A little Icelandic phonetics

« previous post | next post »

Some people are apparently still puzzled by the pronunciation of Eyjafjallajökull. So let's take it a bit at a time. This morning, we'll cover the unexpected (to non-Icelanders) pronunciation of the 'll' at the very end of the word. (I warn you in advance that I don't know anything about Icelandic, I'm just exercising some generic phonetics-fu with a little help from my friends…)

Here's a fragment of a "hi I'm X" dialog from an Icelandic Online course:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

XAVIER:      Já! En hvað heitir þú?    "Yes! But what is your name?"
KATRÍN: Sæll, ég heiti Katrín. "Happy [to meet you], my name is Katrín."
XAVIER: Sæl. "Happy [to meet you]."

Let's zero in on sæl/l, which means something like "happy" or "fortunate". It's apparently used in various formulas of greeting or parting, roughly equivalent to "pleased to meet you" [update: though the adjective agrees with the addressee, with a meaning more like "you come (or may you be) happy/fortunate/blessed"]. Here the formula is abbreviated as the adjective alone.

When Xavier uses it, it's in a feminine form [agreeing with the addressee, Katrín], spelled "sæl" in the standard Icelandic orthography:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

The 's' is unproblematic, and the 'æ' is a rising diphthong (IPA [aj]) rather like the vowel in English why. But the final orthographic 'l' is pronounced as a voiceless lateral fricative, written in IPA as [ɬ]. In a bigger font, that glyph looks like this:

[ɬ]

To make the corresponding sound, put the tip of your tongue against the roof of your mouth as if you were going to say "Language Log", but then just blow some breath out instead. [Note: I agree with John Wells (in disagreeing with the current Wikipedia entry) that this is a voiceless lateral fricative rather than simply a voiceless lateral, which would be written in IPA as an 'l' with a hollow dot underneath it.]

For those who appreciate such things, here's a spectrogram of Xavier's "Sæl." (Click to embiggen, as usual.)

So if you're male, and want to say the Icelandic equivalent of "enchanté" when meeting someone, just use the English word size, substituting [ɬ] for the final [z].

But when Katrín says the same thing, she uses the masculine form [referring to Xavier], spelled "sæll". And that orthographic double 'll' is pronounced [tɬ]:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

To make this sound, put your tongue tip in position to say "talk", let a little air pressure build up, and then release it into the [ɬ] sound that you learned to make just a little while ago.

(In that New York Times article, Andy Newman and Bao Ong describe this by writing that "the 't' at the end kind of sticks for a second and pulls away with a hint of a glottal 'l'". But it's in the nature of every [t] to "kind of stick for a second", and there's nothing "glottal" about the [l], except for the fact that the glottis does remain attached to the rest of the vocal tract during production of the sound.)

Here's Katrín's spectrogram:

There you can see (as well as hear) the honking big "stop gap" (i.e. silence) between Katrín's 'æ' vowel and her lateral fricative. This is an instance of the "pre-stopped lateral" mentioned in my earlier post.

You can also see (and hear) that her vowel is a lot shorter than Xavier's. Vowel length is said to be determined by context in Icelandic, "assigned by [a] very late rule", as Steve Anderson explains in "An outline of the phonology of Modern Icelandic vowels", Foundations of Language 5: 53-72, 1969. But I can't see how to apply his rule to make the vowel long in sæl but short in sæll, so either he left something out, or I don't understand the rule, or something else is going on here. [Update: actually, as discussed below, the rule obviously does the right thing, given that it applies at a "very late" point where the final /l/ is already a voiceless lateral fricative. I warned you that I didn't know anything about Icelandic; but in fact, "something else is going on here" also seems to be a sentiment frequently encountered in examining Icelandic phonetics.]

OK, one last point. The word Eyjafjallajökull has two instances of orthographic 'll'. The final one is rendered as [tɬ], just as we've been discussing. The fact that 'll' is a "pre-stopped lateral" is true of both instances. But the transformation of the final /l/ into a "voiceless lateral fricative" is due to its word-final position. The 'll' in "fjalla" is just [tl], as you can hear in the hyperarticulated pronunciation by Begga (?) from the Icelandic consulate in New York, which I took from Andy Newman and Bao Ong, "Iceland Volcano Spews Consonants and Vowels", NYT 4/16/2010:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

You can see the difference in this spectrogram, where there is a clear phonetic [l] segment at the start of the second syllable of fyalla:

OK, enough for today. Maybe tomorrow we'll take up the performances by Jóhann Heiðar Árnason and Rögnvaldur Ólafsson, where Something Else Is Going On™:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

[Thanks to Joan Maling for Icelandic guidance, though she of course is not responsible for the mistakes that I've probably made here.]

