The orthography of Swedish is pretty similar to that of German. Swedish uses your basic 26-letter Latin alphabet like English, with the addition of the letters å, ä and ö. The latter two have values close to German, and it also shares with German the practice of using double consonants to show the preceding vowel is short and only one consonant to show the preceding vowel is long. Like in German, and unlike in Dutch, in Swedish a word can end with two consonant letters and vowel letters are not doubled to show length; and because of this a word like bok has a long O, while a word like mitt has a short I. Swedish spelling also shares with English and German the use of CK as the doubled version of K; and with German and Dutch the use of J for i̯ / j.
W, Q, Z, the digraph CH and C are rarely used outside of proper names and loanwords, except for the word och – more on C later. The pronunciations of the letters b d f h l m n p s t v x and the digraph ng are pretty self explanatory if you already know English. Swedish, like English and unlike German and Dutch, does not have syllable final devoicing of consonants, and like English and German but unlike standard Dutch, aspirates p t k in certain environments.
The pronunciation of R should not be difficult to anyone who can already pronounce Spanish – it can either be pronounced r or ɾ depending on its position in the word. The distribution of these realizations is very reminiscent of Spanish. Like in Spanish, they are only contrastive intervocalically. One of the interesting things about R in Swedish is that the clusters rd rl rn rs rt, in which you’d expect ɾd ɾl ɾn ɾs ɾt, simplify to the retroflex consonants ɖ ɭ ɳ ʂ ʈ. If you’re learning Swedish and the thought of this is already giving you a headache, you could always learn the Scanian accent instead, which pronounces R as ʁ, otherwise known as the French R.
Let’s move on to the really interesting part as far as I’m concerned: vowels.
In Swedish orthography vowel sounds are represented by the letters A O U Å E I Y Ä Ö. Y and Ö represent front rounded vowels, like German Ü and Ö, and Ä represents a front vowel, like in German. Å represents a back rounded vowel. In the Swedish sound system, A O U Å work like back vowels and E I Y Ä Ö work like front vowels – more on why this is relevant later. First, a quick table to show you the vowel sounds and their corresponding spellings.
|O||ɔ / ʊ||ost / bott||O||uː / oə̯||bok / sova|
As you can see, there is a little overlap, with short O and Å both representing ɔ and short E and Ä both representing e̞; apart from that, it’s a rich vowel system, which is fun for people who like lots of vowels, like me.
On to the meaty stuff: Swedish back vowels show evidence of a vowel shift. Knowing little of Swedish philology, I have no idea what change happened first, and whether we can call it a pull chain or a push chain. But we can see the results. Let’s imagine it like a pull chain. First uː fronted to ɵː, leaving room for oː to close to uː, leaving room for ɔː to close to oː, itself leaving room for ɑː to close and round itself, which in turn allowed a to centralize to ä. This is reminiscent of similar developments in English. The English GOOSE vowel is centralized to ʉ in most accents, and in some can even front all the way to y. This leaves a lot of room in the cardinal u area, which allows for the THOUGHT vowel oː to close all the way to uː in environments that would promote that, such as before l. Listen to Dizzee Rascal here pronouncing GOOSE as yː (movement, improvement) and pre-L THOUGHT as uː (all):
Lest you think this is just a Multicultural London English thing, here’s Ed Sheeran telling us about how people fuːl in love in mysterious ways:
In time perhaps Southern English English will raise the THOUGHT vowel to uː in every environment, as Geoff Lindsey has already predicted. If it happened in Swedish, it can happen here.
This is where knowing about linguistics makes language learning especially fun for me. Having read about how Swedish long U is nothing like, say, Spanish U or German long U, when I encountered bok buːk, instead of huh? I thought hah! vowel shift raised oː to uː.
Another point where knowing about linguistics makes Swedish spelling make sense is palatalization. Palatalization is the reason English cheese tʃʰiz corresponds to Dutch kaas kas and English yesterday ˈjɛstəɹdei to Dutch gisteren ˈχɪstəɾə (both pairs of words are cognates). The front vowels i and ɛ, by virtue of being pronounced further forward in the mouth, caused the preceding velar consonants to also move further in the mouth, becoming palatal and going from k to tʃ and from g (or ɣ) to j.
The same thing happens with the pronunciation of the Swedish letters K, C, SK and G. When they’re followed by the front vowels e i y ä ö, they’re palatalized, and so pronounced differently. Before the back vowels a o u å, K and C are predictably k, SK is sk and G is g. Before the front vowels e i y ä ö, K is ɕ (similar to English SH), C is s, SK is ɧ, and G is j (as in yesterday).
Let’s go back to the Swedish vowel system as a whole. As you could see above, Swedish has long and short vowels. Unlike in American English, the long vowels are, consistently, actually long vowels, and the short vowels are actually short vowels in the pronunciation – they’re not just names handed down by tradition. However, the length differences are also accompanied by quality differences in, it seems, every case. It seems to me that Germanic languages as a whole don’t like to just use length to distinguish vowels, and they need quality differences as well (hey, if it happened in Vulgar Latin, it could happen here too). Going back to how the vowel system as a whole preserves these differences, the rounding of long A sure helps distinguish it from short A. Likewise, difference in openness (and diphtongization) helps distinguish short Å / O from long Å. The fact that short Ä / E seems to be pronounced more like e̞ than like the more traditional transcription ɛ, i.e. more close, helps explain why long E developed into a diphthong – it’s a lot easier to distinguish e̞ from eə̯ than from eː. The realization of short Ä / E as e̞ itself might have come from the need to further distinguish it from long Ä ɛː.
So why do Germanic languages like quality differences in their short / long vowel pairs as opposed to just length?
The answer might lie with the intricacies of Germanic phonotactics. I get the feeling Germanic languages favor a, let’s call it a three-mora syllable structure. This could be CVC, CVV, VCC or VVC. VV here would stand for either a long vowel or a diphthong. Of course, Germanic languages’ propensity for ending syllables with consonants could also result in CVVC – and this is pretty common. On the other hand, fortis voiceless consonants tend to shorten the preceding vowel – what’s commonly referred to in English phonetics as “pre-fortis clipping”. If short / long vowel pairs were just distinguished by length, then their differences would be neutralized in syllables ending in fortis consonants. Considering how often this happens in Germanic languages (particularly German and Dutch), quality differences seem to be essential if the vowel system is to remain complex.
In a vowel system where quality differences coincide with length differences, sometimes the length differences are preserved (like in Southern English English and Swedish), and sometimes they’re eventually lost, or at least lose their systematicity (like in American English or Dutch).
To wrap these thoughts up, I’ll mention that the Swedish system of vowel contrasts also seems to undergo changes before R. However, I’m still murky about the finer points of how it happens, so I’ll do some more research and get back to you on that.