Thank you, Gordon, for your account of the history
and wanderings of the double-dotted o.
On second thought, I am not so sure I have actually
seen the German Fraktur "ö" in Renaissance printed
books. The books I have seen are all from the 17th and
18th centuries, and I cannot now swear that I have
seen Fraktur versions of ö, ü and ö there. That is why
I went to look on the internet, to see if any web pages
carry facsimile images of book-pages from the period.
Thus far I was only able to find a facsimile of the
front page of the Luther Bible of 1534. It carries
the text:

Biblia/ das ist/ die
gantze heilige Sch=
rifft Deudsch.
Mart. Luth.
Wittemberg.
Begnadet mit Kür=
furstlicher zu Sachsen
freiheit.
Gedruckt durch Hans Lufft.
M. D. XXXIIII.

(See the webpage at
http://www.evang.at/amstetten/hauptteil_kirchenboten.html)
I think, if you look very carefully, you will see
something that *might* be a double dot, above the 'u'
in 'Kur='(fursten). If you care to take a look, I
should like to have your opinion. But that would in any
case be in the wrong place, since the right form is Kurfürst.

With regard to Icelandic, if we are lucky, we will find
something at the Arnamagnæan websites either in Copenhagen
or in Reykjavík. But I do not have time right now.
The thing to look for would be the first appearance
of the grapheme ö in Icelandic manuscripts, if any one
would like to try.

With regard to the first umlauts being indicated by
means of juxtaposed miniscule e's, this does
make sense. In fact, if I were to draw a map of the
"positions" of the letters a, e, i, it would have
to look something like this:

........i...e...o...e...i.....
...............aaa............

(don't know how well this comes through)
The reason for two i's and e's is that I'd like
them to represent the corners of the mouth.
O is in the middle, because the mouth is the smallest for o.
Two e's are on each side of o, because in order to change
an o into an e, one needs to draw out the two corners
of the mouth towards each side. Outside the e's you
then see two i's. That is because in order to move from
e to i, one needs to draw the corners of the mouth still
further out.
The a is below the o, because one needs to drop the jaw
in order to move from o to a. But I have drawn 3 a's
below the a, in order to indicate a wider opening of the
mouth.

The umlauted o then fits into the figure somewhere
on the horisontal ..i...e...o...e...i.. axis. But it has
to be inside the e's, because otherwise all reference
to the original o would have disappeared.

Well, the diagram is not 100% accurate, because 'i'
also involves a narrowing of the space between the
tongue and the palate.

But I think the diagram illustrates quite well why
the ö of "röten" develops from the o of "rot" under
the influence of the subsequent e in German.
Gott and Götter is another example.
In Old Icelandic the parallel would be how the o in "koma"
develops into ø in "hann kømr" under the influence of the i.
Similarly with "rót" and "røtr", where a lost i or e
may be assumed.

But on the whole, the u-umlaut seems to be much more
ubiquitous in Old Icelandic or Norse and the i-umlaut
more rare, in contrast to German, where it is opposite.
I wonder why.

Thank you once again for your input of facts!

Xigung

--- In norse_course@yahoogroups.com, Gordon W <gorw@...> wrote:
> Hello Xigung,
>
> About the graphemes <ä> and <ö>.
>
> > Try then to simultaneously answer the two questions
> > 1) why was a "dot" used,
> > 2) why was the dot(s) placed above the letter?
> > The answer, as I have understood it, is that it was
> > simply a question of borrowing from another letter,
> > namely the "i", which is the origin of the dot placed
> > above the letter.
>
> The german (fraktur) letters <ä> and <ö> where initially written as
A and O
> with a small <e> ontop of them, to indicate that the pronounciation
should be
> closer to /e/.
>
> Before the introduction of <ä> and <ö> into swedish writing, we
used <æ>
> and <ø> which originally were ligatures of a+e and o+e.
>
> The swedish grapheme <å> (was designed the same way modelling the
early <ä>.
> So, indicating it should be pronounced closer to /o/. (In modern
Swedish, /å/
> is equivalent to [o:], and it has been for quite some time, since
original Old
> Swedish /a:/ has shifted to /o:/). It was used sporadically in the
15th
> century, but became very common after the book printer Jörgen
Richolff used
> it in all books printed at Kungliga tryckeriet from 1526 and
onwards,
> primarily the New Testament (August 1526).
>
> The reason for the change in Sweden was that the first swedish book
printers
> used german types, this occured in the 16th century. They where not
> incorporated into danish writing. (here is an example:
> http://www.danmark.dk/forside.asp)
>
> <ä> and <ö> with dots, instead of an small e, developed out of "den
> senmedeltida gotiska skrivstilens motsvarande typer", according to
Wessén,
> 1995. It could be difficult to see the difference between ä and o
in fraktur
> printing, while the ä was written with an e. The dots where more
clearer. The
> dots where introduced in antikva in 1730, and in fraktur about a
hundred years
> later, also according to Wessén.
>
> > Then some time later the "double-horned o" must have
> > arrived in Iceland as a Danish export and came in vogue
> > there.
>
> I believe I've read somewhere, that the adoption of <ö> into modern
icelandic
> writing was in part because of a wish to distance themselves from
denmark
> after the independence, to whom iceland for quite some time and
been a
> province. Maybe someone else on the list could clarify this. It is,
in any
> case, not a danish export.
>
> > Any way, as I have understood from perusing some
> > textbooks on modern Icelandic, the ö is now a
> > diphtong - can you confirm this? (pronounced in
> > a way that is best reflected by the combination
> > "ai" relative to Norwegian)
>
> Modern Icelandic <ö> is pronounced as Swedish /ö/. Icelandic <æ>, is
> however pronounced as /ai/.
>
> Kind regards,
> Gordon
> Stockholm, Sweden