Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Scandinavian letter history?

3 views
Skip to first unread message

Jim Heckman

unread,
Sep 7, 2002, 7:26:37 AM9/7/02
to
Pursuant to a recent thread in news:de.etc.sprache.misc
re the Danish spelling reform in the 40's, I realized I know
almost nothing about the history behind the
pan-mainland-Scandinavian grapheme <å> -- other than
that it's etymologically from earlier <aa> (= /a:/ -> /O/?) --
and Danish/Norwegian <ø>.

So what's the story? When did each of these come into
use, and where did they come from? (At a wild guess, is
<å> simply <a> with <o> on top to represent the phonetic
shift, and <ø> just a different development of <oe> than
happened in German/Swedish <ö>?

While we're on the subject, is it safe to assume that
Swedish <ä> and Swedish/Icelandic <ö> were borrowed
from -- or developed in parallel with -- German, while
Danish/Norwegian/Icelandic <æ> and Icelandic <ð> and
<þ> have the same origin as those graphemes in Old
English (<þ> being originally a Germanic rune)?

Final question: How did <y> come to represent the high
front rounded vowel? -- uniquely in Scandinavian as far
as I know.

--
Jim Heckman

Yusuf B Gursey

unread,
Sep 7, 2002, 11:42:46 AM9/7/02
to
Jim Heckman <wnzrfe...@lnubb.pbz> wrote:

: Final question: How did <y> come to represent the high


: front rounded vowel? -- uniquely in Scandinavian as far
: as I know.

i.e. like u"

that's its accepted pronounciation in classical greek.

: --
: Jim Heckman

Henry Churchyard

unread,
Sep 7, 2002, 1:06:38 PM9/7/02
to
In article <unjoirh...@corp.supernews.com>,
Jim Heckman <wnzrfe...@lnubb.pbz> wrote:

> While we're on the subject, is it safe to assume that Danish/


> Norwegian/Icelandic <æ> and Icelandic <ð> and <þ> have the same
> origin as those graphemes in Old English (<þ> being originally a
> Germanic rune)? Final question: How did <y> come to represent the
> high front rounded vowel? -- uniquely in Scandinavian as far as I
> know.

The thing is that Old English was the first Germanic vernacular (by a
hundred years or more) to have a somewhat fixed Latin-alphabet
orthography, and to start building up a body of written literature.
For this reason, the orthography of Old Norse was initially heavily
influenced by the orthography of Old English, and Icelandic has kept
most of the orthography (but not the pronunciation!) of Old Norse.
All the spelling habits you mention above originated in Old English
(as far as Latin-alphabet orthographies of Germanic languages go).

--
Henry Churchyard chu...@crossmyt.com http://www.crossmyt.com/hc/

Yusuf B Gursey

unread,
Sep 7, 2002, 1:37:34 PM9/7/02
to
Yusuf B Gursey <y...@shell01.theworld.com> wrote:
: Jim Heckman <wnzrfe...@lnubb.pbz> wrote:

: i.e. like u"

west greek upsilon became latin V for /u/ and /w/. later latin borrowed
east greek upsilon Y for classical greek /u"/, a foreign sound in latin.


: : --
: : Jim Heckman

Gyula Zsigri

unread,
Sep 7, 2002, 1:45:34 PM9/7/02
to
Yusuf B Gursey <y...@shell01.TheWorld.com> írja:

Greek loanwords with <y> are still pronounced with an [y]
(orthographic <ü>) in German.


Christian Weisgerber

unread,
Sep 7, 2002, 3:13:19 PM9/7/02
to
Yusuf B Gursey <y...@shell01.TheWorld.com> wrote:

> : Final question: How did <y> come to represent the high
> : front rounded vowel? -- uniquely in Scandinavian as far
> : as I know.
>
> i.e. like u"
> that's its accepted pronounciation in classical greek.

It is also the primary pronunciation of <y> in German.

--
Christian "naddy" Weisgerber na...@mips.inka.de

Jukka K. Korpela

unread,
Sep 7, 2002, 4:51:06 PM9/7/02
to
"Jim Heckman" <wnzrfe...@lnubb.pbz> wrote:

> (At a wild guess, is
> <å> simply <a> with <o> on top to represent the phonetic
> shift, and <ø> just a different development of <oe> than
> happened in German/Swedish <ö>?

You guessed right. The character å (a with circle above) was formed from "ao"
(by writing the "o" above the "a") in the 15th century for use in Swedish
orthography. It denotes a sound roughly similar to the one denoted by "a" in
English "hall". The character was taken into use in Norwegian in 1918 and in
Danish in 1948, replacing "aa". For example, the name Håkon was previously
written Haakon, and the old spelling is still widely used e.g. in texts in
English. The character å is also used in Walloon, but I don't know how and
when it came into use there.

