A web search shows that, as I suspected, this professor is Turkish and
his name is Güntürkün. The ü's are Turkish letters, not German
u-umlauts, so it's interesting that they have been converted to "ue"
in some texts (on-line as well as in the Post article).
His e-mail address appears in most on-line resources as
onur.gue...@ruhr-uni-bochum.de
but in one place appears as
onur.güntürk...@ruhr-uni-bochum.de
--
Harlan Messinger
Remove the first dot from my e-mail address.
Veuillez ôter le premier point de mon adresse de courriel.
> An article in today's Washington Post reports on the results of
> research performed by Onur Guentuerkuen of the Ruhr University in
> Bochum, Germany.
>
> A web search shows that, as I suspected, this professor is Turkish and
> his name is Güntürkün. The ü's are Turkish letters, not German
> u-umlauts, so it's interesting that they have been converted to "ue"
> in some texts (on-line as well as in the Post article).
"ü" is also a German letter; it is special in that it has no place in the
ordered alphabet, and that there "ue" exists as a standard way to write it
if "ü" is not available. The sound is like the sound of the Turkish "ü".
Why should the "ue" not be used for it in a German context?
Joachim
A very useful convention in our days of typewriting and computing.
Those who aren't aware of it aren't likely to pronounce correctly
either of the two orthographies anyway, and whoever knows Turkish will
correct automatically.
Most Germans I've seen writing about this don't seem to consider 'ü' a
letter - they just consider it an 'u' with an umlaut, just as I, in
Italian, don't consider 'è' a letter, but an 'e' with a grave accent.
It's just a matter of conventions... to a Swede, 'ö', 'ä' and 'å'
*are* definitely letters on their own right.
by LjL
ljl...@tiscalinet.it
> An article in today's Washington Post reports on the results of
> research performed by Onur Guentuerkuen of the Ruhr University in
> Bochum, Germany.
>
[...]
>
> His e-mail address appears in most on-line resources as
>
> onur.gue...@ruhr-uni-bochum.de
>
> but in one place appears as
>
> onur.güntürk...@ruhr-uni-bochum.de
The umlauted letters are not valid in e-mail addresses, so the second
version must be wrong.
--
Stranded in Canada
[eng]
I am not sure whether non-ASCII characters really make e-mail addresses
invalid. I remember at least one e-mail address with a non-ASCII character
(SMALL LATIN LETTER E WITH ACUTE) that seemed to work.
[epo]
Mi ne scias certe chu neaskiajn signoj vere nevalidigas retmesaghajn
adresojn. Mi rememoras minimume unu retmesaghan adreson kun neaskia signo
(SMALL LATIN LETTER E WITH ACUTE) kiu evidente funkciis.
Gerard van Wilgen
gvanw...@planet.nl
www.majstro.com (Konciza multlingva tradukvortaro)
>
> "O.C." <ocr...@yahoo.com.jp> wrote in message
> news:oprks9cw...@news.uqam.ca...
>> On Mon, 17 Feb 2003 07:40:31 -0500, Harlan Messinger
>> <h.mes...@comcast.net> wrote:
>>
>>
>> The umlauted letters are not valid in e-mail addresses, so the second
>> version must be wrong.
>>
>
> [eng]
>
> I am not sure whether non-ASCII characters really make e-mail addresses
> invalid. I remember at least one e-mail address with a non-ASCII
> character
> (SMALL LATIN LETTER E WITH ACUTE) that seemed to work.
According to RFC821, no character out of ASCII can be used. So '%', '&' and
some other punctuation might be possible, but e acute or umlaut should not
be. That is to say, if they are used, they may work, but it cannot be
guaranteed. In actual practice, '&' doesn't work well.
http://www.faqs.org/rfcs/rfc821.html
--
Stranded in Canada
>> "ü" is also a German letter; it is special in that it has no
>> place in the ordered alphabet, and that there "ue" exists as a
>> standard way to write it if "ü" is not available. The sound is
>> like the sound of the Turkish "ü". Why should the "ue" not be
>> used for it in a German context?
Lorenzo> Most Germans I've seen writing about this don't seem to
Lorenzo> consider 'ü' a letter - they just consider it an 'u' with
Lorenzo> an umlaut,
But there ARE root words in uninflect forms which contain a the ü.