Share:



54 Comments »

  1. Johan Anglemark said,

    April 19, 2010 @ 8:34 am

    I just have to point out the wonderful fact that "sæl(l)" is of course historically the same word as English "silly", which also used to mean "happy, blessed" a few centuries back.

    [(myl) Nice point. I've often used this example in explaining to people why a change in word meaning is not the end of the world, or even the end of the language. Thus in "Nonplussed about nonplussed", 8/6/2008:

    Q: Is it possible for a word to become so commonly misused that the new (wrong) definition becomes acceptable? Has any word like this ever had its new meaning included in a dictionary?

    A: Yes, this happens all the time, though perhaps the loaded term "misuse" should be put aside for the moment.

    Working forward from 1200 to 1600, a certain word went through something like the following sequence of meanings (as given in the OED): "Happy, blissful; fortunate, lucky, well-omened, auspicious"; "Spiritually blessed, enjoying the blessing of God"; "Pious, holy, good"; "Innocent, harmless"; "Deserving of pity or sympathy; pitiable, miserable, ‘poor’; helpless, defenceless"; "Insignificant, trifling; mean, poor; feeble"; "Frail, worn-out, crazy"; "Foolish, simple, silly".

    The word? Silly.

    ]

  2. language hat said,

    April 19, 2010 @ 8:44 am

    Nothing to do with Icelandic, but do you really pronounce scythe with a voiceless final? It's not a word that you hear often, so one doesn't get many opportunities to check one's own usage, but I've always said it with voiced /ð/.

    [(myl) You're right. Fixed now.

    On reflection, I've changed the example to "size", since people pronounce scythe in a variety of ways, including without a final consonant, and others are likely to be uncertain about how to pronounce the name of this no-longer-everyday implement.]

  3. Colin John said,

    April 19, 2010 @ 8:47 am

    So the Icelandic single l in 'Sæl' is the same as the Welsh 'll'.

  4. Andrew said,

    April 19, 2010 @ 9:00 am

    The word sæl(l) isn't analogous to enchanté(e) — it agrees with the addressee, rather than the person who's speaking. So the masculine form, sæll, is used to address men, and the feminine sæl to address women. It's an ellipsis komdu sæl(l), literally something like "(you) come happy".

    [(myl) Oops. Told you I didn't know anything about Icelandic. I'll revise (i.e. reverse) the exposition appropriately.]

    I don't really know anything about Icelandic phonology either, but as far as I'm aware, vowels are generally long when followed by a single consonant (as in "sæl"), but short before geminates and most consonant clusters (such as [tɬ], I guess). Don't take my word for it though!

    [(myl) The cited paper by Anderson says that

    On reflection, this does seem to predict the observed pattern of length in sæl/l, given that the final /l/ becomes a voiceless fricative before the rule applies (which it would, since the rule is "very late").]

  5. Trond Engen said,

    April 19, 2010 @ 9:04 am

    I don't know much Icelandic either, and risk a spectacular fail, but haven't you switched the masculine and the feminine here? I'd believe the double l to be an assimilation of the ON/OIc masculine nominative ending -r (which of course in turn is from *-z, from *-uz, from IE *-us). I understand the expression as a wish for happiness (but now, of course, just a hi).

    [(myl) You're right -- the text was sexually inverted, but is now (I hope) fixed.]

  6. Helgi Briem said,

    April 19, 2010 @ 9:13 am

    You can also see (and hear) that her vowel is a lot shorter than Xavier's. Vowel length is said to be determined by context in Icelandic, "assigned by [a] very late rule", as Steve Anderson explains in "An outline of the phonology of Modern Icelandic vowels", Foundations of Language 5: 53-72, 1969. But I can't see how to apply his rule to make the vowel long in sælbut short i sæll, so either he left something out, or I don't understand the rule, or something else is going on here.

    The rule that modifies the length of vowels in Icelandic is that if the vowel is followed by a double (i.e. "ll") or two (i.e. "rt", "fl", "st" etc) consonants, the vowel is shortened.

    I say this only as an Icelander, not a linguist.

