying system is what you bring the uninformed reader you're not necessarily doing her any favors. You've probably smoothed over a lot of strange things and irregularities that make the language unique and interesting. When I'm learning a new language I'd like to get the straight dope about a defective distribution of a phoneme. And if there are known historical reasons for this defective distribution I might like to know those too. What I don't need is some esoteric theory where there isn't any defective distribution because some wise guy who sleeps with SPE under his pillow has explained it away. It's very interesting that unvoiced m is, in the Prague school sense, a phoneme in Icelandic and if you, like Eiríkur Rögnvaldsson, present the reader with a picture of the Icelandic phoneme system where unvoiced m doesn't appear you're just obscuring information. (I use 'you' as an indefinite pronoun.) Sorry for the rant, I know people can legitimately disagree. I might even be arguing against a strawman - I don't have ER's book at hand and I'm not 100% sure how he handled unvoiced m in particular. - Haukur Þorgeirsson 14:36, 11 November 2005 (UTC)

I'm glad you're using "you" as an indefinite pronoun, because I personally also am rather in favor of a more concrete view of phonology. For Icelandic in particular, I'm quite ready to believe that /c/ and /k/ are separate phonemes, but I'm much more skeptical regarding [m̥] since AFAIK there aren't even any surface environments in which [m̥] and [m] contrast, are there? If not, why would [m̥] be considered a phoneme even by the Prague school? --Angr/_{tɔk tə mi} 15:16, 11 November 2005 (UTC)

There are minimal pairs like skemmt/skemmd (neutral singular and feminine singular of 'skemmdur', "damaged") which are distinguished in pronunciation only by the voicing of the m. Historically, of course, the difference was in the voicing of the plosive but that distinction was lost centuries ago. - Haukur Þorgeirsson 15:20, 11 November 2005 (UTC)

Oh okay. --Angr/_{tɔk tə mi} 15:23, 11 November 2005 (UTC)

Is the table of consonants of phones or phonemes? If the latter, please point this out. While there are always disagreements about the amount of phonemes in a language, we can't duck the issue entirely by just saying that they're phones. If there are controversies and disputes among linguists, explain these rather than going for hyper-neutrality.

Peter ^Isotalo 13:47, 19 November 2005 (UTC)

It's phones, the voicing of the dental and velar fricatives is not phonemic so θ vs. ð and x vs. ɣ do not have minimal pairs. There is also a question (I guess) about how to classify the palatal and velar fricatives. The palatal fricatives can both occur word-initially and the voicing is phonemic there. In other locations the unvoiced palatal fricative can not occur and the voiced palatal one and the two velar ones can only occur in complementary positions, I guess that makes three phonems, two palatal and one velar. Everything else is phonemic. But we could say a lot more about Icelandic sounds, it's a lot of work though and I don't have any references available. But I know a lot of words in Icelandic so here are minimal pairs for everything (except the velar fricatives):

Correction: the palatal nasals are allophones of the velar ones occuring before a palatal stop. Also, I think that the velar fricatives can only occur in positions where velar stops can not occur so it might be that there are no velar fricative phonems. Somebody more knowledgeable please chech this all. Stefán Ingi 11:04, 20 November 2005 (UTC)

I think one velar fricative phoneme is present in the normal analysis, you can't really treat them as allophones of velar stops because of minimal pairs like ragur/rakur. You can, however, treat them as allophones of each other. I think everything else you said is also pretty standard - even if it looks like you figured some of it out by yourself :) - Haukur Þorgeirsson 12:55, 20 November 2005 (UTC)

Minimal pairs

Plosives: bóla [pouːla] "spot (on skin)" póla [pʰouːla] "poles" dóla [touːla] as in "dóla sér" "hang around" tóla [tʰouːla] "tools" gjóla [couːla] "light wind" kjóla [cʰouːla] "dresses" góla [kouːla] "howl" kóla [kʰouːla] "cola"

Nasals: geymd [ceint] "stored (feminine)" geymt [cein̥t] "stored (neuter)" leynd [leint] "kept secret (feminine)" leynt [lein̥t] "kept secret (neuter)" lengd [leiŋt] "lengthened (feminine)" lengt [leiŋ̊t] "lengthened (neuter)" banginn [pauɲcin] "afraid" bankinn [pauɲ̊cin] "the bank"