The medieval style of writing "oe" as a ligature was preserved in French but
in some areas the "e" was written above the "o", later changed to a dieresis,
resulting in "ö". Presumably this took place in German, from which it was
adopted to Swedish, Hungarian etc. In Danish and Norwegian, the ligature
changed to "ø" (o with stroke), where the stroke represents the original "e"
component, in a sense.

> Danish/Norwegian/Icelandic <æ> and Icelandic <ð> and
> <þ> have the same origin as those graphemes in Old
> English (<þ> being originally a Germanic rune)?

Yes.

--
Yucca, http://www.cs.tut.fi/~jkorpela/

Torsten

unread,
Sep 7, 2002, 8:22:45 PM9/7/02
to
Jim Heckman skrev:

> is [...] <ø> just a different development of <oe> than


> happened in German/Swedish <ö>?

Probably not. Looking at the lyrics of a song in a medieval
Danish manuscript written with runes reveals that it too used a
seperate grapheme for <ø>: drømde mik en drøm i nat om silki ok
ærlik pæl.

The runes used for <ø> and <o> look, respectively, roughly like
this:

# #
# #
####### ####
# #
####### ####
# #
# #

The use of a stroke was not unique to the letter <ø> in Danish.
For a while <y> had a competitor in Danish manuscripts: a <u>
with a horizontal bar looking like this:

# #
# #
#########
# #
# #
#### #

Rather than being variants of <oe> and <ue> as <ö> and <ü> are,
they may owe their origin to a practice similar to that of the
dotted runes. Note that early Swedish texts also used <ø> and
<æ>.

--
Torsten

Brian M. Scott

unread,
Sep 7, 2002, 10:26:49 PM9/7/02
to
On Sat, 7 Sep 2002 11:26:37 GMT, "Jim Heckman"
<wnzrfe...@lnubb.pbz> wrote:

>Pursuant to a recent thread in news:de.etc.sprache.misc
>re the Danish spelling reform in the 40's, I realized I know
>almost nothing about the history behind the
>pan-mainland-Scandinavian grapheme <å> -- other than
>that it's etymologically from earlier <aa> (= /a:/ -> /O/?) --

According to Haugen's Scandinavian Language Structures, it's mostly
from OWSc/OESc /O:/, from /a:/ by u-umlaut; it corresponds pretty well
to Icelandic and OIc <á>.

>and Danish/Norwegian <ø>.

That last can't be right: Norwegian has <å> as well (in both
varieties), and so has Danish. Danish and Norwegian <ø> mostly
represents earlier /ø/ and /ø:/, from i-umlaut of /o/ and /o:/.

>So what's the story? When did each of these come into
>use, and where did they come from? (At a wild guess, is
><å> simply <a> with <o> on top to represent the phonetic
>shift,

It's certainly possible; if memory serves, the First Grammarian
described /O:/ as a blending of the sounds of <a> and <o>. Ah, here
we go: '<Ö> hefir lykkju af <ae>, en hringinn af <oe>, því at hann er
af þeira hljóði tveggja saman blandinn, kveðinn minnr opnum munni en
<a>, en meirr en <o>'. (This is from the online version at
<http://home.nvg.org/~gjerde/norn/gramm/>, which substitutes available
characters for those of the MS.) '<Ö> has the loop of <ae> and the
ring of <oe>, because it is blended of these two sounds, pronounced
with less open mouth than <a> but more than <o>.' (I'm not entirely
certain of the sense of the first clause, but I'm pretty sure that it
refers to the shape of the MS. letter, which in any case varied
considerably.)

>and <ø> just a different development of <oe> than
>happened in German/Swedish <ö>?

Not clear; see below.

>While we're on the subject, is it safe to assume that
>Swedish <ä> and Swedish/Icelandic <ö> were borrowed
>from -- or developed in parallel with -- German, while
>Danish/Norwegian/Icelandic <æ> and Icelandic <ð> and
><þ> have the same origin as those graphemes in Old
>English (<þ> being originally a Germanic rune)?

Thorn and edh were borrowed from Old and early Middle English. Gordon
says that <ø> occurs in a few 11th c. OE MSS., that this seems to be
its source in Denmark, and that its use then spread from Denmark
throughout Scandinavia. I've not encountered the late OE <ø>, and I
don't know how it was used.

>Final question: How did <y> come to represent the high
>front rounded vowel? -- uniquely in Scandinavian as far
>as I know.