    [(myl) Steve's rule also makes the vowel long if the second consonant is a sonorant continuant.]

  7. Amy West said,

    April 19, 2010 @ 9:15 am

    I don't do modern Icelandic, I do Old Norse. But, many medievalists working on Old Norse/Old Icelandic materials use a modern Icelandic pronunciation — instead of a reconstructed historical ON pronunciation. (I don't: I stick to the best-guess historical)

    The very brief guide to modern Icelandic pron. in the standard _New Intro to Old Icelandic and Norse_, gives two pronunciations for [ll]: (1) as in English "leaf" when it's before consonants other than _n_ or _r_ as in "illt" (bad); (2) as in English badly (both the d and l) when it's between vowels or before _n_ or _r_ as in "hellir" (cave). I know that colleagues who use the modern Icelandic pron. pronounce names like "Egill" as "ey-id." So, in "Eyjafjallajokull," that first -ll- gets the -dl- sound, the final -ll, gets just that -d sound.

    [(myl) This advice doesn't accord with the pronunciations in the various recordings (by three different native speakers), which are consistent in having [tl] for the first 'll' and [tɬ] for the second one.]

  8. Adam said,

    April 19, 2010 @ 9:53 am

    How does the [ɬ] sound compare with the Welsh "ll"? (It sounds rather like it to me, but I admit I may not have a very good ear for phonetics.)

    [(myl) They're both described as "voiceless lateral fricatives". Since in addition to not knowing anything about Icelandic, I also don't know anything about Welsh, I can confidently assert that they're the same.]

  9. Army1987 said,

    April 19, 2010 @ 10:26 am

    The vowel length rule as given by the Wikipedia article, IMHO, sounds like "stressed vowels in open syllables are long" where what follows is a list of the consonant clusters allowed in syllable onsets, doesn't it?

    [(myl) The idea makes sense, but the facts are against it -- for example, vowels in front of a single word-final consonant, which by default would certainly count as closed syllables, are long.]

  10. Kobey said,

    April 19, 2010 @ 10:55 am

    re: pronunciation of scythe

    I definitely say it with a final voiceless [θ], and I have 'canadian' raising to confirm it — I say [sʌɪθ].

    [(myl) I think vary between [θ] and [ð] for the noun, with invariant [ð] for the verb. (Maybe some influence of bath/bathe?) Not that I've had many occasions to pronounce this word over the past few decades. When I was growing up, we had a scythe in the barn, which I used to use from time to time when some of the grass got too long for a lawnmower to manage it; and I recall a discussion with a neighbor child who pronounced it [saj], with no final consonant at all. But anyhow, the dictionaries all seem to be have [ð] or nothing.]

  11. Sili said,

    April 19, 2010 @ 11:13 am

    How does the [ɬ] sound compare with the Welsh "ll"? (It sounds rather like it to me, but I admit I may not have a very good ear for phonetics.)

    John Wells does speak Welsh.

  12. Joe said,

    April 19, 2010 @ 12:18 pm

    John Wells' Longman Pronunciation Dictionary does give an alternate pronunciation of "scythe" with a voiceless final consonant (although Cambridge prnouncing dictionary does not). And "Time's scythe" does appear in Shakespeare's Sonnet 12…

    [(myl) Nice to know that I'm not completely idiosyncratic.]

  13. Lane said,

    April 19, 2010 @ 1:14 pm

    You forget that "glottal", like "guttural" and "nasal", is layperson for "sound I don't like or can't make".

    [(myl) Good point.]

  14. Amy West said,

    April 19, 2010 @ 1:30 pm

    Well, the variance between what NION gives re: modern Icelandic [ll] and what you hear from the native speakers may be that the NION guidelines are just approximations for (British) English speakers.

    And I have a tin ear, so what I think may sound like that terminal /t/ isn't really.

    I don't do the modern Icelandic pron for Old Norse because that /tl/ or /t/ for [ll] just makes me crazy.

  15. Troy S. said,

    April 19, 2010 @ 2:05 pm

    Re: scythe Am I the only one whose instinct is to voice it when it's a verb and leave it voiceless when it's a noun?

    On a related note, I remember reading in one of my textbooks that those two sounds sounds are almost never phonemic in English, but except for noun/verb distinctions like that one, I haven't been able think of an example.

