Discussion:
German umlauted vowels different from Swedish counterparts?
(too old to reply)
Dingbat
2018-01-29 05:41:03 UTC
Permalink
Raw Message
German <ü> I hear as similar to Swedish <y>
German <ö>, when long, I hear as similar to Swedish <u>, not Swedish <ö>;
i.e., I don't hear in German <schön> the sound of the first vowel in
Swedish <Öland>

Has anyone else made similar observations?
Ruud Harmsen
2018-01-29 20:17:53 UTC
Permalink
Raw Message
Sun, 28 Jan 2018 21:41:03 -0800 (PST): Dingbat
Post by Dingbat
German <ü> I hear as similar to Swedish <y>
I don't. The Swedish (and Norwegian) vowel is really front, the German
(and French, Danish, Turkish, Hungarian, Dutch) vowel is almost front,
or between front and central. That is why they sound so different.
Post by Dingbat
German <ö>, when long, I hear as similar to Swedish <u>, not Swedish <ö>;
Likewise, the German ö (and Turkish, Hungarian etc.) vowels are not
front, but almost front. I'm not sure if the Swedish ö is front, but
it might.
Post by Dingbat
i.e., I don't hear in German <schön> the sound of the first vowel in
Swedish <Öland>
Has anyone else made similar observations?
I hear Swedish <u> (which is central) more like Dutch <uu> (almost
front), than Swedish <y> (front).
--
Ruud Harmsen, http://rudhar.com
Ruud Harmsen
2018-01-29 20:28:34 UTC
Permalink
Raw Message
Post by Ruud Harmsen
Sun, 28 Jan 2018 21:41:03 -0800 (PST): Dingbat
Post by Dingbat
German <ü> I hear as similar to Swedish <y>
If you listen to a speaking IPA charts (links here
http://rudhar.com/foneport/en/lingglos.htm), you;ll find that cardinal
[y] sounds quite like a Swedish/Norwegian <y>, but nothing like a
German etc. <ü>.

Also, the opener vowels in the series "front rounded", which are
supposed to occur in French (e.g. in "un peu", "coeur"), don't sound
like that at all. Because the French vowel aren't front, but more in
the direction of central, without being fully central.
--
Ruud Harmsen, http://rudhar.com
Ruud Harmsen
2018-01-29 20:43:39 UTC
Permalink
Raw Message
My theory to explain this, is in a natural tendency in languages of
phonemes to maintain maximum distance to other phonemes, because
maximum contrast makes recognition easier.

Swedish and Norwegian have a three way distinction in close rounded
vowels: <y> <u> <o>, front, central, back. <y> must remain fully front
to keep it apart from the central <u>, even though that makes /y/ more
similar to /i/.

In other languages (all others? I know of no other examples than
se/no, of languages in which /y/ is really phonetically front), like
German, French, Dutch, Danish, Turkish, Hungarian, because there is no
rounded central high vowel, but only rounded back high vowels, their
/y/ phoneme(s) "can afford" to drift more towards a central position,
so they are better differentiated from their fully front (and
unrounded) /i/ phoneme(s).
--
Ruud Harmsen, http://rudhar.com
António Marques
2018-01-29 23:25:39 UTC
Permalink
Raw Message
Post by Ruud Harmsen
My theory to explain this, is in a natural tendency in languages of
phonemes to maintain maximum distance to other phonemes, because
maximum contrast makes recognition easier.
Swedish and Norwegian have a three way distinction in close rounded
vowels: <y> <u> <o>, front, central, back. <y> must remain fully front
to keep it apart from the central <u>, even though that makes /y/ more
similar to /i/.
In other languages (all others? I know of no other examples than
se/no, of languages in which /y/ is really phonetically front), like
German, French, Dutch, Danish, Turkish, Hungarian, because there is no
rounded central high vowel,
Isn’t Turkish dotless-i supposed to be barred-u?

Yusuf could tell us, but I haven’t seen him lately.
Post by Ruud Harmsen
but only rounded back high vowels, their
/y/ phoneme(s) "can afford" to drift more towards a central position,
so they are better differentiated from their fully front (and
unrounded) /i/ phoneme(s).
Peter T. Daniels
2018-01-30 04:07:06 UTC
Permalink
Raw Message
Post by António Marques
Post by Ruud Harmsen
My theory to explain this, is in a natural tendency in languages of
phonemes to maintain maximum distance to other phonemes, because
maximum contrast makes recognition easier.
Swedish and Norwegian have a three way distinction in close rounded
vowels: <y> <u> <o>, front, central, back. <y> must remain fully front
to keep it apart from the central <u>, even though that makes /y/ more
similar to /i/.
In other languages (all others? I know of no other examples than
se/no, of languages in which /y/ is really phonetically front), like
German, French, Dutch, Danish, Turkish, Hungarian, because there is no
rounded central high vowel,
Isn’t Turkish dotless-i supposed to be barred-u?
No, it patterns with the unroundeds.
Post by António Marques
Yusuf could tell us, but I haven’t seen him lately.
Post by Ruud Harmsen
but only rounded back high vowels, their
/y/ phoneme(s) "can afford" to drift more towards a central position,
so they are better differentiated from their fully front (and
unrounded) /i/ phoneme(s).
Arnaud Fournet
2018-01-30 07:33:40 UTC
Permalink
Raw Message
Post by António Marques
Post by Ruud Harmsen
My theory to explain this, is in a natural tendency in languages of
phonemes to maintain maximum distance to other phonemes, because
maximum contrast makes recognition easier.
Swedish and Norwegian have a three way distinction in close rounded
vowels: <y> <u> <o>, front, central, back. <y> must remain fully front
to keep it apart from the central <u>, even though that makes /y/ more
similar to /i/.
In other languages (all others? I know of no other examples than
se/no, of languages in which /y/ is really phonetically front), like
German, French, Dutch, Danish, Turkish, Hungarian, because there is no
rounded central high vowel,
Isn’t Turkish dotless-i supposed to be barred-u?
barred i = unrounded closed back vowel
something like Russian y in бык byk "ox, bull".
Daud Deden
2018-01-30 15:50:57 UTC
Permalink
Raw Message
A.: "something like Russian y in бык byk "ox, bull". "
-
bison/wisent Bos beef bovid
***@Malay: cow/cu(d.der)

