Re: Three questions concerning the Swedish language
- From: "Tor Aschan" <tor826@xxxxxxxxx>
- Date: 23 Sep 2005 09:31:48 -0700
FB wrote:
> I've read the f.a.q. "What is sci.lang for?", but am still doubtful whether
> the following questions about Swedish are allowed. If they aren't, let me
> know; and I'd appreciate if you'd recommend other newsgroups, too, since I
> asked the same questions on se.humaniora.svenska recently but have received
> no replies yet.
Please excuse the lack of replies on se.humaniora.svenska. Your
questions are very interesting but also very difficult to answer. What
has been said so far by others on sci.lang is quite good, but I'd like
to add to it.
> Conversely, if they are IT, I should point out that I haven't a proper
> Swedish grammar book at the moment---I'd appreciate recommendations on
> that, too (I'm considering buying the large Routledge one).
The large Routledge reference grammar (_Swedish: A Comprehensive
Grammar_) doesn't deal with the Swedish sound system, but I do
recommend it for everything else. The small Routledge reference
grammar (_Swedish: An Essential Grammar_) deals with the sound system
in the chapters "Pronunciation" and "Stress and Accent". These
chapters are, as far as I know (but I could be wrong), no worse than
anything else on the subject that is in print and generally available
in a language other than Swedish.
If you buy the small grammar book, be aware that the phonetic
transcription (within square brackets) is misleading (at least in the
1997 edition that I have) in that it doesn't distinguish between long
and short consonants. This is not a problem if you know that a single
consonant after a short vowel must be pronounced long in most varieties
of Swedish. (The distinction between long and short consonants is
about the same as in Italian.) There are a few other oddities as well:
for <säga> 'say', <trädgård> 'garden', and <ladugård> 'cowshed,
barn', the book gives [sE:ja], [trE:go:d], and [lA:go:d.], though they
are normally pronounced [sEjja], [trEggo:d.] (or trEggod.), and
[laggo:d.] (or [lA:du"go:d.] or [lA.:du"go:d.]). ([u"] is a high
central rounded vowel, not very different from English [U] in <good>.)
> 1) To what extent are vowels weakened in Swedish? I have very very little
> Swedish, and I used to think they weren't at all, unlike, say, in English,
> and German, to a lesser degree. This until I realized that "hade" is
> pronounced [hA.:d@] ([A.] being something like Swedish long "a"; [@] ==>
> schwa), or can be pronounced like that.
The word <hade> 'had' is sometimes pronounced [hA:dE>] (or [hA.:dE>]),
especially when it is uttered carefully, but most often it is
pronounced [haddE>]. (I invented "E>" to represent a sound a little
further back than [E].)
> Hence my original question: what
> vowels are weakened in unsteressed syllables, and how?
Vowels are not usually weakened much, except /E/ in certain unstressed
positions. For example, /E/ is quite often weak in <handen> /'handEn/
['hand@n] 'the hand'.
When /E/ has the primary stress, it is similar in pronunciation to the
German, American English, and standard British English varieties of
[E]. When it precedes the primary stress, it is pronounced [E] or [e],
except in unstressed /bE-/, /jE-/, and /SE-/, when it is mostly
weakened to [@] but may also be pronounced [E>] (the sound that is a
little further back than [E]). When /E/ follows the primary stress and
is both unstressed and weakened, it is pronounced [@]. But when it
follows the primary stress and is stressed to some degree or
unweakened, it is usually pronounced [E>], except before /r/, when its
pronunciation is between [E>] and [&], at about the level of French [E]
in <Pière>.
For example, <brunnen> /'bru"nnEn/ 'the well' has no stress on the
second syllable, so when /E/ is weakened there, as it quite often is,
the word is pronounced ['bru"nn@n], but <brunnen> /`bru"nnEn/ 'burned'
has secondary stress on the second syllable, so it is usually
pronounced [`bru"n,nE>n]. Note that the latter word has accent 2,
which is conventionally indicated by using ` instead of ' to mark the
primary stress. Accent 2 causes the second syllable to have the
secondary stress that favors the pronunciation with [E>].