  16. Mona said,

    April 19, 2010 @ 2:20 pm

    First, I'm not a linguist, just a (non-native) speaker of Icelandic. That being said, a note on sæl(l): This greeting is actually a shortened version of "Vertu sæll!", which literally translates as "Be (you) blessed/fortunate". Another greeting used in exactly the same manner is "blessaður!" (fem. blessuð), which in turn is short for "Vertu blessaður", or "be blessed". In casual speech you mostly hear the even shorter version "bless!". In other words, not "you come happy", but "be lucky"

  17. naddy said,

    April 19, 2010 @ 2:50 pm

    On a related note, I remember reading in one of my textbooks that those two sounds [[θ], [ð]] are almost never phonemic in English, but except for noun/verb distinctions like that one, I haven't been able think of an example.

    Commonly cited minimal pairs: mouthto mouth, teethto teethe. A non-verb example is thighthy.

  18. Icelander said,

    April 19, 2010 @ 2:56 pm

    Vertu sæl(l) is not a greeting, but a farewell. Komdu sæl(l) is the greeting.

    Komdu sæl(l) ≈ [I hope you] come content/happy/fortunate/blissful
    Vertu sæl(l) ≈ [I hope you will] be content/happy/fortunate/blissful

    Then you can go even further and say:

    Komdu/Vertu sæl(l) og blessuð/blessaður ≈ Come/Be happy and blessed

  19. Mona said,

    April 19, 2010 @ 3:14 pm

    Icelander, thanks for the correction! Of course you are right. I typed too hastily and did not triple-check what I had written. I have also heard (read?) "Kom þú sæll, heill, og blessaður" (heill=healthy, whole), but that is going a bit poetic I guess…

  20. Bob Ladd said,

    April 19, 2010 @ 4:16 pm

    Another English th voicing minimal pair that involve neither archaic words nor noun-verb relations: either – ether, provided you say either with the vowel of beat rather than the vowel of bite.

  21. Pekka K. said,

    April 19, 2010 @ 4:20 pm

    A minor issue: In BeggaA.png, don't both ya and fja[t] have the same sound for y and j? Wouldn't it be consistent to use one symbol for both?

    [(myl) Indeed. I was clearly inadequately caffeinated, or perhaps excessively rushed, at the time.]

  22. dw said,

    April 19, 2010 @ 5:26 pm

    According to the Wikipedia Icelandic phonetics page, orthographic NN can also be pronounced /tn/ under certain circumstances. Maybe the same process at work?

  23. Nathan Myers said,

    April 19, 2010 @ 5:27 pm

    I think some genderitionary corrections remain to be performed in the paragraph after the first sonogram.

    Perhaps a good idiomatic English translation for "sael(l)" is "good on ya" ? I have only heard it in British English.

  24. Beth G. said,

    April 19, 2010 @ 6:45 pm

    @dw

    Not a native Icelander, but someone who's studied Old Norse and Modern Icelandic (and ftr, I've never met someone who spoke Old Norse with the "classic" pronunciation; all my professors pronounced it more or less as they would Modern Iceland).

    I was taught that orthographic double NN is pronounced [tn] when preceded by a historically long vowel (any vowel with an acute accent over it; ON had phonemic stress, MIce probably not) or a diphthong (well, an orthographic diphthong; [e:] and [o:] are generally diphthongized in current speech) and as [n] when preceded by a (historically) short vowel (orthographically: i, e, a, o, u, y). So steinninn 'the stone, masc nom sing' is pronounced [steitnɪn]. Word-finally, the /n/ is devoiced: steinn 'stone' is [steitn̥] and can even be dropped completely (with what sounds to my non-native ears as a compensatory non-release or gemination of the /t/).

  25. Michael said,

    April 19, 2010 @ 6:55 pm

    One minor point: The difference in voicing of the "l" in the "tl"-clusters is not necessarily due to the word final position of the latter one. Icelandic does have phrase final devoicing (therefore the "l" in "sæl" is devoiced when the word is pronounced in isolation), and this devoicing obviously applies when "Eyjafjallajökull" is pronounced in isolation.
    But if you consider the word "fjallganga" 'mountain climbing' the "ll" is pronounced with a voiceless lateral fricative, too, although it's not in wordfinal position. Icelandic just "hates" syllabic consonants and chooses to devoice sonorants if they "threaten to become syllabic". In "-fjalla-" the "l" is in the onset so it's safe.