The Lemba of W Africa are cattle raising (matriarchal).
Daud Deden
2018-02-01 00:17:35 UTC
Permalink
Raw Message
Post by Daud Deden
A.: "something like Russian y in бык byk "ox, bull". "
-
bison/wisent Bos beef bovid
The Lemba of W Africa are cattle raising (matriarchal).
Not sure if Russian byk ~ buck, yak.
Yusuf B Gursey
2018-02-03 01:17:46 UTC
Permalink
Raw Message
Russian byk seems like Hungarian bika which comes from a Chuvash type reflex of Turkic boğa boka buka etc.
Yusuf B Gursey
2018-02-03 01:01:35 UTC
Permalink
Raw Message
The Turkic / Turkish vowel is further back than the Russian vowel but Cyriilic orthographies use the same symbol
Ruud Harmsen
2018-01-30 08:03:20 UTC
Permalink
Raw Message
Post by Ruud Harmsen
In other languages (all others? I know of no other examples than
se/no, of languages in which /y/ is really phonetically front), like
German, French, Dutch, Danish, Turkish, Hungarian, because there is no
rounded central high vowel,
Mon, 29 Jan 2018 23:25:39 -0000 (UTC): António Marques
Isn’t Turkish dotless-i supposed to be barred-u?
No. It's a back vowel, not central. As a phoneme, that is. It might
have allophones that are more central, because Turkish vowel have lots
of room: there are only four, on the four corners of the diagram (each
rounded and unrounded, which makes eight).

Moreover, I meant Turkish ü.
Yusuf could tell us, but I haven’t seen him lately.
Post by Ruud Harmsen
but only rounded back high vowels, their
/y/ phoneme(s) "can afford" to drift more towards a central position,
so they are better differentiated from their fully front (and
unrounded) /i/ phoneme(s).
--
Ruud Harmsen, http://rudhar.com
Peter T. Daniels
2018-01-30 13:50:18 UTC
Permalink
Raw Message
Post by Ruud Harmsen
Post by António Marques
Post by Ruud Harmsen
In other languages (all others? I know of no other examples than
se/no, of languages in which /y/ is really phonetically front), like
German, French, Dutch, Danish, Turkish, Hungarian, because there is no
rounded central high vowel,
Mon, 29 Jan 2018 23:25:39 -0000 (UTC): António Marques
Post by António Marques
Isn’t Turkish dotless-i supposed to be barred-u?
No. It's a back vowel, not central. As a phoneme, that is. It might
have allophones that are more central, because Turkish vowel have lots
of room: there are only four, on the four corners of the diagram (each
rounded and unrounded, which makes eight).
That makes 8, not 4. The rounded vowels are not allophones of the unrounded phonemes.
Post by Ruud Harmsen
Moreover, I meant Turkish ü.
Post by António Marques
Yusuf could tell us, but I haven’t seen him lately.
Post by Ruud Harmsen
but only rounded back high vowels, their
/y/ phoneme(s) "can afford" to drift more towards a central position,
so they are better differentiated from their fully front (and
unrounded) /i/ phoneme(s).
Ruud Harmsen
2018-01-30 16:10:40 UTC
Permalink
Raw Message
Tue, 30 Jan 2018 05:50:18 -0800 (PST): "Peter T. Daniels"
Post by Ruud Harmsen
No. It's a back vowel, not central. As a phoneme, that is. It might
have allophones that are more central, because Turkish vowel have lots
of room: there are only four, on the four corners of the diagram (each
rounded and unrounded, which makes eight).
That makes 8, not 4. The rounded vowels are not allophones of the unrounded=
phonemes.
Indeed that is not what I meant to say, although admittedly I
expressed myself unclearly.
Yusuf B Gursey
2018-02-03 01:09:43 UTC
Permalink
Raw Message
In proper it is an unrounded back high vowel. It is however usually of shorter duration than the other vowels
Christian Weisgerber
2018-01-30 21:10:24 UTC
Permalink
Raw Message
Post by António Marques
Isn’t Turkish dotless-i supposed to be barred-u?
I highly recommend the vowel diagrams increasingly found on Wikipedia's
“<language> phonology” pages. Citing research papers, they show
the actual position of the vowels in vowel space, which frequently
diverges from that implied by the IPA symbol.

(Think of the ridiculous situation in English where the back vowel
symbol [ʌ] is habitually used for the STRUT vowel, although none
of General American, Received Pronunciation, or Australian English
actually have a back vowel there.)

https://en.wikipedia.org/wiki/Turkish_phonology#Vowels
For Turkish ı the symbol [ɯ] suggests a back vowel, although its
actual position appears more central than back. Similar position
to [ʉ], but unrounded.
--
Christian "naddy" Weisgerber ***@mips.inka.de
Ruud Harmsen
2018-01-31 15:36:53 UTC
Permalink
Raw Message
Tue, 30 Jan 2018 21:10:24 -0000 (UTC): Christian Weisgerber
Post by Christian Weisgerber
Isn’t Turkish dotless-i supposed to be barred-u?
I highly recommend the vowel diagrams increasingly found on Wikipedia's
“<language> phonology” pages.
So do I. The English Wikipedia nowadays is often very detailed,
accurate and well-founded. On this and a lot of other topics.

The German Wikipedia too, other languages less so. Best is to combine
knowledge from the articles in as many languages as one can read.
Post by Christian Weisgerber
Citing research papers, they show
the actual position of the vowels in vowel space, which frequently
diverges from that implied by the IPA symbol.
(Think of the ridiculous situation in English where the back vowel
symbol [?] is habitually used for the STRUT vowel, although none
of General American, Received Pronunciation, or Australian English
actually have a back vowel there.)
True.

