WTPh? (What the Phonics) – Pronouncing street names in Denmark:

Created by by Momo Miyazaki and Andrew Spitz at the CIID, WTPh? (What the Phonics) is an installation which helps passers-by to learn the correct way to pronounce street names in Denmark.
Street names in Denmark are close to impossible for foreigners to pronounce, so we did a little intervention in the touristic areas of Copenhagen. We recorded a Danish person speaking the street names then split up each syllable. In true karaoke style, we placed lights above the matching syllable so that in real-time, you can see which part of the word is being spoken. When participants lift the speaker off the wall, it starts playing.

A very cool idea (Danish really is impossible to pronounce if you don’t know Danish); you can watch a brief video about it, and hear some of the street signs, here. (Brief, but not brief enough for my taste; too much artsiness and too little actual information. But hey, it’s only a couple of minutes.)
Via MetaFilter, where people share other interesting links, like this Danish dialect site.


  1. Danish really is impossible to pronounce if you don’t know Danish

    Correction: Danish is impossible to pronounce even if you know Danish.

  2. In the movie Haevnen you get to hear one character speak Swedish, because he’s Swedish, and everyone else speak Danish, becsause it’s set in Denmark. The effect is straight up creepy. The Danish in comparison to Swedish sounds like a deaf person trying to vocalize and the movie works like one unending Helen Keller joke, at least on one level.

  3. forstårikke says

    Ka-me-lå-se. Kamelåse.

  4. That’s very cool, I really like that, I would think any language nerd would.
    It’s very welcoming to foreigners as well, it encourages them to speak the local language but does so in just the right way, by trying to help them.

  5. John Emerson says

    I’ve been told that Norwegians say that Danes talk as though their mouths were full of potatoes. Didn’t say whether mashed, raw, baked, or what.

  6. doesntunderstand illustrates my point nicely.
    As for this gizmo, it’s not bad for what it does. Given that this is a design project (exam?), it’s actually very good. It identifies a need, and sets about solving it very cleverly. I rather hope the tourist board pours some money into implementing this. The artsiness, like it or not, has to be the focus of this presentation.

  7. Jeffry House says

    John, it is potatoes “i halsen”—deep in the throat. Small, round ones which have been boiled with the skins on.

  8. weatherbeatenvessel says

    This cries out for the creation of a PronunciationManual-style counterpart.

  9. LH, Sili, do you mean Danish is difficult to articulate, or that it’s difficult to infer pronunciation from spelling? Or both?
    My German friends used to tell me that I sounded like I had Kartoffeln im Mund when I spoke English, so I thought the potatoes-in-the-mouth allegation was a pretty conventional characterisation of all kinds of foreign speech. But maybe it’s just the Danes and me.

  10. michael farris says

    Cute idea though I agree that there wasn’t so much content.
    As for Danish pronunciation, it’s my understanding that while Danish orthography is mostly archaic and the spoken language has moved in some odd directions that mean there’s not much of a fit between writing and speech.
    As a non-expert, my impression is that spoken Danish (as opposed to written) can be characterized by (among other things):
    – syllable mutation, both collapse (as two or three written syllables turn into one) and reconstruction (sometimes the phonetic shape of the word is very different from what the orthography would suggest
    – consonant mutation (partly reflected in the script) as unvoiced consonants are voiced and voiced consonants spirantized, but also t often becomes something like [ts] and d becomes something like [I have no idea what the symbol is, it sounds halfway between a voiced interdental and a lateral]
    – vowel mutation, compared with other Scandivavian languages whose vowels move more back, in Danish they’ve moved to the front, also some weird stuff like unstressed -er becoming (for some speakers) something like [ɔ]
    All those combined at native speed can make the idea of learning to understand Danish kind of depressing for a non-native.
    Also, it might not seem as weird as it does if Swedish and (semi-official) Norwegian didn’t exist where the fit between speaking and orthography is a lot closer. Imagine a spoken national language very much like written French but with all the consonants (and vowels) restored and you might have an idea of the comparison between Swedish/Norwegian and Danish (S/N also have their weirdnesses but they’re not in the same league).