Fricatives vara [va:ra] "warn" fara [fa:ra] "go" Sara [sa:ra] proper noun þara [θa:ra] "sea weed" jóla [jouːla] "christmas" hjóla [çouːla] "cycle" hóla [houːla] "hills"

Other hráka [r̥au:ka] "spit" ráka [rau:ka] "creases" hláka [l̥au:ka] "defrost" Láka [lau:ka] proper noun

Stefán Ingi 20:07, 19 November 2005 (UTC)

Addition

I added the consonants ɬ and ɮ as both are found in the language (think long and hard about how you pronounce "sigldi" and "siglt"), although they probably don't form minimal pairs anywhere. Cheers Io

Hmm... So 'sigldi' and 'vildi' don't rhyme? I suppose this varies depending on the speaker. - Haukur Þorgeirsson 20:12, 26 November 2005 (UTC)

It does, but in my pronuncuation they don't.

Sorry, I forgot to sign off. Cheers Io 20:17, 26 November 2005 (UTC)

Okay - but what's with the glottal sound? :) - Haukur Þorgeirsson 23:12, 26 November 2005 (UTC)

Also, my feeling is that most Icelanders would rhyme sigldi and vildi. If people agree with that, then perhaps it would be more appropriate to mention the alternative pronounciation in the comments rather than including the symbols in the table. Stefán Ingi 23:28, 26 November 2005 (UTC)

The glottal was my mistake and has been removed. Icelanders would rhyme "sigldi" and "vildi" if it says so in a poem, but the sounds are not the same to my ears in everyday speech, although this would have to be confirmed by phonologists. (As I remember, it has but too vaguely to name a source.) Somebody said that one own's ear applied to one own's speech is the poorest phonetic guide of all. He may have been right. The table is about existing sounds in Icelandics, regardless of whether they're allophones or not, so if the sounds actually do occur, they belong here. Cheers Io 15:05, 27 November 2005 (UTC)

OK, I misunderstood your point. If there is a difference in everyday speech then of course the sounds should be in the table. I'd be curious to have this confirmed (I have thinking about it to much so I can hear it either way now), possibly the new book which was published a few days ago would be useful here. Until then, we can go around and ask people to read the sentence Yrðir þú ánægðari ef ég veldi/velgdi grautinn? (or maybe not, maybe the sound would only occur in the cluster gld and not in lgd). Anyway, thanks for the clarification. Stefán Ingi 16:46, 27 November 2005 (UTC)

After jogging my memory quite a bit, I believe Höskuldur Þráinsson mentioned it someplace, but that's as precise as I can make it. Perhaps someone could mail him? (Preferably someone with at least minimal linguistic credentials, which excludes me. :) Cheerio Io 17:06, 28 November 2005 (UTC)

PS: As for the clusters gld and lgd, they sound the same to me, for instance fylgdi and sigldi. Io 17:14, 28 November 2005 (UTC)

I had a look at the first volume of Íslensk tunga ISBN 9979219009 today. From the brief time I had with it, it looked very good. There is a table of the Icelandic consonants there and it does not include the lateral velars. So I have removed them from the table. The glottal stop is included. Stefán Ingi 01:50, 24 December 2005 (UTC)

Then you must have looked only at the table on page 161. Read the top of page 160. The lateral velars are in, if we intend the table to represent all actually occurring sounds. I'll take the liberty of reinserting them. I agree that Íslensk tunga appears to be a remarkable work. Cheers and gleðileg jól Io 16:29, 25 December 2005 (UTC)

Sorry, I overlooked the table earlier on, which is probably the one you had in mind, and I also overlooked that you don't have the book anymore. My apologies. Would it be OK with you if I sort this out to the best of my ability with citations and then we can discuss the matter? Cheerio Io 17:17, 25 December 2005 (UTC)

I would be very pleased if you would do this, especially since you have the book, in fact I was only able to have a quick look at it in the bookshop when I was doing the last of my christmas shopping. I'm sorry I removed information based on this brief and inaccurate look. I highly value your input. Ég óska þér sömuleiðis gleðilegra jóla, en núna ætla ég að fá mér hangikjöt. Stefán Ingi 18:20, 25 December 2005 (UTC)

It turned out that I had got a couple of things wrong. But I'm finished for now pending criticism. Verði þér að góðu. :) Cheerio Io 18:24, 25 December 2005 (UTC)