Many (nothern or other) variaties of England English have, by the way.
--
Ruud Harmsen, http://rudhar.com
António Marques
2018-01-31 22:51:32 UTC
Permalink
Raw Message
Post by Ruud Harmsen
Tue, 30 Jan 2018 21:10:24 -0000 (UTC): Christian Weisgerber
Post by Christian Weisgerber
Isn’t Turkish dotless-i supposed to be barred-u?
I highly recommend the vowel diagrams increasingly found on Wikipedia's
“<language> phonology” pages.
So do I. The English Wikipedia nowadays is often very detailed,
accurate and well-founded. On this and a lot of other topics.
The German Wikipedia too, other languages less so. Best is to combine
knowledge from the articles in as many languages as one can read.
Unless the subject is specifically Portuguese or Brazilian in nature,
you’re well advised to stay clear of pt.wikipedia.
Post by Ruud Harmsen
Post by Christian Weisgerber
Citing research papers, they show
the actual position of the vowels in vowel space, which frequently
diverges from that implied by the IPA symbol.
(Think of the ridiculous situation in English where the back vowel
symbol [?] is habitually used for the STRUT vowel, although none
of General American, Received Pronunciation, or Australian English
actually have a back vowel there.)
True.
Many (nothern or other) variaties of England English have, by the way.
But those have phonologies that are wholesale different from ‘common
English’.
Ruud Harmsen
2018-01-31 15:46:46 UTC
Permalink
Raw Message
Tue, 30 Jan 2018 21:10:24 -0000 (UTC): Christian Weisgerber
Post by Christian Weisgerber
https://en.wikipedia.org/wiki/Turkish_phonology#Vowels
Yes. That diagram there,
https://en.wikipedia.org/wiki/Turkish_phonology#/media/File:Turkish_vowel_chart.svg
show a nice example of how the systematics in a language are still
valid, but actual realisation may be quite different, optimalising the
contrasts.

Turkish has eight vowels, resulting from 3 phonological contrasts:
back-front (a, dotless i, o, u -- e, i, ö, ü)
high-low (u, dotless i, ü, i -- o, a, ö, e)
rounded-unrounded (ü, u, ö, o -- i, dotless i, e, a)

but the reality results from quite some clarifying shifts, as shown in
the diagram. Most notably /a/ and /o/, which phonologically differ
only in rounding, are actually quite far apart on the front-back and
on the low-high (open-close) scales.
--
Ruud Harmsen, http://rudhar.com
Yusuf B Gursey
2018-02-01 21:16:20 UTC
Permalink
Raw Message
Post by Ruud Harmsen
Tue, 30 Jan 2018 21:10:24 -0000 (UTC): Christian Weisgerber
Post by Christian Weisgerber
https://en.wikipedia.org/wiki/Turkish_phonology#Vowels
Yes. That diagram there,
https://en.wikipedia.org/wiki/Turkish_phonology#/media/File:Turkish_vowel_chart.svg
show a nice example of how the systematics in a language are still
valid, but actual realisation may be quite different, optimalising the
contrasts.
back-front (a, dotless i, o, u -- e, i, ö, ü)
high-low (u, dotless i, ü, i -- o, a, ö, e)
rounded-unrounded (ü, u, ö, o -- i, dotless i, e, a)
but the reality results from quite some clarifying shifts, as shown in
the diagram. Most notably /a/ and /o/, which phonologically differ
only in rounding, are actually quite far apart on the front-back and
on the low-high (open-close) scales.
I think Turkish a is misrepresented and should be further. There is a rare
a found in some loanwords that is fronted and would correspond to what
is represented here.
Post by Ruud Harmsen
--
Ruud Harmsen, http://rudhar.com
Ruud Harmsen
2018-02-02 07:35:51 UTC
Permalink
Raw Message
Thu, 1 Feb 2018 13:16:20 -0800 (PST): Yusuf B Gursey
Post by Yusuf B Gursey
Post by Ruud Harmsen
https://en.wikipedia.org/wiki/Turkish_phonology#/media/File:Turkish_vowel_chart.svg
back-front (a, dotless i, o, u -- e, i, ö, ü)
high-low (u, dotless i, ü, i -- o, a, ö, e)
rounded-unrounded (ü, u, ö, o -- i, dotless i, e, a)
but the reality results from quite some clarifying shifts, as shown in
the diagram. Most notably /a/ and /o/, which phonologically differ
only in rounding, are actually quite far apart on the front-back and
on the low-high (open-close) scales.
I think Turkish a is misrepresented and should be further.
Further east? :)
Further back. I think so too, from what I have occasionally heard.
Post by Yusuf B Gursey
There is a rare a found in some loanwords that is fronted and would
correspond to what is represented here.
--
Ruud Harmsen, http://rudhar.com
Arnaud Fournet
2018-01-30 07:37:24 UTC
Permalink
Raw Message
Post by Ruud Harmsen
My theory to explain this, is in a natural tendency in languages of
phonemes to maintain maximum distance to other phonemes, because
maximum contrast makes recognition easier.
Swedish and Norwegian have a three way distinction in close rounded
vowels: <y> <u> <o>, front, central, back. <y> must remain fully front
to keep it apart from the central <u>, even though that makes /y/ more
similar to /i/.
In other languages (all others? I know of no other examples than
se/no, of languages in which /y/ is really phonetically front), like
German, French, Dutch, Danish, Turkish, Hungarian, because there is no
rounded central high vowel, but only rounded back high vowels, their
/y/ phoneme(s) "can afford" to drift more towards a central position,
so they are better differentiated from their fully front (and
unrounded) /i/ phoneme(s).
French u [ü] is definitely a front vowel,
it's exactly the same as i, but for lip rounding.
And this also applies to the pair eu vs é.
No doubt about that.
The drift toward central position is a fiction.
A.
Ruud Harmsen
2018-01-30 08:20:50 UTC
Permalink
Raw Message
Mon, 29 Jan 2018 23:37:24 -0800 (PST): Arnaud Fournet
French u [ü] is definitely a front vowel, it's exactly the same as i, but for lip rounding.
And this also applies to the pair eu vs é.
No doubt about that.
The drift toward central position is a fiction.
OK, if you say so, you're the expert, because you are French.

So now tell me, if in https://jbdowse.com/ipa you click on the y
symbol, in the upper left, does it sound like the vowel in French "la
lune"?

Then also try https://en.wikipedia.org/wiki/Swedish_phonology, last
row but one of the sample table, phonemes /y:/ and /y/. Please click
on the arrow to play the sample. Do you hear something similar to a
French "sule"? Or more like "sile"?

And the sample of Swedisch <ful>, is it more like French foule, or
French fule?