  11. For a Dane, it’s not so much pronouncing what you read that’s hard, it’s spelling what you hear. Lots of homophones, more on the way.
    Danish vowel phonology is quite complex, with a number of very similar sounds as conditioned allophones of different phonemes – postvocalic r is mainly realized as lowering of the preceding vowel, for instance – but parts of that system seems to be collapsing. I’ve had to teach my kids where some r’s go in spelling by the singular/plural distinction, they simply don’t hear the difference that I do in some contexts. No wonder that non-natives have a hard time!
    @Michael Farris: The strange ‘soft d’ is a non-sibilant postalveolar fricative (or approximant) – not lateral in production. though there is a certain similarity in sound. (A high front semivowel is the usual substitute when people can’t pronounce it). The IPA doesn’t have a symbol for it, your best bet is edh with the retracted diacritic.

  12. Thanks for the Danish dialect link, Language.
    Sili: doesntunderstand illustrates my point nicely.
    Does kamelåse actually mean something in Danish? I thought it was just a fake word invented by Harald Eia & co.
    Sili: The artsiness, like it or not, has to be the focus of this presentation.
    Why? It’s a useful & well executed idea, with long stretches of music I didn’t like & unfunny jokes.
    I wouldn’t want anyone to form the impression that because of the hot-potato effect Danes don’t articulate the words. They distinguish between different vowels instead of making them all into one amorphous schwa, like we do in English.

  13. Quoting myself again:

    What I don’t understand is why Danes don’t spit out the potato at birth. Surely it must make suckling difficult? But no, it remains in place until it’s removed by the undertaker and ceremonially planted in a separate tiny grave next to the big one.

    And then we went off on potatoes vs. foreskins.

  14. And I want to plug The Killing here again, because it’s Danish (with subtitles, so you can test your understanding of the language). It’s the best, most depressing tv detective thriller in the history of the universe, it makes Sam Spade look like a Jimmy Stewart character.

  15. We missed out on placentas, although they too have been buried ceremoniously in some cultures.

  16. Christopher Burd says

    (S[wedish]/N[orwegian] also have their weirdnesses but they’re not in the same league)
    A lot of very common words in Swedish have irregular orthographies: silent letters in “jag”, “med”, etc.; “mig”, “dig”, “sig” = [mej], [dej], [sej]; and “de”/”dem” pronounced [dom] (?!); and so on. (No IPA glyphs, sorry.) I suppose if this was typical of the whole language, Swedish would be on par with Danish. In fact, for most of the vocabulary the spelling is fairly sane.

  17. Why? It’s a useful & well executed idea, with long stretches of music I didn’t like & unfunny jokes.

    Again. It’s a presentation from a design school. They have to speak the vernacular to get taken seriously by their peers.

    Does kamelåse actually mean something in Danish? I thought it was just a fake word invented by Harald Eia & co.

    Never heard it before, but if I were to pronounce it, I’d go for ka-mel-åse (camel – girlsname).

    LH, Sili, do you mean Danish is difficult to articulate, or that it’s difficult to infer pronunciation from spelling? Or both?

    Incidentally, Skellerup on that dialect map is the parish next to where I grew up. Not that I ever picked up the dialect, myself.

  18. Why not? Was your speech influenced by a more pervasive dialect?

  19. Yeah. I’m of mixed marriage and we always spoke RP(DK) at home.

  20. What’s RP(DK) called, på dansk?

  21. dearieme says

    I watched an archaeology show on the telly this evening: one of the things found was a “viking” whorl – a wee device used in spinning wool. Alas, the people discussing it were all speakers of Lunnun English, and they pronounced it “wel” so that I didn’t know what the hell they were talking about until the word appeared on screen.
    Really we need subtitles so that speakers of euphonious English – like me – can understand what the hell Londoners are talking about.

  22. You’re right. It’s time to heavily tax the use of schwas. English speakers will only be able to use them for free on international flights and at airports, before they go through customs.

  23. Bathrobe says

    How do you get ‘wel’ out of ‘whorl’. Even in Australian it would be pronounced the same as ‘wall’, not ‘wel’ (whatever that is).

  24. It’s all just one schwa letter in southern English. I don’t know why they bother writing AEIOU, it’s a waste of vowels.

  25. Dearieme: I can understand London “whorl” being mistaken for “whirl”, and even “world”, but all those include the tongue touching the hard palate at the end of the word, albeit barely noticeably. London “well” involves no movement of the tongue at all.