Removal

I also took the liberty of removing the reference to High Icelandic, as High Icelandic has very little to do with the Icelandic language as is. Feel free to restore the link, if you wish. Cheers Io 20:05, 26 November 2005 (UTC)

You're right. Though I'm fine with having an article on it we must be careful not to imply that the phenomenon is more important than it is. - Haukur Þorgeirsson 20:12, 26 November 2005 (UTC)

I noticed that the link to High Icelandic just reappeared in the article. Honestly, I know very little about Icelandic at all, let alone its relationship to "High Icelandic", but, at any rate, I feel that it's reference is somewhat awkward in the flow of the article. It also seems to contradict the statement in the article that follows, because the text reads: "An extremely purist form of Icelandic is High Icelandic or Háfrónska. / Icelandic does not have any dialect differences that can cause misunderstanding." To me, one sentence says that there is a "different form" of Icelandic and the next says that there are no different forms (dialects, rather) to cause confusion... this only causes me confusion as someone who knows little about either! However, I do think the reference probably has a place roughly where it is now. I'd like to propose a sentence something like the following: "Although Icelandic does not have any widely recognized dialectical variants, the language does have a modified ultrapurist form -- called High Icelandic or Háfrónska -- which has not gained much popularity as yet." or something similar. This sentence gets rid of some of the awkwardness of the two seemingly unrelated potpourri facts in the article, and the only reason I have not taken it upon myself to make these changes is that I am not sure of their wording because I really know nothing about either Icelandic or "high icelandic" and I'd rather not mislead readers... thoughts, anyone? Jxn 05:24, 2 December 2005 (UTC)

Also, I'd like to propose the removal of the high icelandic offsite link in the "external links" section because it almost caused me a little confusion when trying to research the language more. Yes, it belongs in the High Icelandic page, but not here. Edit: Nevermind, I removed it myself. Revert if appropriate. Thanks Jxn 05:32, 2 December 2005 (UTC)

Just to clarify this, "High Icelandic" is as far as I know a hobby of two individuals and not realy a language, since no one actually speaks it. It is simply a webpage with some creative word lists that relate to Icelandic.

--Sindri 13:37, 22 December 2005 (UTC)

Glottal stop

Just out of curiosity, has it been investigated whether the preaspiration could be interpreted as a glottal stop or not? Also, if a sentence starts with a vowel (or any part of a sentence after a pause starts with a vowel), is there a glottal stop present? I'm reminded of the troubles Germans learning French have dropping their glottal stops, which French does not possess. (The canonical example is "ein Ei" vs. "un œuf".) Cheers Io 19:23, 10 December 2005 (UTC)

I added the glottal plosive out of my own feeling for the language, as I did with the velar lateral approximates approximants. If anyone disagrees, or even better, can point to to studies to the contrary, feel free to remove it. Personally, I'm not a linguist, so I relied on my ears and vocal cords to include the sounds, but try saying "einn er" with emphasis, first as it comes to you naturally and then try saying it without a stop. Cheerio Io 20:40, 10 December 2005 (UTC)

As an additional Christmas present this year, Kristján Árnason agrees with me regarding the glottal stop (ʔ). In Íslensk tunga I. Hljóð he says: The glottal stop is more ubiquitous in Icelandic speech than often assumed. In addition to playing a part in the formation of some stops, it often occurs at the beginning of a stressed syllable beginning with a vowel. It is extremely satifying to have one's suspicions confirmed. Cheerio Io 22:42, 24 December 2005 (UTC)

Excellent. You should add the work to the references section of the article. - Haukur 16:46, 25 December 2005 (UTC)

Will do, when I've sorted out the velars. Both me and everybody and else (including the authors of the IPA) seem to be a bit confused on the subject. In the meantime I heartily recommend the entire work (although I got the impression that Hölskuldur Þráinsson, who wrote about syntax, isn't quite sure whether to be a Chomskyist or not :) ). In any case, all authors strive as they can for a balanced point of view. Bear in mind, that I only got the work yesterday and it runs to about 1700-1800 pages. :) Gleðileg jól Io 17:17, 25 December 2005 (UTC)

I haven't seen it yet but I'm in Iceland now and I've found Eiríkur Rögnvaldsson's phonetics book. Will check it when I have time :) Gleðileg jól! - Haukur 18:39, 25 December 2005 (UTC)