Swedish [@] is similar to German [@] in <kommen> and English [@] in
<totem>. Beware of French [@] (as in a careful pronunciation of
<dessus>), because it is often too rounded, and don't use the English
varieties of [@] that you hear in <gopher> and <sofa>, because they are
too central or too central and low. Swedish [E>] is between German [@]
and German [E].
> 2) All verb forms have accent 2, don't they? What if that verb has a
> prefix? Do accents somehow shift? Or what else happens?
A polysyllabic verb form with primary stress on the first syllable has
accent 2 (a.k.a. the grave accent) unless it's a noncompound
present-tense verb form ending with <-er> (e.g., <lyfter> ['lyft@r]
'lifts'). Any other verb form has accent 1 (a.k.a. the acute accent).
Primary stress doesn't fall on the prefixes <be-> and <för->, so verbs
with those prefixes have accent 1. For example, <tala> /`ta:la/
[`tA:,la] 'speak' has accent 2, but <betala> /bE'ta:la/ [b@'tA:la]
'pay' has accent 1. Similarly, <söka> /`sY:ka/ [`sY:,ka] 'look for,
seek' has accent 2, but <försöka> /fY's.Y:ka/ [fO"'s.Y:ka] 'try' has
accent 1. ("Y" is here equivalent to IPA "ø".)
> 3) Assuming a stress consists of three elements, pitch, intensity and time
> (I hope I'm not mistaken), when do those elements occur in a word with
> accent 2 ( '\___'/ )?
The distinction between accent 1 and accent 2 is phonemic. Thus
there's a difference in meaning between <brunnen> /'bru"nnEn/
['bru"nn@n] 'the well' with accent 1 and <brunnen> /`bru"nnEn/
[`bru"n,nE>n] 'burned' with accent 2. But it's important to know that
a failure to make a distinction between the two accents will rarely
cause any confusion. There are a few dialects that don't distinguish
between the accents and have intonation patterns that are similar to
those of English. Thus students of Swedish will not sound completely
un-Swedish if they fall back on an English-like intonation.
The combinations of pitch and intensity that make up the accents vary
somewhat from region to region. Furthermore, the realization of an
accent on a word will depend on whether the word is by itself or in a
sentence, whether it's uttered quickly or slowly, whether it's
emphasized or unstressed, whether it's in a question or a statement,
and, if in a statement, whether it's at the end or not. One should
also realize that Swedes are not completely consistent from one moment
to the next in how they produce pitch in a particular word, phrase, or
sentence. So if your imitation of something in Swedish isn't quite
exact, chances are that it's nevertheless a satisfactory, albeit
perhaps infrequent, variant of what you've heard.
I'm not an expert on Swedish accents, but I can tell you how their
typical realizations sound to me. Here are three rules of thumb:
Accent 1: The intensity and pitch patterns are roughly similar to
those in an English word (that is not a compound). There is no
secondary stress but there is often a tertiary stress when a word has
more than two syllables. Note that all monosyllables have accent 1.
Accent 2 in noncompounds: There is always a secondary or tertiary
stress in addition to the primary stress. The primary stress is
usually but not always on the first syllable. In a two-syllable word,
the second stress (usually secondary) is on the second syllable. In a
word with more than two syllables, the second stress (most often
tertiary rather than secondary) is on the last syllable. In the
syllable with the primary stress, pitch falls from a mid level to a low
level, and in the next syllable, pitch is most often high and falling
(when the word is at the end of a statement) or rising (in other
contexts).
Accent 2 in compounds: There is always a secondary stress in addition
to the primary stress. Pitch falls from a mid level to a low level on
the syllable with the primary stress, which is in the first component
of the compound. Pitch is most often high on the syllable with the
secondary stress, which is in the last component of the compound.