Also please compare Swedish <sil> (top row) with <syl>.
--
Ruud Harmsen, http://rudhar.com
Arnaud Fournet
2018-01-30 21:41:55 UTC
Permalink
Raw Message
Post by Ruud Harmsen
Mon, 29 Jan 2018 23:37:24 -0800 (PST): Arnaud Fournet
French u [ü] is definitely a front vowel, it's exactly the same as i, but for lip rounding.
And this also applies to the pair eu vs é.
No doubt about that.
The drift toward central position is a fiction.
OK, if you say so, you're the expert, because you are French.
So now tell me, if in https://jbdowse.com/ipa you click on the y
symbol, in the upper left, does it sound like the vowel in French "la
lune"?
There's clearly a problem in that table,
as sounds are paired in a way that does not make sense.
French i and ü, which differ in nothing else than lip-rounding (and not in tongue or jaw position) are not paired together correctly and are associated with two vowels that are clearly not the same,
French ü = Y is not English i plus lip-rounding, neither is their y the right partner of French i.
This table and recording are completely nonsensical.
A.
Post by Ruud Harmsen
Then also try https://en.wikipedia.org/wiki/Swedish_phonology, last
row but one of the sample table, phonemes /y:/ and /y/. Please click
on the arrow to play the sample. Do you hear something similar to a
French "sule"? Or more like "sile"?
And the sample of Swedisch <ful>, is it more like French foule, or
French fule?
Also please compare Swedish <sil> (top row) with <syl>.
The word herring is not /sɪlː/ with English ɪ, but French i, it's not far from French word cil "eye-lash".
To me, the distinction between syl "awl" and sil "sieve" is not lip-rounding as in French ü vs i but a kind of pharyngeal feature, syl sounds low-tone and pharyngealized, but not lip-rounded.
The description in this table is aberrant.
Actually the vowel closest to French ü is that of ful 'ugly' and I hear the same vowel as in German Gefühl. It has nothing to do with /fʉːl/ it's /fy(:)l/.
full, 'full' sounds like French foule, it's not /fɵlː/ but /fu:l/.
I completely disagree with the transcriptions.
A.
Ruud Harmsen
2018-01-31 14:30:31 UTC
Permalink
Raw Message
Tue, 30 Jan 2018 13:41:55 -0800 (PST): Arnaud Fournet
Post by Arnaud Fournet
Post by Ruud Harmsen
Mon, 29 Jan 2018 23:37:24 -0800 (PST): Arnaud Fournet
French u [ü] is definitely a front vowel, it's exactly the same as i, but for lip rounding.
And this also applies to the pair eu vs é.
No doubt about that.
The drift toward central position is a fiction.
OK, if you say so, you're the expert, because you are French.
So now tell me, if in https://jbdowse.com/ipa you click on the y
symbol, in the upper left, does it sound like the vowel in French "la
lune"?
There's clearly a problem in that table,
as sounds are paired in a way that does not make sense.
French i and ü, which differ in nothing else than lip-rounding (and not in tongue or jaw position) are not paired together correctly and are associated with two vowels that are clearly not the same,
French ü = Y is not English i plus lip-rounding, neither is their y the right partner of French i.
This table and recording are completely nonsensical.
Well, the tabel is the IPA, and the recordings are the consequence of
how the IPA vowels are described in articulatory terms.

My conclusion: because you misunderstand and misinterpret the
articulatory features of some French vowels (which is quite natural,
and expected to happen, because understanding those is especially
difficult for native speakers; and easier for non-natives), you demand
the IPA to be changed, even though the IPA and what it stands for, are
correct.

Verdict: your request is denied.
--
Ruud Harmsen, http://rudhar.com
Arnaud Fournet
2018-02-01 07:17:03 UTC
Permalink
Raw Message
Post by Ruud Harmsen
Tue, 30 Jan 2018 13:41:55 -0800 (PST): Arnaud Fournet
Post by Arnaud Fournet
Post by Ruud Harmsen
Mon, 29 Jan 2018 23:37:24 -0800 (PST): Arnaud Fournet
French u [ü] is definitely a front vowel, it's exactly the same as i, but for lip rounding.
And this also applies to the pair eu vs é.
No doubt about that.
The drift toward central position is a fiction.
OK, if you say so, you're the expert, because you are French.
So now tell me, if in https://jbdowse.com/ipa you click on the y
symbol, in the upper left, does it sound like the vowel in French "la
lune"?
There's clearly a problem in that table,
as sounds are paired in a way that does not make sense.
French i and ü, which differ in nothing else than lip-rounding (and not in tongue or jaw position) are not paired together correctly and are associated with two vowels that are clearly not the same,
French ü = Y is not English i plus lip-rounding, neither is their y the right partner of French i.
This table and recording are completely nonsensical.
Well, the tabel is the IPA, and the recordings are the consequence of
how the IPA vowels are described in articulatory terms.
My conclusion: because you misunderstand and misinterpret the
articulatory features of some French vowels (which is quite natural,
and expected to happen, because understanding those is especially
difficult for native speakers; and easier for non-natives), you demand
the IPA to be changed, even though the IPA and what it stands for, are
correct.
yes, you seem to be right, after all.
What is troublesome is that "serious" books on French phonetics completely misuse the IPA symbols, misleading people (like me) about what IPA y is supposed to be.
A.
Ruud Harmsen
2018-02-01 09:21:37 UTC
Permalink
Raw Message
Wed, 31 Jan 2018 23:17:03 -0800 (PST): Arnaud Fournet
Post by Arnaud Fournet
Post by Ruud Harmsen
Well, the tabel is the IPA, and the recordings are the consequence of
how the IPA vowels are described in articulatory terms.
My conclusion: because you misunderstand and misinterpret the
articulatory features of some French vowels (which is quite natural,
and expected to happen, because understanding those is especially
difficult for native speakers; and easier for non-natives), you demand
the IPA to be changed, even though the IPA and what it stands for, are
correct.
yes, you seem to be right, after all.
What is troublesome is that "serious" books on French phonetics completely misuse the IPA symbols, misleading people (like me) about what IPA y is supposed to be.
Using IPA for phonemes is legitimate (for example /y/ in French), but
it doesn't mean the actual phonetic realisation has to be exactly [y],
in those same IPA symbols. It would only be relevant if French has a
fully front rounded /y/ AND a more centralised rounded phoneme. But it
hasn't.
Ruud Harmsen
2018-01-31 15:02:29 UTC
Permalink
Raw Message
Post by Arnaud Fournet
Post by Ruud Harmsen
Then also try https://en.wikipedia.org/wiki/Swedish_phonology, last
row but one of the sample table, phonemes /y:/ and /y/. Please click
on the arrow to play the sample. Do you hear something similar to a
French "sule"? Or more like "sile"?
And the sample of Swedisch <ful>, is it more like French foule, or
French fule?
Also please compare Swedish <sil> (top row) with <syl>.
The word herring is not /sɪlː/ with English ɪ, but French i, it's not
far from French word cil "eye-lash".
Yes, I agree, I also hear [i] rather than [ɪ]. But maybe that difference is not really important in Swedish, I don't know.
Post by Arnaud Fournet
To me, the distinction between syl "awl" and sil "sieve" is
not lip-rounding as in French ü vs i but a kind of pharyngeal
feature, syl sounds low-tone and pharyngealized, but not
lip-rounded.
I tried to say <sil> /si:l/ as in the example, and then added pharyngalisation as in Arabic. But the result really sounds different from the <syl> /sy:l/.