Velarized, alveolar, lateral approximants

The table of consonants now contains four entries in the field "Alveolar, lateral approximants". The relevant thing is (see Íslensk tunga I p. 160), that the Icelandic lateral approximates l̥ and l may become velarized. The resulting sound is quite different from the original. This is not a very common, but a fairly regular phenomenon, namely when gl or lg are followed by a consonant. The IPA-convention for velarization is to put a small latin gamma superscript on the velarized sound (e. g. l^ɣ). I decided to put those two additional sounds in the alveolar field, since that should probably still be considered the place of articulation. Personally I feel that those sounds have a place in the table itself, but I'll concede, that it might be a topic of discussion, whether to have them in the table or mention them in explanations below. Cheerio Io 18:13, 25 December 2005 (UTC)

Preaspiration, again

It turns out that your example (Haukur), tott, really is [t^hɔht], but according to spectral analysis the word is not entirely symmetrical, although the difference, when played backwards, may not be discernible to the normal, human ear. Cheers Io 18:48, 25 December 2005 (UTC)

High Icelandic

I removed the sentence about High Icelandic from the article but Timbur-Helgi reinserted it. My view is that although the project deserves its own page on Wikipedia it is not important enough that it should be mentioned in this article, basically because nobody speaks it. Other opinions? Stefán Ingi 01:04, 26 December 2005 (UTC)

Yes, it seems too specialized for a mention in an overview article like this one. On the other hand a short mention would not be amiss in an article about Icelandic language purism and neologisms in general. Maybe Timbur-Helgi can write such an article. - Haukur 01:07, 26 December 2005 (UTC)

My personal opinion is that the words "artificially created" automatically disqualify High Icelandic from being mentioned in the context of dialects, which was where Timbur-Helgi put his addition. A mention of High Icelandic in the body of the article might become acceptable as it expands, but then in its proper place, not as a dialect. Cheers Io 15:07, 27 December 2005 (UTC)

A good idea. I will write an article about Icelandic language purism. I was surprised to find that the German wikipedia has one (see Ísländischer Sprachpurismus). I will base the article on the book 'Íslensk málhreinsun' by Kjartan G. Ottósson, which is, for all I know, the most comprehensive source on the subject.

Spoken in

Where would you say Icelandic is spoken. Currently the infobox says Iceland, Denmark, Norway, parts of Canada (Manitoba) which I think is a bit of a stretch. Maybe just Iceland would be better. Stefán Ingi 15:24, 27 December 2005 (UTC)

Indeed. It's spoken in Iceland and then wherever Icelanders happen to be - a lot of us are in Scandinavia at any given time, certainly. I'll support listing only Iceland. - Haukur 15:46, 27 December 2005 (UTC)

I mostly agree with Haukur, but parts of Canada should probably be kept in. It's still alive there (barely) as a medium of communication with a history, as opposed to the Icelanders elsewhere, who come and go. Cheers Io 22:36, 27 December 2005 (UTC)

I have removed Denmark and Norway from the list. I can see the logic in keeping Canada in but sadly I believe that it is dying out there so at some point in the future it will have to be removed. Stefán Ingi 13:20, 28 December 2005 (UTC)

<v>

It seems to me that <v> often becomes [V\] between a consonant and certain vowels, such as in "tvö" and "svo". I have never, however, seen anything that claims [V\] even exists in Icelandic. Am I mishearing things? --Crushti

What is [V\] in IPA? I've looked for it in the SAMPA article and I found nothing like. Ciacchi 20:53, 27 January 2006 (UTC)

I assume the contributor above meant [v\] (otherwise written [P]) - the labiodental approximant. But I can't comment on whether it's a phone in Icelandic. --Dave ~ (talk) 22:47, 27 January 2006 (UTC)

Yeah, my apologies, I meant [v\]. --Crushti 23:37, 27 January 2006 (UTC)

Pronunciation of g

I have a lot of questions about how the letter g is pronounced in Icelandic. The different pronunciation guides floating around the Internet are often contradictory and seldom use IPA transcriptions, and the current article does not have enough information about this either. So far, using this article and relying on the pronunciations given for Icelandic words in Edmund Gussman's Phonology: Analysis and Theory (Cambridge Univ. Press, 2002), I've come up with some guesses about how to pronounce g:

• word-initially as [k], or if followed by e or i [c], [c] also in the combination gj
• intervocally as [ɣ], at least when the following vowel isn't e or i
• in the combinations gl, gn, gg, lg, and ng as [k] (or in the latter cases, [c] if followed by e or i)
• in combinations such as gr and gð as [ɣ] (the word vogris [voːkris] seems to be an exception)
• in the combination gs as [x]
• word-finally as [x], at least when a single g follows a vowel

Are these guesses correct? Or at least, if I follow those rules, would the resulting pronunciations be acceptable enough with only a few exceptions?