  26. Jangari said,

    April 19, 2010 @ 8:12 pm

    Surely the final "ll" could be analysed as a laterally released /t/. In fact they both can; the only difference being there's a following vowel in the medial "ll" in -fjalla- for the stop to laterally release into. The final instance (as I observed in the post the other day) occasionally has an epenthetic vowel in some people's pronunciation, but not in others. That is, some pronounced it [kœtlə] it seemed to me.

    Thus it could be exactly what most RP English speakers tend to say for 'little'. The /t/ releasing into the /l/ (merely by dropping the sides of the tongue) can sound a bit like a velar sound. I never heard it much myself (being an Australian male who universally taps his intervocalic alveolar stops) until I started watching Top Gear and started noticing Jeremy's pronunciation.

    As for why it fricates a bit at the end of a word into [tɬ], well, my phonotactic knowledge isn't terribly good, but I'd want to claim that without a vowel (or syllabic /l/), the lateral release devoices, leaving a whole voiceless cluster (Unicode doesn't allow me to transcribe a voiceless lateral release, but imaging a "t" followed by a superscript "l" with a voicelessness circle underneath it). This cluster I reckon would be perceived as a lateral-alveolar affricate [tɬ].

    On [ð] versus [θ], they were historically allophonic in English and most Germanic languages, surfacing as voiced intervocalically, and voiceless elsewhere (I think). This accords with typical Germanic final devoicing (hund [hʊnt] but hunde [hyndə]). This was apparently why we get the noun/verb distinction with verbs ending with voiced [ð] and nouns ending with [θ], because verbs used to have another vowel (since dropped) allowing the fricative to voice.

    I don't quite understand though where the function words come from. All words with initial [ð] are function words (pronouns, determiners, demonstratives), basically. So: the, thy, thou, though, this ,that, them, they, thine, etc., all are voiced initially, but thing, thigh, through, thought, think, thin, thistle.

    Ah, another minimal pair (besides thy~thigh): this'll ~ thistle.

    However, as we are in the company of a great many brilliant phoneticians, and I'm just some lowly syntactician, I'll happy concede defeat on any of these matters.

  27. Icelander said,

    April 19, 2010 @ 8:24 pm

    Yes Mona, "Kom þú sæll, heill og blessaður" would be over the top. It's even hard to say it's poetic, as it becomes too staccato for a fluid phrase. "Komdu sæll og blessaður" has on the other hand been used very memorably in a lyrical sense. The song "Ofboðslega frægur" (overwhelmingly famous) with Stuðmenn really hammers it home. To be heard here: http://www.youtube.com/watch?v=F0oZyOdbjns

    Although on the subject of LL being pronounced as DL, it is not a 100% rule. If I take an example where the same letters have different pronunciation: "Ella" (a female name derived from Helen) is pronounced with the LL as in "HeLa cells". Long L followed by a short A. Ella (meaning otherwise) is pronounced "edla".

  28. Jangari said,

    April 19, 2010 @ 9:13 pm

    Icelandic just "hates" syllabic consonants and chooses to devoice sonorants if they "threaten to become syllabic". In "-fjalla-" the "l" is in the onset so it's safe.

    Aha! This is basically what I was getting at above when I said "without a vowel (or syllabic /l/), the lateral release devoices, leaving a whole voiceless cluster" (just without actually knowing any Icelandic phonotactics).

  29. Kenny V said,

    April 20, 2010 @ 12:22 am

    language hat: I pronounce scythe with a voiceless final.

  30. Joyce Melton said,

    April 20, 2010 @ 4:02 am

    The pair thin-then is phonemic for voiced, unvoiced labio-dental fricatives for me since Arkansawyers pronounce both vowels alike and do so in other pairs, too. Tin-ten, pin-pen — but not fin-fen since fen is a furrin word for a sort of loblolly.

  31. Jongseong Park said,

    April 20, 2010 @ 6:08 am

    Slightly off topic, but do most scholars of Old Norse use the Modern Icelandic pronunciation? If so, would they nevertheless use anglicized or continental Scandinavian pronunciation when they discuss Old Norse literature with people who don't speak Icelandic? I certainly learnt Egill as 'EGG-ill', not 'AY-yitl' in a general course on Old Norse literature at an American university.

  32. bfwebster said,

    April 20, 2010 @ 8:45 am

    My first thought listening to Begga's pronunciation was that this sounds like a quiet English phrase recorded and then played in reverse. My second was to burst out laughing at my hopes that I would be able to pronounce "Eyjafjallajökull". Someone needs to come up with a short, punchy nickname. ..bruce..