In the examples that follow, I'm going to distinguish among four
degrees of intensity (3, 2, 1, and 0) and four levels of pitch (extra
high, high, mid, and low). Primary stress has intensity 4 or 3,
secondary stress usually has intensity 2, tertiary stress has intensity
1, and minimal stress has intensity 0. It seems that for any degree of
intensity and any level of pitch, there is some context in which the
two may occur together.
Accent 1
------------
<sitter> ['sitt@r] 'sits'.
Intensity: 4, 0.
Pitch in isolation or at end: high, mid - low.
Pitch before end: low - mid, high.
Pitch in question: mid - extra high, extra high.
Cf. English <sitter>. Intensity: 3, 0. Pitch: high - low, low. (Also:
high, mid - low.)
<komponist> [kOmpu'nist] 'composer'.
Intensity: 1, 0, 4 (also 0, 0, 4).
Pitch in isolation or at end: low, low, high - low.
(Also: low, low, high - mid; low, mid, hig - mid)
Pitch before end: low, mid, high.
Pitch in question: mid, high, extra high.
<händerna> ['hEnd@xxx] 'the hands'.
Intensity: 4, 0, 1.
Pitch in isolation or at end: high, mid, low.
Pitch before end: low, mid, high.
Pitch in question: mid - high, extra high, extra high.
Accent 2 in Words That Are Not Compounds
---------------------------------------------
<flicka> [`flik,ka] 'girl'.
Intensity: 3, 2 (sometimes 3, 3 when emphasized, and sometimes 3, 1).
Pitch in isolation: mid - low, high - low.
(Also: mid - low, high - mid.)
Pitch at end: high - mid, high - low.
(Also: high - mid, mid - low.)
Pitch before end: mid - low, high.
Pitch in question: high - mid, extra high.
<flickorna> [`flikkOn.a] (variants: [`flikkun.a], [`flikk@xxx]) 'the
girls'.
Intensity: 3, 0, 1 (sometimes 3, 0, 2).
Pitch in isolation: mid - low, high, low.
(Also: mid - low, high, mid.)
Pitch at end: high - mid, high, low.
(Also: high - mid, mid, low.)
Pitch before end: mid - low, mid, high.
(Also: mid - low, high, high.)
Pitch in question: high - mid, extra high, extra high.
<tidigare> [`ti:digarE>] 'earlier'.
Intensity: 3, 0, 0, 1.
Pitch in isolation: mid - low, high, mid, low.
(Also: mid - low, high, high, mid; mid - low, low, high, low; mid -
low, low, high, mid.)
Pitch at end: high - mid, high, mid, low.
(Also: high - mid, mid, high, low.)
Pitch before end: mid - low, low, mid, high.
(Also: mid - low, mid, high, high.)
Pitch in question: high - mid, high, extra high, extra high.
Accent 2 in Compound Words
-------------------------------
The vast majority of compounds have accent 2. However, their accent
patterns differ a little from those of accent 2 words that are not
compounds. Note also that the accent patterns of the compound as a
whole supersede the accent patterns of any part of the compound.
Phonologically, a word with an affix is often treated as a compound
word. Affixes from group 1 below are treated as components of compound
words whereas affixes from group 2 are not:
Affix group 1. This includes derivational affixes that always have
primary stress (e.g., <miss-> and <upp->) or secondary stress (e.g.,
<-het> and <-sam>) when the words that contain them are not parts of
larger compounds. For example, <-sam> has secondary stress in
<lönsam> /`lY:n,samm/ 'profitable' and also when a group 2 affix such
as <-ast> is added: <lönsammast> /`lY:n,sammast/ 'most profitable'.
However, it may lose its secondary stress when another group 1 affix or
another word is attached: <lönsamhet> /`lY:nsamm,he:t/
'profitableness'.
