2002-05-22 10:36

Swedish Pronunciation, Part 1

Swedish Pronunciation Part 1

[ By ancient mathematical tradition I began at Part 0. That was yesterday. ]

The pronunciation guide in Routledge's Colloquial Swedish is OK, so far as it goes. The cassettes are OK, too, but they feature real live Swedish people talking at pretty much real conversational speeds, which is a bit daunting if you're not used to it. Which if you're listening to them if the first place you're pretty much guaranteed not to be.

Even after two years of Swedish classes with a teacher taking care to chop sentences into the words nicely for us I still find decoding honest-to-goodness full-speed Swedish hard work hard. Isolated words are easy enough, but when they start congealing into clumps large enough to be worth saying it's a different matter.

The problem is that words in sentences get changed and run together in all sorts of ways. Linguists call this prosody, but knowing that doesn't help much. What does help is when a textbook, such as the excellent P� Svenska!, goes to the trouble of explaining the prosodical rules of the language, with careful and accurate use of phonetic notation.

A lot of people seem to react badly to the IPA. They think it looks complicated and difficult, and more trouble than it's worth. There's certainly a lot of it - Catford's Practical Introduction to Phonetics has the full gory details, if you want them - but for any given language you only need a small subset, and it uses exactly one symbol for each sound. This means that the job of pronouncing Swedish can be split up into

• knowing how to pronounce each symbol (i.e., each sound in the language); and
• knowing how to map ideas into symbols (i.e., sounds).

Note that you would still have to do this without the IPA, it just makes some aspects more explicit. In particular, if I don't know how to write down an arbitrary chunk of Swedish in IPA notation that's the same as not knowing how to say it, and that's bad.

So my way of working on Swedish pronounciation is:

• Pick a dialogue/text recorded on the Colloquial Swedish cassettes;
• Transcribe it into IPA;
• Practice saying that;
• Compare with the tape.

I hope it goes without saying that transcribing into IPA is and should be a collaboration between voice and pencil.

This may sound like a lot of work. It is quite a lot of work, but it's also quite satisfying when you listen to the tape and understand exactly how it got to sound like that.

For example the sentence:

Eller finns det n�gon r�d avg�ng f�rresten?

when spoken, sounds more like:

[ El@fInsdEnOn'rY:'dA:v'gON:f&.rEst@n ]

(NB: in Swedish transcriptions I propose to use the Swedish ASCII IPA as used on se.humaniora.svenska)

When you know what to listen for - when you're listening with a trained mouth, rather than baffled foreign ears - you can hear the difference in quality between '�' in 'r�d' (which comes out [Y:]) and '�' in 'f�rresten' (which is [&.]). There's a difference in length, but there's also (and for a different reason) a difference in quality. And how many English-speakers would infer for themselves the distinction between dental consonants ([d] and friends) and the retroflex versions ([d.] and friends)? I'm fairly confident that I wouldn't have done so unaided, although I can hear them well enough now that I've got into the habit of using them myself.

So I'm proposing to have a running series of Swedish Sounds of the Day (not guaranteed to be daily, though) introducing sounds and their ASCII IPA symbols, and then maybe to describe enough prosodic rules to allow the interested reader (hem hem) to understand conversions such as are involved in the sentence above.