On the other hand, if I pronounce a Dutch <uu> (with quite a lot of lip-rounding) and continue that sound, then try to fully front that vowel, without changing anything to the lips position, I get a result that is very much like the sample of <syl>. It is a very un-Dutch vowel, very strange, not known (to me) from any other languages than Swedish and Norwegian.

The same result of course should be obtainable by starting from a Dutch <ie> = [i], making it overlong, and then adding lip-rounding WITHOUT doing the slight centralisation that is automatic in Dutch and French and lots of other languages. That is hard for me, because the connexion between the two (lip rounding and centralisation) is so automatic and unconscious.

Yet, when really concentrating, I can do that too, and the result is, again, a Swedish /y:/. Very un-Dutch, but clearly different from unrounded [i::].
Post by Arnaud Fournet
The description in this table is aberrant.
Actually the vowel closest to French ü is that of ful 'ugly' and
I hear the same vowel as in German Gefühl.
That's right, Germand and French (and Danish, Dutch, Turkish and Hungarian do not differ in this respect.
Post by Arnaud Fournet
It has nothing to do with /fʉːl/
That's right, because that Swedish vowel (written <u>) is really central, whereas the German, French etc. /y/ phonemes are realised somewhere in between central and front.
Post by Arnaud Fournet
it's /fy(:)l/.
It is, phonologically, but the realisation of such phonemes (in French, German etc.) is NOT IPA [y(:)]. The Swedish and Norwegian phoneme realisations are (or very nearly so).
Post by Arnaud Fournet
full, 'full' sounds like French foule, it's not /fɵlː/ but /fu:l/.
Now you are confusing French rounded back vowels with Swedish rounded central vowels. That's understandable, because French does not have such vowels, so you mind has to look for a nearest 'equivalent' in your language. Then you think they're the same, but they are not.

You can hear that from the Swedish /u:/ phoneme, spelt <o>. That one is really back, like the <ou> in French <foule>.
Post by Arnaud Fournet
I completely disagree with the transcriptions.
The transcriptions are correct, but your interpretation ability is influenced by your native language. So is mine (but then by Dutch, of course) -- but I trained myself to be able to switch that natural effect off temporarily. It's hard, but it can be done.
Peter T. Daniels
2018-01-30 13:48:21 UTC
Permalink
Raw Message
Post by Arnaud Fournet
Post by Ruud Harmsen
My theory to explain this, is in a natural tendency in languages of
phonemes to maintain maximum distance to other phonemes, because
maximum contrast makes recognition easier.
Swedish and Norwegian have a three way distinction in close rounded
vowels: <y> <u> <o>, front, central, back. <y> must remain fully front
to keep it apart from the central <u>, even though that makes /y/ more
similar to /i/.
In other languages (all others? I know of no other examples than
se/no, of languages in which /y/ is really phonetically front), like
German, French, Dutch, Danish, Turkish, Hungarian, because there is no
rounded central high vowel, but only rounded back high vowels, their
/y/ phoneme(s) "can afford" to drift more towards a central position,
so they are better differentiated from their fully front (and
unrounded) /i/ phoneme(s).
French u [ü] is definitely a front vowel,
it's exactly the same as i, but for lip rounding.
And this also applies to the pair eu vs é.
No doubt about that.
The drift toward central position is a fiction.
He is not talking about phonemics, but about phonetics.
Ruud Harmsen
2018-01-30 16:15:54 UTC
Permalink
Raw Message
Tue, 30 Jan 2018 05:48:21 -0800 (PST): "Peter T. Daniels"
Post by Peter T. Daniels
Post by Arnaud Fournet
French u [ü] is definitely a front vowel,
it's exactly the same as i, but for lip rounding.
And this also applies to the pair eu vs =C3=A9.
No doubt about that.
The drift toward central position is a fiction.
He is not talking about phonemics, but about phonetics.
Both. The tendency to obtain optimal contrast has to do with
phonemics. Phonetics in its role to encode information is phonemics.

The difference between Dutch <ie> and <uu> isn't just lip rounding,
and with good reason. The centralisation is just as important for
recognition, which is proven by the fact that I tend to interpret a
Swedish /y/ as Dutch /i/, until I hear a real Swedish /i/ and must
admit that it differs from Swedish /y/.
Christian Weisgerber
2018-01-30 21:16:10 UTC
Permalink
Raw Message
Post by Arnaud Fournet
French u [ü] is definitely a front vowel,
it's exactly the same as i, but for lip rounding.
And this also applies to the pair eu vs é.
No doubt about that.
The drift toward central position is a fiction.
Actual phoneticians seem to differ:
https://en.wikipedia.org/wiki/French_phonology#Vowels
--
Christian "naddy" Weisgerber ***@mips.inka.de
Ruud Harmsen
2018-01-31 15:56:15 UTC
Permalink
Raw Message
Tue, 30 Jan 2018 21:16:10 -0000 (UTC): Christian Weisgerber
Post by Christian Weisgerber
Post by Arnaud Fournet
French u [ü] is definitely a front vowel,
it's exactly the same as i, but for lip rounding.
And this also applies to the pair eu vs é.
No doubt about that.
The drift toward central position is a fiction.
https://en.wikipedia.org/wiki/French_phonology#Vowels
Now compare this with
https://en.wikipedia.org/wiki/Dutch_phonology#Monophthongs
https://en.wikipedia.org/wiki/Swedish_phonology#Vowels ,
then we can understand why I (Ruud Harmsen) tend to misinterpret
Swedish <u> as Dutch /y/ (in the diagrams, they are actually in
exactly the same place!), and why Arnaud Fournet misinterprets that
same Swedish <u> as French <ou> = /u/. That's because the French /y/
is less centralised than the Dutch /y/, and French /u/ is slightly
centralised while Dutch /u/ is not.