Well, it was originally borrowed from Greek upsilon, representing
/y(:)/. The First Grammarian says: '<Y> er af röddu <is> ok <us> görr
at einni röddu, kveðinn minnr opnum munni en <i> ok meirr en <u>, ...
': '<Y> is a single vowel made from the vowels <i> and <u>, pronounced
with a less open mouth than <i> and more than <u>, ...'. He also
knows that it was Greek: '<Y>, hann er grikkskr stafr ok heitir þar
<ui>, en látínumenn hafa hann fyr <i>, ok í grikkskum orðum at eins
þó, ef skynsamliga er ritit': '<Y>, that is a Greek letter and is
called there <ui>, but Latin-men have it for <i>, and only in Greek
words, if writing rationally'.

Brian

Jim Heckman

unread,
Sep 8, 2002, 5:36:31 AM9/8/02
to
Thanks to everyone who responded!

--
Jim Heckman

Philip Newton

unread,
Sep 8, 2002, 7:52:47 AM9/8/02
to
On Sat, 7 Sep 2002 19:45:34 +0200, "Gyula Zsigri"
<zsi...@hung.u-szeged.hu> wrote:

> Greek loanwords with <y> are still pronounced with an [y]
> (orthographic <ü>) in German.

Or [I.] (especially in closed syllables?), as in "Ypsilon". (That's IPA
small cap Y.)

Cheers,
Philip
--
Philip Newton <nospam...@gmx.li>
That really is my address; no need to remove anything to reply.
If you're not part of the solution, you're part of the precipitate.

Prai Jei

unread,
Sep 8, 2002, 12:12:56 PM9/8/02
to
Translation please? "I dreamed a dream in [the] night about silk and ... ?"

"Torsten" <t_usen...@hotmail.com> wrote in message
news:ale58l$15vm$1...@news.cybercity.dk...

Brian M. Scott

unread,
Sep 8, 2002, 12:37:41 PM9/8/02
to
On Sun, 8 Sep 2002 17:12:56 +0100, "Prai Jei"
<pvsto...@prai-jei.fsnet.co.uk> wrote:

>"Torsten" <t_usen...@hotmail.com> wrote in message
>news:ale58l$15vm$1...@news.cybercity.dk...

>> drømde mik en drøm i nat om silki ok ærlik pæl.

>Translation please? "I dreamed a dream in [the] night about silk and ... ?"

... fine fur. (And <i nat> is probably better translated 'last
night', though Old Norse <í nátt> can be either 'tonight' or
'last night'.)

You can see the relevant part of the runic MS. and hear the
melody that goes with it at
<http://www.hum.ku.dk/ami/28-100r-dt.html>.

Brian

Torsten

unread,
Sep 8, 2002, 5:38:35 PM9/8/02
to
Brian M. Scott skrev:

>>> drømde mik en drøm i nat om silki ok ærlik pæl.

>> Translation please? "I dreamed a dream in [the] night about
>> silk and ... ?"

> ... fine fur. (And <i nat> is probably better translated 'last
> night', though Old Norse <í nátt> can be either 'tonight' or
> 'last night'.)

Yes. Even in contemporary Danish <i nat> means both things. Allan
Karker¹ translates <ærlik pæl> as 'prægtig brokade', i.e., fine
or magnificent brocade. A possible translation could then be
'dreamed me a dream last night about silk and fine brocade'.
Otto Kalkar's 'Ordbog til det ældre danske sprog (1300-1700)'
published in four volumes from 1881-1907 unfortunately does not
resolve the problem of the meaning of <pæl>. Fur is 'pels' in
Danish. From latin 'pellicia' via Low German.

> You can see the relevant part of the runic MS. and hear the
> melody that goes with it at
> <http://www.hum.ku.dk/ami/28-100r-dt.html>.

Which also reveals that I made a slight mistake and wrote <om>
instead of <um>, despite the fact that I was looking at the same
picture in a book. With modern orthography the text would look
like this: 'drømte mig en drøm i nat om silke og ærlig pæl'. The
melody was used by Radio Denmark for many years.

¹ Dansk i tusind år, 1993.
--
Torsten

Brian M. Scott

unread,
Sep 8, 2002, 10:59:09 PM9/8/02
to
On Sun, 8 Sep 2002 21:38:35 +0000 (UTC), Torsten
<t_usen...@hotmail.com> wrote:

>Brian M. Scott skrev:

>>>> drømde mik en drøm i nat om silki ok ærlik pæl.

>>> Translation please? "I dreamed a dream in [the] night about
>>> silk and ... ?"