  33. Dan Lufkin said,

    April 20, 2010 @ 9:07 am

    And how does Keflavík become KePlavík?

    Just wondering. I find Icelandic pretty easy to read but very hard to understand when spoken.

  34. Army1987 said,

    April 20, 2010 @ 10:53 am

    Another thing which threw me off is…
    The “k” is softer than an English “k,” almost like a hard “g.”
    Apparently, "soft" means "unaspirated" but "hard" means velar. I got somewhat confused by that as I had heard that /k/ can be [c] before front vowels, so I thought they were saying *that*, before getting to the end of the sentence.

  35. Jongseong Park said,

    April 20, 2010 @ 12:12 pm

    And how does Keflavík become KePlavík?

    The weird thing for non-speakers is that Icelandic does have the sequence /fl/ word-initially (does it also occur word-internally but morpheme-initially?). Icelandic just seems liable to insert stops into sequences of unsuspecting consonants followed by /l/ or /n/, except that with /f/ the original consonant is completely replaced by the stop.

  36. Dan Lufkin said,

    April 20, 2010 @ 8:40 pm

    Further thought on Keflavík:

    Hmm, maybe it's an over-reaction to the OHG P–>PF–>F shift. The timing would be about right.

    When you visit Iceland, after it quiets down, do so in the early spring, before the grounds crew has spread fresh gravel on the paths at Þingvellir. You'll see a pattern of fine cracks in the soil. That's where Europe and North America are pulling apart about as fast as a fingernail grows. Eyjafjällajökull is only about 60 miles from Þhingvellir and 40 miles from Skálholt, the site of the old bishopric of Iceland. Geysir and Gullfoss (Europe's biggest waterfall) are in the same area. I spent a grand week there some years ago with Njáls Saga as a guidebook.

  37. Leonardo Boiko said,

    April 21, 2010 @ 7:40 am

    Man, I would SO love to study Icelandic, try to read the Sagas and eat poisonous, rotten shark… but my university has no Icelandic or Old Norse courses at all, and apparently there’s no student exchange program between Iceland and Brazil :/

  38. Maame said,

    April 21, 2010 @ 8:46 am

    Icelander:

    My Icelandic teacher once said that spoken as [l:] is the general rule for nicknames. Or: is always pronounced [dl] (voiceless or not or whatever) except for if the occurs in a nickname.
    That's interesting because Swedish, too, has special phonetic rules to the formation of nicknames.

    Jongseong Park:
    /fl/ does occur word-internally but morpheme-initially, e.g. Alþjóðaflugmálastofnunin.
    You need too know where a morpheme begins and ends to be able to pronounce an unknown icelandic word. I recently misprounounced eflaust 'without doubt' as [ɛplœy̯st] because I didn't know that it's a contraction of efalaust.

    The pronounciation of written or original ll as dl, nn as dn and rn as dn isn't unique in Icelandic. It's also found in Faroese and in several Norwegian dialects.

  39. Maame said,

    April 21, 2010 @ 8:49 am

    Okay, all graphemes seem to missing from my previous post. Here I go again:

    Icelander:

    My Icelandic teacher once said that ll spoken as [l:] is the general rule for nicknames. Or: ll is always pronounced [dl] (voiceless or not or whatever) except for if the ll occurs in a nickname.
    That's interesting because Swedish, too, has special phonetic rules to the formation of nicknames.

    Jongseong Park:
    /fl/ does occur word-internally but morpheme-initially, e.g. Alþjóðaflugmálastofnunin.
    You need too know where a morpheme begins and ends to be able to pronounce an unknown icelandic word. I recently misprounounced eflaust 'without doubt' as [ɛplœy̯st] because I didn't know that it's a contraction of efalaust.

    The pronounciation of written or original ll as dl, nn as dn and rn as dn isn't unique in Icelandic. It's also found in Faroese and in several Norwegian dialects.

  40. Jongseong Park said,

    April 21, 2010 @ 9:43 am

    Thanks, Maame for the answer to my question. The example of eflaust intrigues me. Is the f pronounced as a voiceless [f] there? Wouldn't the f in efa ("of doubt") and efalaust be pronounced [v]?

  41. Maame said,

    April 21, 2010 @ 3:03 pm

    Jongseong, I'm not 100% sure. I think it's voiceless and I think my teacher pronounced it as voiceless. But you are right: efa and efalaust are pronunced with a [v]. I'll see if I can look it up somewhere.