But then, according to these diagrams, Swedish /y:/ and /y/ should
have the same sound (barring length) as French /y/, which everybody
agrees is not true.
--
Ruud Harmsen, http://rudhar.com
Peter T. Daniels
2018-01-29 21:25:09 UTC
Permalink
Raw Message
Post by Ruud Harmsen
Post by Ruud Harmsen
Sun, 28 Jan 2018 21:41:03 -0800 (PST): Dingbat
Post by Dingbat
German <ü> I hear as similar to Swedish <y>
If you listen to a speaking IPA charts (links here
http://rudhar.com/foneport/en/lingglos.htm), you;ll find that cardinal
[y] sounds quite like a Swedish/Norwegian <y>, but nothing like a
German etc. <ü>.
It depends who recorded them. Daniel Jones, who invented the concept, made several recordings
over the decades, and they differ noticeably.
Post by Ruud Harmsen
Also, the opener vowels in the series "front rounded", which are
supposed to occur in French (e.g. in "un peu", "coeur"), don't sound
like that at all. Because the French vowel aren't front, but more in
the direction of central, without being fully central.
Ruud Harmsen
2018-01-30 08:29:49 UTC
Permalink
Raw Message
Mon, 29 Jan 2018 13:25:09 -0800 (PST): "Peter T. Daniels"
Post by Peter T. Daniels
Post by Ruud Harmsen
Post by Ruud Harmsen
Sun, 28 Jan 2018 21:41:03 -0800 (PST): Dingbat
Post by Dingbat
German <ü> I hear as similar to Swedish <y>
If you listen to a speaking IPA charts (links here
http://rudhar.com/foneport/en/lingglos.htm), you;ll find that cardinal
[y] sounds quite like a Swedish/Norwegian <y>, but nothing like a
German etc. <ü>.
It depends who recorded them. Daniel Jones, who invented the concept, made several recordings
over the decades, and they differ noticeably.
Peter Ladefoged didn't do it right, here:
http://www.phonetics.ucla.edu/course/chapter1/vowels.html
Not front enough.

The 1956 Daniel Jones samples I found, unfortunately only contain the
first 8, not [y] which is probably no. 9.
Post by Peter T. Daniels
Post by Ruud Harmsen
Also, the opener vowels in the series "front rounded", which are
supposed to occur in French (e.g. in "un peu", "coeur"), don't sound
like that at all. Because the French vowel aren't front, but more in
the direction of central, without being fully central.
--
Ruud Harmsen, http://rudhar.com
Ruud Harmsen
2018-01-30 08:36:50 UTC
Permalink
Raw Message
Post by Ruud Harmsen
http://www.phonetics.ucla.edu/course/chapter1/vowels.html
Not front enough.
The 1956 Daniel Jones samples I found, unfortunately only contain the
first 8, not [y] which is probably no. 9.
Here's the full set of 18 cardinal vowel.


From 1:17. Daniel Jones did it right: definitely the Swedish type
vowel, not 'la lune' or "die Düse" of German. And also definitely not
the <uu> in Ruud and (the Dutch pronunciation of) Rudolf.

At 2:02 the Swedish <u>, which is much more like a Dutch/German/French
/y/, although "too dark" for that.
--
Ruud Harmsen, http://rudhar.com
Ruud Harmsen
2018-01-30 08:51:25 UTC
Permalink
Raw Message
Post by Ruud Harmsen
Post by Ruud Harmsen
http://www.phonetics.ucla.edu/course/chapter1/vowels.html
Not front enough.
The 1956 Daniel Jones samples I found, unfortunately only contain the
first 8, not [y] which is probably no. 9.
Here's the full set of 18 cardinal vowel.
http://youtu.be/6UIAe4p2I74
Also http://audiufon.hum.uu.nl/data/e_secundary_long.html .
--
Ruud Harmsen, http://rudhar.com
Peter T. Daniels
2018-01-30 13:56:40 UTC
Permalink
Raw Message
Post by Ruud Harmsen
Mon, 29 Jan 2018 13:25:09 -0800 (PST): "Peter T. Daniels"
Post by Peter T. Daniels
Post by Ruud Harmsen
Post by Ruud Harmsen
Sun, 28 Jan 2018 21:41:03 -0800 (PST): Dingbat
Post by Dingbat
German <ü> I hear as similar to Swedish <y>
If you listen to a speaking IPA charts (links here
http://rudhar.com/foneport/en/lingglos.htm), you;ll find that cardinal
[y] sounds quite like a Swedish/Norwegian <y>, but nothing like a
German etc. <ü>.
It depends who recorded them. Daniel Jones, who invented the concept, made several recordings
over the decades, and they differ noticeably.
http://www.phonetics.ucla.edu/course/chapter1/vowels.html
Not front enough.
Peter Ladefoged was one of Jones's last students. If anyone of his generation could
reproduce Jones's (last) conception of the concept, it was he.
Post by Ruud Harmsen
The 1956 Daniel Jones samples I found, unfortunately only contain the
first 8, not [y] which is probably no. 9.
One of Ladefoged's and/or Maddieson's books included a CD with three of Jones's, one of
Ladefoged's, and maybe a couple of others. I don't know what happened to my copy -- and
can't determine which book it was packaged with. (Most of the CD was native speakers
demonstrating "exotic" sounds.) The surviving copies of Jones's original 78s weren't in great
condition. They may well date from the "acoustic" and not yet the "electric" era of recording.
Post by Ruud Harmsen
Post by Peter T. Daniels
Post by Ruud Harmsen
Also, the opener vowels in the series "front rounded", which are
supposed to occur in French (e.g. in "un peu", "coeur"), don't sound
like that at all. Because the French vowel aren't front, but more in
the direction of central, without being fully central
The two mid front rounded vowels of French were merging already 50 years ago when we were being
taught the difference but told that the only place we might hear it was at the Comédie Française.
Ruud Harmsen
2018-01-30 16:20:41 UTC
Permalink
Raw Message
Tue, 30 Jan 2018 05:56:40 -0800 (PST): "Peter T. Daniels"
Post by Peter T. Daniels
Post by Ruud Harmsen
Mon, 29 Jan 2018 13:25:09 -0800 (PST): "Peter T. Daniels"
Post by Peter T. Daniels
Post by Ruud Harmsen
Post by Ruud Harmsen
Sun, 28 Jan 2018 21:41:03 -0800 (PST): Dingbat
Post by Dingbat
German <ü> I hear as similar to Swedish <y>
If you listen to a speaking IPA charts (links here
http://rudhar.com/foneport/en/lingglos.htm), you;ll find that cardinal
[y] sounds quite like a Swedish/Norwegian <y>, but nothing like a
German etc. <ü>.
It depends who recorded them. Daniel Jones, who invented the concept, made several recordings
over the decades, and they differ noticeably.
http://www.phonetics.ucla.edu/course/chapter1/vowels.html
Not front enough.
Peter Ladefoged was one of Jones's last students. If anyone of his generation could
reproduce Jones's (last) conception of the concept, it was he.
No two speech organs are the same, and all articulatory
characteristics are gliding scales and binary oppositions. And
nobody's perfect.