Also, could anyone tell me how to deal with the some cases that don't fit into any of the cases above? For instance:

• When g occurs intervocally but the following vowel is e or i, is it still pronounced as [ɣ]?

• No, when the following vowel is i you get a diphthong with [i] as the second element; "magi" is [mai:I] for example. Intervocalic g with the following vowel as e doesn't occur except in combinations and loanwords. The word "lager" is, counterintuitively, pronounced with a [ɣ]. Haukur

• When g occurs at the end of the word but follows another consonant, as in Ingibjörg, is it still pronounced as [x]? Are there other cases, like Icelandic words ending in gg or lg? Oh, and word-final g in ng is pronounced as [k], right?

• The g in Ingibjörg is a [k]. So is the word-final one in the ng-combination. Haukur

• When is the g in ng (or for that matter, the k in nk) not pronounced? The words lengd and svangt seem to have [ŋ] directly followed by [t] or [tʰ]. The article currently says only that "[ŋ] appears (in addition to before velar stops) also before [l] and [s] through the deletion of [k] in the consonant clusters [ŋkl] and [ŋks]." But does the deletion always occur? Wouldn't the g in tungl be pronounced?

• The g in tungl only surfaces as [k] in a dying dialect. Haukur

• What is it with guð being pronounced as [kvʏð], as in the pronunciation guide for Björk Guðmundsdóttir, ['pjœr̥k 'kvʏðmʏnstoʊhtɪr]? Is there a rule for the insertion of [v]? Does this occur in words like Guðrún? Gunnar? Sigur? Or is the insertion of [v] optional so that ['kʏðmʏnstoʊhtɪr] is a perfectly acceptable pronunciation?

• The word "guð" (god) is irregular and always pronounced [kvʏ(ː)ð], so is any combination with it. Haukur

• What is the pronunciation of g in other letter combinations, such as in Ásgeir, Hallgrímsson, and Vigdís? Is there a general rule to predict how g would be pronounced in all these different combinations?

Finally, on an unrelated note, is the d sound deleted in the names Davidsdóttir and Rögnvaldsson? It certainly seems to be for Guðmundsdóttir according to the pronunciation given. Is there a rule for this, such as that d is not pronounced in the combination ds?

I realise I've asked a whole lot of questions, but I'd appreciate any help, even little things like adding the pronunciations for the words and names I asked about. --Iceager 15:30, 7 March 2006 (UTC)

You're right on most counts, you've clearly done a lot of research and have a solid background in phonetics. I added answers to a few of your questions, I hope others will step up to help. Haukur 15:52, 7 March 2006 (UTC)

I'll also take a stab, but intersperse my additions between the numbered items (forgive me). I'll say, this is a fine effort, since g is probably the most complicated letter in the Icelandic alphabet. Io 20:08, 9 March 2006 (UTC)

If I take a stab at it somebody will come along shortly and correct me.

1. word initially 'g' is an unvoiced plosive, a palatal one [c] if followed by j, e, é, i, í, y, ý or æ but velar [k] otherwise (gaf, gefa)
2. also 'lg', 'ng' and 'rg' follow the same pattern (this can only occur after a vowel) (belgur, löng, berg)

   This seems to be an unnecessary item or at least should be reworded.

   Yes, reading it again now, it doesn't make much sense. Perhaps: the 'g' in 'lg', 'ng' and 'rg' is pronouced [k] or [c] (following the above pattern) if followed by a vowel or word finally. If followed by a consonant the 'g' in these combinations is silent but in the combination 'ng' the 'g' colors the 'n' to [ŋ] and the 'g' in 'lg' colors the 'l' to [l^ɣ].

3. word finally 'gg' is an unvoiced velar plosive [k] (bygg)
4. between vowels, 'gg' is an unvoiced plosive, a palatal one [c] if followed by j, e, é, i, í, y, ý or æ but velar [k] otherwise (leggur, leggir)

   Is there an example of g followed by é?

   Hmm, perhaps the name of the letter, "gé". I'm not sure though.