Affix group 2. This includes derivational and inflectional affixes
that are always unstressed (e.g., <be-> and <för-> in verbs) or that
are unstressed by the addition of any other affix (e.g., <-ning>,
<-ast>, and <-ar>; they have secondary or tertiary stress when they are
final but they are otherwise unstressed). For example, <-ning> has
secondary stress in <ritning> /`ri:t,niN/ 'drawing' but is unstressed
when <-ar> is added: <ritningar> /`ri:tniNar/ 'drawings'.
(The affixes in each group are listed in the 1990 edition of
Routledge's introductory textbook _Colloquial Swedish_.)
In a compound word, secondary stress (most often with high pitch) falls
on the last component, on the syllable that carries primary stress when
the component is uttered separately. A component of a lower-level
compound most often counts as a component of the highest-level
compound.
<brandbilarna> [`brand,bi:lan.a] 'the fire engines'.
>>From <brand> 'conflagration' and <bilar> 'cars, vehicles'.
Intensity: 3, 2, 0, 1.
Pitch in isolation: mid - low, high, mid, low.
(Also: mid - low, high, high, mid; mid - low, low, high, low; mid -
low, low, high, mid.)
Secondary stress falls on <bilarna> because <bilarna> is the last
component of the compound. The suffixes <-ar> and <-na> are from group
2 and don't count as components of the compound. Secondary stress
falls on the first syllable of <bilarna> because that is the syllable
that carries the primary stress when <bilarna> is uttered separately.
When separate, <bilarna> is pronounced [`bi:lan.a] and has the same
accent patterns as <flickorna> above.
<verksamhet> [`v&rksam,he:t] 'activity'.
>>From <verksam> 'working, active' and <-het>, a suffix from group 1.
Intensity: 3, 0, 2.
Pitch in isolation: mid - low, low, high - low.
(Also: mid - low, low, high - mid.)
<Verksam> is itself a compound of <verk> 'work' and <-sam>, a suffix
from group 1, but this fact has no effect here. Secondary stress falls
on <-het> because it is the last component of <verksamhet>. When
<verksam> is uttered separately, it is pronounced [`v&rk,sam] and has
the same accent patterns as <flicka> above.
<försäkringsavgift> 'insurance fee'.
>>From <försäkring> 'insurance' and <avgift> 'fee'. <Avgift> is itself
a compound of <av> 'off' and <gift> 'something given'.
This compound word has two types of pronunciation:
(1 - most often) [fO"`s.E:kriNsA:v,jift].
Intensity: 0, 3, 0, 1, 2.
Pitch in isolation: low, mid - low, low, low, high - low.
(Also: low, mid - low, low, low, high - mid.)
Here the compound is analyzed as having the components <försäkring>,
<av>, and <gift>, so secondary stress falls on <gift>, the last
component.
(2 - not infrequently) [fO"`s.E:kriNs,A:vjift].
Intensity: 0, 3, 0, 2, 0 (also 0, 3, 0, 2, 1).
Pitch in isolation: low, mid - low, low, high, low.
(Also: low, mid - low, low, high, mid.)
Here the compound is analyzed as having only the components
<försäkring> and <avgift>, so secondary stress falls on <avgift>. It
falls on the first syllable of <avgift> because that is the syllable
that carries the primary stress when <avgift> is uttered separately.
When separate, <försäkring> has accent 1 and is pronounced
[fO"'s.E:kriN], and <avgift> has accent 2 and is pronounced
[`A:v,jift].
-----
Tor
.
- Follow-Ups:
- Re: Three questions concerning the Swedish language
- From: Tor Aschan
- Re: Three questions concerning the Swedish language
- References:
- Prev by Date: Re: Easy as ABG
- Next by Date: Re: Easy as ABG
- Previous by thread: Re: [Possibly OT] Three questions concerning the Swedish language
- Next by thread: Re: Three questions concerning the Swedish language
- Index(es):
Relevant Pages
|