Here's a comparative talking chart, by three different speakers:
http://www.phonetics.ucla.edu/course/chapter1/wells/wells.html

There are notable differences in some places.
Arnaud Fournet
2018-01-30 07:40:28 UTC
Permalink
Raw Message
Post by Ruud Harmsen
Post by Ruud Harmsen
Sun, 28 Jan 2018 21:41:03 -0800 (PST): Dingbat
Post by Dingbat
German <ü> I hear as similar to Swedish <y>
If you listen to a speaking IPA charts (links here
http://rudhar.com/foneport/en/lingglos.htm), you;ll find that cardinal
[y] sounds quite like a Swedish/Norwegian <y>, but nothing like a
German etc. <ü>.
Also, the opener vowels in the series "front rounded", which are
supposed to occur in French (e.g. in "un peu", "coeur"), don't sound
like that at all. Because the French vowel aren't front, but more in
the direction of central, without being fully central.
This is nonsense.
A.
Ruud Harmsen
2018-01-30 08:37:50 UTC
Permalink
Raw Message
Mon, 29 Jan 2018 23:40:28 -0800 (PST): Arnaud Fournet
Post by Arnaud Fournet
Post by Ruud Harmsen
Post by Ruud Harmsen
Sun, 28 Jan 2018 21:41:03 -0800 (PST): Dingbat
Post by Dingbat
German <ü> I hear as similar to Swedish <y>
If you listen to a speaking IPA charts (links here
http://rudhar.com/foneport/en/lingglos.htm), you;ll find that cardinal
[y] sounds quite like a Swedish/Norwegian <y>, but nothing like a
German etc. <ü>.
Also, the opener vowels in the series "front rounded", which are
supposed to occur in French (e.g. in "un peu", "coeur"), don't sound
like that at all. Because the French vowel aren't front, but more in
the direction of central, without being fully central.
This is nonsense.
Before of after you actually listened?
--
Ruud Harmsen, http://rudhar.com
Ruud Harmsen
2018-01-30 16:30:55 UTC
Permalink
Raw Message
About the two sound for French <eu>, one half close, one half open:
clear difference here:

visible and audible.

It may not be phonemic, because generally the opener sound occurs in
closed syllable (when considering the actual pronunciation, not the
spelling) (in coleur, jeune, beure) and more close in open syllables
(peu, blue, feu, veut, veux, jeu(x)).
Ruud Harmsen
2018-01-30 16:32:52 UTC
Permalink
Raw Message
Post by Ruud Harmsen
clear difference here: http://youtu.be/n94F-DMvvLM
visible and audible.
It may not be phonemic, because generally the opener sound occurs in
closed syllable (when considering the actual pronunciation, not the
spelling) (in coleur, jeune, beure) and more close in open syllables
(peu, blue, feu, veut, veux, jeu(x)).
See also
same opener
sound, different spelling.
Dingbat
2018-01-30 07:19:56 UTC
Permalink
Raw Message
Post by Ruud Harmsen
Sun, 28 Jan 2018 21:41:03 -0800 (PST): Dingbat
Post by Dingbat
German <ü> I hear as similar to Swedish <y>
I don't.
The word that brought out the similarity was über. Audio clip:
https://d2db9axx8hb6k.cloudfront.net/audio/lessons/3/31.mp3