>> ... fine fur. (And <i nat> is probably better translated 'last
>> night', though Old Norse <í nátt> can be either 'tonight' or
>> 'last night'.)

>Yes. Even in contemporary Danish <i nat> means both things. Allan
>Karker¹ translates <ærlik pæl> as 'prægtig brokade', i.e., fine
>or magnificent brocade. A possible translation could then be
>'dreamed me a dream last night about silk and fine brocade'.
>Otto Kalkar's 'Ordbog til det ældre danske sprog (1300-1700)'
>published in four volumes from 1881-1907 unfortunately does not
>resolve the problem of the meaning of <pæl>.

I'll bet that it's the same as Swedish <päll>, which is covered
by the SAOB (on-line at <http://g3.spraakdata.gu.se/saob/>); its
meaning seems to be 'costly fabric'. And now that I've had a
chance to look further, I see that Old Norse <pell> is 'some sort
of costly fabric'; the word, like Sw. <päll>, is from Latin
<pallium>. (But SAOB gives the Danish as <peld>.)

[...]

Brian

Torsten

unread,
Sep 9, 2002, 6:15:59 AM9/9/02
to
Brian M. Scott skrev:

> I'll bet that it's the same as Swedish <päll>, which is covered
> by the SAOB (on-line at <http://g3.spraakdata.gu.se/saob/>);
> its meaning seems to be 'costly fabric'. And now that I've had
> a chance to look further, I see that Old Norse <pell> is 'some
> sort of costly fabric'; the word, like Sw. <päll>, is from
> Latin <pallium>. (But SAOB gives the Danish as <peld>.)

'Ordbog over det danske Sprog' has <Peld> with the variant
spellings <Pel>, <Pelt>, <Pjeld>, and <Pjæld>. It lists the
following older forms: <pield>, <pield>, <pelt>, <peld>, <pell>,
<pæll>, and <pæl>. Like the SAOB it mentions Old Norse <pell> and
writes that it is from Old English <pæll>, from Latin <pallium>.
It means both some kind of costly fabric and a garment made of
the material. The form <Peld> can also mean a 'bear's fur' but
the citation given is from 1835. That is likely to be the source
of the (probably incorrect) translation on the Arnamagnaean
Institute's web site.

The various other meanings of Swedish <päll> are not shared by
<Peld>. The word is archaic and has not been in continuous use in
Danish. It was revived in the 19th century but has fallen out of
use again.

--
Torsten

Torsten

unread,
Sep 9, 2002, 6:33:00 AM9/9/02
to
Torsten wrote:
> Jim Heckman wrote:

>> is [...] <ř> just a different development of <oe> than


>> happened in German/Swedish <ö>?

> Probably not.

I was wrong. According to <http://www.dsn.dk/nfs/2000-3.htm>,
quoting Erik Kroman 'Dansk Palćografi' (in 'Nordisk Kultur
XXVIII'), <ř> is thought to have originated as a combination of
<o> and <e>. Besides the diagonal stroke, vertical or horizontal
ones were also in use in manuscripts.

--
Torsten

Brian M. Scott

unread,
Sep 9, 2002, 11:49:14 AM9/9/02
to
On 09 Sep 2002 Torsten <t_usen...@hotmail.com> wrote in
news:alhscv$27e0$1...@news.cybercity.dk in sci.lang:

> Brian M. Scott skrev:

>> I'll bet that it's the same as Swedish <päll>, which is
>> covered by the SAOB (on-line at
>> <http://g3.spraakdata.gu.se/saob/>); its meaning seems to
>> be 'costly fabric'. And now that I've had a chance to look
>> further, I see that Old Norse <pell> is 'some sort of
>> costly fabric'; the word, like Sw. <päll>, is from Latin
>> <pallium>. (But SAOB gives the Danish as <peld>.)

> 'Ordbog over det danske Sprog' has <Peld> with the variant
> spellings <Pel>, <Pelt>, <Pjeld>, and <Pjæld>. It lists the
> following older forms: <pield>, <pield>, <pelt>, <peld>,
> <pell>, <pæll>, and <pæl>. Like the SAOB it mentions Old
> Norse <pell> and writes that it is from Old English <pæll>,
> from Latin <pallium>. It means both some kind of costly
> fabric and a garment made of the material. The form <Peld>
> can also mean a 'bear's fur' but the citation given is from
> 1835. That is likely to be the source of the (probably
> incorrect) translation on the Arnamagnaean Institute's web
> site.

That makes sense, and the confusion is understandable. Thanks
for the information.

[...]

Brian

0 new messages