  26. If you haven’t seen this… (the origin of Kamelåså?)

  27. i) They omitted the “h”, so it starts with “w” alone. (Just as their whales = Wales: this I expect in England, but it still has the ability to throw me when combined with other infelicities.)
    ii) The “r” wasn’t non-rhotic, it was non-existent.
    iii) The vowel – which I wrote as “e” – was sometimes indeterminate, sometimes a diphthong: “e” is simply an approximation.
    iv) The “l” was scarcely pronounced – just a hint of it was present. This softening of consonants, but especially “l”, is becoming (it seems to me) pretty common. You’ll sometimes see sarcastic people writing it as a “w”, so that “well” might be written as “wew”.
    That last one isn’t unique to London, since I heard it too when we lived in Adelaide (South Australia) in the late 80s.
    Fifteen years ago an Italian complained to me about a public announcement at Heathrow directing him to bus stop “I” – that’s a capital “eye” by the way. (Dear God, why do we use such idiotic fonts?) Anyway, to “I” he went, but it turned out that the announcer had meant the “I” that comes before “B”. Whereas more recently, as AJP implies, he’d have been lucky to make out any vowel at all, right or wrong.

  28. I think that too much was thrown overboard along with Received Pronunciation. It’s all of course deliciously egalitarian that speech in various media is now not Received, but more Thrown-In-Your-Face. However, in train stations and airports one wants clear diction, not democracy.
    This problem exists in Germany as well. Several times, in the best and most conciliatory spirit (believe it or not), I have suggested to various officials in the Cologne train station that certain aspects of the English announcements (spoken by Germans) are hard to understand. In particular I pointed out that the difference between “fifteen” and “fifty” is crucial when the subject is how many minutes a train will be delayed.
    I explained that, regardless of what Germans may have learned in schools about the pronunciation of “fifteen” and “fifty” as counting words, in loudspeaker announcements you must say “two fifTEEN” and “two FIFty”. What comes through the speakers for 2:15 is something “two FIFtee(n)” with inaudible “n” (depending on background noise), so you don’t know whether 2:15 or 2:50 is meant.
    But all of this came to nothing. None of these officials gave a damn about the intelligibility of the announcements, and seemed to think I was some sort of weird pedant.
    It is unexpected, I admit, that we say in America (and I think in GB also) “FOURteen, FIFteen, SIXteen” but (clock times) “two fifTEEN, two sixTEEN”. I wasn’t aware of it myself until I analyzed why those loudspeaker announcements were hard to understand.

  29. In German there is never any stress shift in number words, whether they are used for counting, telling time or any other purpose (is there another purpose ??)

  30. Let me make that more precise: I meant no stress shift within a “simple” counting word, which I define to be those from 1 to 19. Composita like zweiundvierzig, when they are used for counting (instead of denoting, as in clock times) can be pronounced ZWEIundvierzig, DREIundvierzig or zweiundVIERZIG, zweiundFÜNFZIG. It depends on the decimal place you want to stress.
    I hope that is intelligible as is – it would be rather time-consuming to explain in detail how it works. Actually, it occurs to me that we can do exactly the same thing in English.

  31. John Wells did a blog a good while back pointing out that automated announcements on English stations suffered from wrong stress as well, because the numbers had been recorded in isolation.

    What’s RP(DK) called, på dansk?

    Rigsdansk. Danish of the realm – or Reich, if you prefer.

  32. Sili: some of the announcements in the Cologne train station are recorded, some are spoken live. They all make the mistake I described. I spoke to the officials only about the recorded announcements (they are not electronically patched together), since I didn’t want to offend a live person …
    These announcements are the ones that come through loudspeakers on the platforms. It’s interesting that the multi-lingual announcements inside the station are impeccable (the tracks are a floor up from the halls and pedestrian zones with shops, ticket counters etc.) The (recorded) English and French versions of these are spoken by native speakers, of that there is absolutely no doubt. They warn you to be on your guard against pickpockets.

  33. FIFteen vs fifTEEN: I think that the latter somes up in contexts other than telling time. How old are you? FifTEEN. (Well, that’s a kind of time, innit? OK, try again.) How much money is in your pocket? FIFteen CENTS. How many pennies do have? FifTEEN.
    In my youth I was fooled by a Brit’s pronunciation of “eighteen”. It sounded like “eighty”, but not because of the stress. The problem was that (1) her “n” sound was a little quiet and (2) where I come from we really say eightteen.
    I know that at least once in a German train station I heard “Gleis zwo” for “Track two”. I wonder, was that a case of local dialect making an appearance (as deplored in However, in train stations and airports one wants clear diction, not democracy) or a case of sensibly using an alternative to “zwei” to avoid confusion with the same-vowel word “drei”?