In Glück, ü sounds nothing like a Swedish <y>. I haven;t saved a link to a clip for this word.
Post by Ruud Harmsen
The Swedish (and Norwegian) vowel is really front, the German
(and French, Danish, Turkish, Hungarian, Dutch) vowel is almost front,
or between front and central. That is why they sound so different.
Post by Dingbat
German <ö>, when long, I hear as similar to Swedish <u>, not Swedish <ö>;
Likewise, the German ö (and Turkish, Hungarian etc.) vowels are not
front, but almost front. I'm not sure if the Swedish ö is front, but
it might.
Post by Dingbat
i.e., I don't hear in German <schön> the sound of the first vowel in
Swedish <Öland>
Has anyone else made similar observations?
I hear Swedish <u> (which is central) more like Dutch <uu> (almost
front), than Swedish <y> (front).
--
Ruud Harmsen, http://rudhar.com
Arnaud Fournet
2018-01-30 07:50:39 UTC
Permalink
Raw Message
Post by Dingbat
German <ü> I hear as similar to Swedish <y>
German <ö>, when long, I hear as similar to Swedish <u>, not Swedish <ö>;
i.e., I don't hear in German <schön> the sound of the first vowel in
Swedish <Öland>
Has anyone else made similar observations?
I'm not familiar with Swedish,
judging from the recordings on translate.google, Swedish ö is indeed not the same sound as German schön,
to me Swedish öl sounds like a diphthong eu-è-l
It's even worse with Norwegian öl, to me it's vowel è not eu. This vowel is not rounded at all.
A.
Ruud Harmsen
2018-01-30 08:50:28 UTC
Permalink
Raw Message
Mon, 29 Jan 2018 23:50:39 -0800 (PST): Arnaud Fournet
Post by Arnaud Fournet
I'm not familiar with Swedish,
judging from the recordings on translate.google, Swedish ö is indeed not the same sound as German schön,
to me Swedish öl sounds like a diphthong eu-è-l
It's even worse with Norwegian öl, to me it's vowel è not eu. This vowel is not rounded at all.
It is rounded, but more front than you are used to. Because in French
the rounding and the slight centralisation are always coupled, you are
unable to hear and make the two effects separately. Anything you hear
tends to be modelled by the first language you learnt. It's only
natural.
--
Ruud Harmsen, http://rudhar.com
Dingbat
2018-01-30 11:48:06 UTC
Permalink
Raw Message
Post by Arnaud Fournet
Post by Dingbat
German <ü> I hear as similar to Swedish <y>
German <ö>, when long, I hear as similar to Swedish <u>, not Swedish <ö>;
i.e., I don't hear in German <schön> the sound of the first vowel in
Swedish <Öland>
Has anyone else made similar observations?
I'm not familiar with Swedish,
judging from the recordings on translate.google, Swedish ö is indeed not the same sound as German schön,
Öland is pronounced (.ogg clip) at this link:
https://en.wikipedia.org/wiki/%C3%96land
Post by Arnaud Fournet
to me Swedish öl sounds like a diphthong eu-è-l
It's even worse with Norwegian öl, to me it's vowel è not eu. This vowel is not rounded at all.
A.
https://en.wikipedia.org/wiki/%C3%96land
Arnaud Fournet
2018-01-30 21:46:14 UTC
Permalink
Raw Message
Post by Dingbat
Post by Arnaud Fournet
Post by Dingbat
German <ü> I hear as similar to Swedish <y>
German <ö>, when long, I hear as similar to Swedish <u>, not Swedish <ö>;
i.e., I don't hear in German <schön> the sound of the first vowel in
Swedish <Öland>
Has anyone else made similar observations?
I'm not familiar with Swedish,
judging from the recordings on translate.google, Swedish ö is indeed not the same sound as German schön,
https://en.wikipedia.org/wiki/%C3%96land
Post by Arnaud Fournet
to me Swedish öl sounds like a diphthong eu-è-l
It's even worse with Norwegian öl, to me it's vowel è not eu. This vowel is not rounded at all.
A.
https://en.wikipedia.org/wiki/%C3%96land
ok, this sounds like French euh.
It's indeed ö = French <eu>
I suppose google.translate recordings are inadequate.
A.
Mścisław Wojna-Bojewski
2018-01-30 20:53:54 UTC
Permalink
Raw Message
Post by Arnaud Fournet
Post by Dingbat
German <ü> I hear as similar to Swedish <y>
German <ö>, when long, I hear as similar to Swedish <u>, not Swedish <ö>;
i.e., I don't hear in German <schön> the sound of the first vowel in
Swedish <Öland>
Has anyone else made similar observations?
I'm not familiar with Swedish,
judging from the recordings on translate.google, Swedish ö is indeed not the same sound as German schön,
to me Swedish öl sounds like a diphthong eu-è-l
Many Swedish dialects both in Sweden and in Finland diphthongize such vowels.
Arnaud Fournet
2018-01-30 21:50:26 UTC
Permalink
Raw Message
Post by Mścisław Wojna-Bojewski
Post by Arnaud Fournet
Post by Dingbat
German <ü> I hear as similar to Swedish <y>
German <ö>, when long, I hear as similar to Swedish <u>, not Swedish <ö>;
i.e., I don't hear in German <schön> the sound of the first vowel in
Swedish <Öland>
Has anyone else made similar observations?
I'm not familiar with Swedish,
judging from the recordings on translate.google, Swedish ö is indeed not the same sound as German schön,
to me Swedish öl sounds like a diphthong eu-è-l
Many Swedish dialects both in Sweden and in Finland diphthongize such vowels.
ok, that's clearly what I perceive.
A.
Ruud Harmsen
2018-01-31 15:02:59 UTC
Permalink
Raw Message
Tue, 30 Jan 2018 13:50:26 -0800 (PST): Arnaud Fournet
Post by Arnaud Fournet
Post by Mścisław Wojna-Bojewski
Post by Arnaud Fournet
Post by Dingbat
German <ü> I hear as similar to Swedish <y>
German <ö>, when long, I hear as similar to Swedish <u>, not Swedish <ö>;
i.e., I don't hear in German <schön> the sound of the first vowel in
Swedish <Öland>
Has anyone else made similar observations?
I'm not familiar with Swedish,
judging from the recordings on translate.google, Swedish ö is indeed not the same sound as German schön,
to me Swedish öl sounds like a diphthong eu-è-l
Many Swedish dialects both in Sweden and in Finland diphthongize such vowels.
ok, that's clearly what I perceive.
Me too.
--
Ruud Harmsen, http://rudhar.com
Hans Aberg
2018-02-02 21:54:09 UTC
Permalink
Raw Message
Post by Dingbat
German <ü> I hear as similar to Swedish <y>
German <ö>, when long, I hear as similar to Swedish <u>, not Swedish <ö>;
i.e., I don't hear in German <schön> the sound of the first vowel in
Swedish <Öland>
Has anyone else made similar observations?
The Swedish letters åäö are not diacritics, but ligatures, and separate
letters from ao with different alphabetical sorting order, not "umlaut"
as in German means inflection vowel alterations.

There are some discussions of Swedish and Germany phonology at [1-2]: [
[yː] and [øː] are similar, but Swedish long "u" has no German
counterpart, and the short one, [2] gives as a diphthong. This is for
the main Swedish dialect, which does not have diphthongs, but some
dialects do, for example the Scanian dialect (Skånska) [4].


1. https://en.wikipedia.org/wiki/Swedish_phonology
2. https://en.wikipedia.org/wiki/Standard_German_phonology
3. https://en.wikipedia.org/wiki/Standard_Swedish
4. https://en.wikipedia.org/wiki/Scanian_dialect
Loading...