5. word finally 'g' is an unvoiced (half voiced) velar fricative [x] (lag)

   unless followed by a word starting with a vowel.

6. between a vowel and an unvoiced consonant 'g' and 'gg' is an unvoiced (half voiced) velar fricative [x] (lagt, byggt)
7. after a vowel and before i 'g' is a voiced palatal approximant [j]
8. after a vowel and before anything else 'g' is a voiced velar fricative [ɣ]

   If followed by a consonant, it is unvoiced.

   :PS: It also depends on the consonant. Compare lagt [laxt], hagl [hakl ̥] and lagði [laɣði].

   I've often felt dumb, but rarely more than now. Disregard the comment I struck. I didn't read the article closely enough. With blushing cheeks and a hanging head. Io 18:00, 11 March 2006 (UTC)

   We need to unstrike the hagl [hakl ̥] example, it's not covered by anything else I said. So, in the combination 'gl', 'g' is a velar plosive.

9. 'ng' followed by a plosive and [l] is just pronounced [ŋ] (lengt, lengd, tungl)

   Perhaps this is getting reasonably accurate now but I have made such a horrible mess of writing it so it will be very difficult to get any useful information out of this. Stefán Ingi 23:47, 11 March 2006 (UTC)

Additionally, 'guð' is an exception, pronounced [kvʏð] although there is nothing to suggest the [v] sound, the same goes for anything starting with 'guð' such as 'Guðrún'. 'Ásgeir' 'Hallgrímsson' and 'Vigdís' are combined words, Ás-geir, Hall-grímsson, Vig-dís and the 'g' in each of those is pronounced as if these were separate words. For words such as 'legi' see rule 7. For 'Ingibjörg' see rule 2 for both g's. 'Davídsdóttir' is a corruption, it should be 'Davíðsdóttir', the 'ð' is probably usually silent, but pronounced [ð] in careful speech. I cannot make my mind up whether the 'd' is silent in 'Rögnvaldsson'. In general the combination 'ds' usually has a silent 'd' in everyday speech but not in careful speech. Stefán Ingi 16:09, 7 March 2006 (UTC)

In my speech, careful or not, neither the [ð] nor the [d] are silent. But I admit that I may not be typical. Cheers Io 20:11, 9 March 2006 (UTC)

I've unstruck hagl. Apparently only the original author can strike or unstrike his comments according to Wikicode. You could have done it without complaints from me. As it happens, I posted a message on the originator's user page and offered him to undertake some research on the matter. If noone objects, I will do so in the next few days and post a pre-pre-pre-definitive list. Cheers Io 20:36, 13 March 2006 (UTC) (Why does g have to be so bloody complicated? Those of you who learned German with the aid of the Little Yellow Hen, will remember the chapter about Icelandic pronunciation, where g had at least 8 varieties, including one, where it disappears altogether.)

More on g

This g-stuff is even more complex than I imagined before thinking about it. Do we have a chance of setting up a definite set of rules, or must we be content with saying, "well, that's how we pronounce it" in every individual case? Cheers Io 20:37, 9 March 2006 (UTC)

The first thing might be to get a grip on where 'g' can occur. Let's try and make a list.

Word initially, and then rule 1 above is accurate.

Otherwise we can have the following combinations, following a stressed vowel:

1. g (lag), gt (lagt), ga (laga), gu (legur)
2. gi (lagi)
3. gg (bygg), ggj (byggja), ggt (byggt)
4. gl (hagl)
5. rg (berg), rgs (bergs), rgt (byrgt), rgi (byrgi), rgj (byrgja)
6. ng (löng), ngj (lengja), ngt (lengt)
7. lg (belg), lgu (belgur), lgja (belgja), lgt (belgt), lgu (elgur), lga (elga)

I cannot think of any more now, and I cannot see any way of organising this neatly. :( Stefán Ingi 00:04, 12 March 2006 (UTC)

I have more time on my hands now than you have. Give me a day or two to come up with a system — after all we were probably both brought up in the the German systematic way — one way or other, regardless of whether you chose French or German in school. :) Io 20:36, 13 March 2006 (UTC)

I don't think ggt (byggt) belongs there with gg and ggj. Pési 22:35, 8 April 2006 (UTC)

Sounds like the pronunciation of "ough" in English :-) Nyttend 03:43, 4 October 2006 (UTC)