  34. The latter.

  35. Missed another trick. I wish I had written: “The ladder”

  36. FIFteen CENTS. How many pennies do have? FifTEEN.
    You’re right, but I can’t figure out “how to formulate the rule” for that. Or rather “how to formulate the regularity”, since it may well not be a hard and fast rule, dependin’ on where you come from.

  37. Wow, I just checked out that alternate stress thing by eliciting it from my teenage children using the questions in ø’s post, and, of course, it checked out. I’m a native (and apparently naive) English speaker and a trained linguist, and I have never noticed this phenomenon in 40 years. Wow.
    The only difficulty I encountered in eliciting the correct response was when my non-Autism-spectrum (and usually non-smart-aleck) daughter responded to the the question about how much money she had in her pockets by looking at me across the table and proclaiming, “None. I don’t have any pockets.” [Eye roll from me.] while her ASD younger brother simply answered the questions as expected. I think she has something against linguistics.

  38. The London Underground has very good clear announcements in a soothing female voice. There’s one about trains being delayed due to a body on the line.
    Stu, I don’t know about emphasis but German has zwo (I don’t know how you spell it) to make sure you don’t hear zwei as “drei”.
    They had reichsmål or riksmål in Norway but they changed to to “bokmål” – I suppose to make it sound less scary.

  39. Never mind, I see you’ve covered zwo while I was walking the dog.
    where I come from we really say eightteen
    They do that in some British dialects. Liverpool is one.

  40. Are there really large numbers of people – in Britain, the USA, anywhere English is spoken – who say “eigh-teen” or “eight-een” ? I thought everybody said “eight-teen”.
    The only exception I can reconstruct in my idiolect is when I count very fast: “seven-teen-eigh-teen-nine-teen” (or is it “…eight-een..”, I can’t tell by introaudio).

  41. Or intraudition, as opposed to recording my speech and playing it back – extraudition.

  42. dearieme says

    ‘I thought everybody said “eight-teen”.’ Not me.
    About fifteen years ago our post office underwent an overhaul. I reported to my wife that it had one unexpected feature: the recorded loudspeaker announcements had gone continental. It would invite the next person in the queue to go to guichet five, as might be.
    She replied, with a quite unnecessary scorn, “I think you’ll find that that is a local mispronunciation of cashier five”.

  43. C’mon, dearie, don’t keep us in suspenders. How do you pronounce eighteen: “eigh-teen” or “eight-een”, or something even more exscottish ?

  44. dearieme says

    Which I use may depend on context: I’m getting self-conscious now, so it’s getting harder to be sure. Usually, I think, eigh-teen.
    It makes sense: thir-teen, four-teen, fif-teen, six-teen, seven-teen, eigh-teen, nine-teen.
    I use twenty not twenny, thirty not thirry, forty not forry seventy not sevenny , ninety not niney. Nor do I use a glottal stop in eighty. I use hundred, not my old playground hunnert. Come to that, I don’t use my old playground yin, twy, thry either. And that’s either, not eether.

  45. When I said “eightteen”, I meant the first of the two “t”s to be a glottal stop, I believe.

  46. I say ‘eigh-teen’, in case anybody’s inner rested.

  47. For me it’s eigh?-teen. Though I don’t glottalize my /t/s in most cases, /tt/ across a morpheme boundary becomes /?t/.

  48. Christopher Burd says

    I say not only eightteen, but also thirtteen, fourtten, ninetteen, and twunny. There’s probably a more sophisticated way of analysing this.

  49. My pronunciation is identical to Christopher Burd’s, or at least it was until I noticed that I was doing it. The “extra t” adds another way of distinguishing the -teens from the -tys (along with the n, the stress and, in American, the flapped t).
    On the stress question, I would say that the default is fifTEEN, but it becomes FIFteen either when it is contrasted with other -teen numbers (as in counting up or down) or when it modifies a noun.
    The latter process isn’t unique to numbers; compare “good afterNOON” with “AFTERnoon TEA”.

  50. Rodger C says

    I was just about to post the same thing John Cowan said. Just saying it’s not his idiosyncrasy or regional pronunciation.

Speak Your Mind