  42. Michael Rank said,

    April 21, 2010 @ 7:15 pm

    I took a few informal Mongolian lessons a few years ago and can report that Mongolian also has a voiceless lateral fricative (or similar).

  43. Amy West said,

    April 21, 2010 @ 7:58 pm

    For folks wondering how the modern Icelandic pron. got so different from the orthography there's a great little book, _The Icelandic Language_ by Stefan Karlsson, translated by Rory McTurk, available from the Viking Society for Northern Research's publications Web page:
    http://vsnrweb-publications.org.uk/

  44. PaulB said,

    April 22, 2010 @ 9:19 am

    >>>
    So if you're male, and want to say the Icelandic equivalent of "enchanté" when meeting someone…
    >>>
    I think you overlooked this bit in your revision. Unless it's implicit that male Icelandic speakers are only ever pleased to meet women.

  45. Jason L. said,

    April 22, 2010 @ 4:49 pm

    I find that when I say "scythe" (or any other word I can think of with a final /ð/) in isolation, the /ð/ is devoiced but noticeably more lenis than words with the same nucleus and /θ/ in the coda. I think the same is also true when I say "the farmer was scything the grain". The fortis/lenis distinction is also manifested in a longer preceding vowel for the lenis cases: "She was bathing" has a longer [ej] than a pretend-lisp "she was basing". FWIW, I'm a 28-year-old from San Francisco without, as far as I can tell, the California vowel shifts or much uptalk.

  46. Julie said,

    April 23, 2010 @ 4:39 am

    I'm also from Northern California, and just determined the other day that my mother (a lifelong resident of Mendocino County) doesn't pronounce any final consonant in scythe. I've picked one up, but for me it's a spelling pronunciation adopted in young adulthood.

  47. SPROGMUSEET » Eyjafjallajökull said,

    April 23, 2010 @ 1:38 pm

    [...] Om det helt nøjagtigt udtales tl eller hvad, diskuteres i teknisk detalje på Language Log her og her og på John Wells's phonetic blog [...]

  48. Qov said,

    April 25, 2010 @ 1:57 pm

    I lived in (mostly English-speaking) south Wales, and the Welsh LL sound I was taught for Llan had a voiced component, while the l in the clip sounds exactly like the unvoiced tlh sound I know from Klingon.

  49. Henry said,

    April 26, 2010 @ 8:27 pm

    http://www.youtube.com/watch?v=55UMIidyecA&feature=related conveniently depicts a realistic robot mouth saying it. Is there a machine that you can feed IPA into and have language-like sounds come out?

    I might devote my life to making such a machine if not.

  50. Census-taking, volcano-pronouncing, and why Thais win at Scrabble « the world in words said,

    May 4, 2010 @ 4:44 pm

    [...] volcanologists believe Eyjafjallajökull isn't done belching yet.  More pronunciation tips here and here. photo: [...]

  51. Leonardo Boiko’s Diary said,

    June 11, 2010 @ 1:24 pm

    [...] The real revenge, of course, is not the volcano eruption per si, but watching the poor anglophones trying to pronounce such a simple, logical word as Eyjafjallajökull (1, 2). [...]

  52. Yokult « Literal-Minded said,

    April 27, 2012 @ 4:10 pm

    [...] literally, "glacier of the mountains of the islands"). On Language Log, Mark Liberman explained the pronunciation, and on his Phonetic Blog, John Wells gave some additional [...]

  53. Liam McGee said,

    January 5, 2013 @ 5:27 pm

    Side note from a second-language welsh speaker… The Welsh 'll' has no suggestion of 'tl' or 'dl' in it (or 'cl' for that matter), there is no dental component nor is it voiced (at least not in West Wales where I grew up). It is a 'purer' lateral fricative than the Icelandic 'll' I'm hearing in Reykjavik, which seems to have a stop before the lateral. Eyafjallajokull sounds like a 'tl' at the end to me…

  54. asker said,

    June 2, 2013 @ 5:46 pm

    May I ask what the phoetics of the Icelandic word 'Heill'? In Iceland, do they say it with the same Germanic phonetics? I heard an Icelander say it as "Heek". And is it sometimes used as 'Hello or hey'?
    I know it's a very basic question regarding the forum and what you discuss here but I was very curious.

RSS feed for comments on this post · TrackBack URI

Leave a Comment