E.g. Gebühr, München, fühlen, etc. These ARE NOT inflected from
*Gebuhr, *Munchen, *fuhlen. So, the ü in these cases are letters, not
umlauts (= sound changes).
Lorenzo> just as I, in Italian, don't consider 'è' a
Lorenzo> letter, but an 'e' with a grave accent. It's just a
Lorenzo> matter of conventions...
In Italian, the grave accent mark is used to indicate stress when the
stress is not on the penultimate syllable. (Often, you only use it
when the stress falls on the last syllable.)
In German, there are sound change rules that convert a u to a ü, but
there ARE words that have a ü sound in the first place, not a change
from any u. Consider ü to be always a sound change is absurd.
Lorenzo> to a Swede, 'ö', 'ä' and 'å' *are* definitely letters on
Lorenzo> their own right.
Why?
--
Lee Sau Dan 李守敦(Big5) ~{@nJX6X~}(HZ)
E-mail: dan...@informatik.uni-freiburg.de
Home page: http://www.informatik.uni-freiburg.de/~danlee
Gerard> [eng]
Gerard> I am not sure whether non-ASCII characters really make
Gerard> e-mail addresses invalid.
They are invalid.
Gerard> I remember at least one e-mail
Gerard> address with a non-ASCII character (SMALL LATIN LETTER E
Gerard> WITH ACUTE) that seemed to work.
That's an invalid address. That it works was luck on you. It is NOT
GUARANTEED to work, as some intermediate relay mail-servers may fail
to handle it and scramble it, causing it to be undeliverable.
Gerard> [epo]
Gerard> Mi ne scias certe chu neaskiajn signoj vere nevalidigas
Gerard> retmesaghajn adresojn.
Mi certas, ke gxi estas nevalida.
Gerard> Mi rememoras minimume unu
Gerard> retmesaghan adreson kun neaskia signo (SMALL LATIN LETTER
Gerard> E WITH ACUTE) kiu evidente funkciis.
Nur kiam vi havus bonsxancon. Tiaj adresoj ne cxiam funkcias. Tio ja
dependas je, cxu cxiuj mesagx-serviloj inter vi kaj la ricevanto povus
la neaskiajn signojn manipuli korekte. Se ja, vi havus bonsxancon.
Se ne, la mesagxon perdus.
Are you asking why he makes that claim, or why they came to be
considered distinct letters? In the former case, there is no question:
they come at the end of the alphabet, after Z. Words beginning with
these letters are found at the end of the dictionary.
As to why they should be distinct letters, it's because no umlaut is
involved at all. There is no inflectional shift o > ö, etc. The three
symbols were created to represent three sounds distinct from those
represented by A, E, I, O, U, Y. They just happen to have been based
on the letters A and O rather than having been invented from scratch.
There are a few words in Swedish where the inflected form has <ä> instead
of <a>, an example the plural of <stad> 'town, city', <städer>. Some other
words undergo the change <å>:<ä> (as far as I know, the letter <å> was
earlier long /a/): <låg> : <lägre> 'low: lower'; or <o>:<ö>: <rot> :
<rötter> 'root; or, finally <u>:<y>, where no "diacritical" marks are
used: <ung> : <yngre> 'young: younger'.
Of course, these changes are less common than in German. The
pronunciations have diverged, too. <å> sounds like the German <o> and the
Swedish <o> usually is, in open syllables, pronounced like the German <u>.
In closed syllables, it generally resembles the German <o> in closed
syllables. In other words, the openness of a syllable has influenced the
pronunciation, as has happened in German and English.
Toni
--
# change .invalid -> .fi only for personal mail #
http://www.st.jyu.fi/~tpkeskit/
"Kuka nyt uskoo tähän kaikkeen? Kuka vielä uskoo käärmeeseen, joka
kahdella kielellä myrkkyä iskee unelmien eteiseen?" Ismo Alanko
How is it relevant to the discussion? The question is, can the script
represent accurately enough for a native whether a back or front vowel
is required.
For more or less the same reason that an Albanian "ë" would not be
rendered in a German text as "ee".
This is all quite correct. Still one may ask why Germans don't consider umlaut
letters as letters in their own right. I think, it must be that alphabet is a
fairly traditional thing and Germans are used to make umlaut letters equal with
A, O, U, even when this is morphologically/lexically not the case - it's their
tradition. In Swedish umlauting is much rarer so Swedes have decided to give the
umlaut letters the rights of separate letters - and a place in the alphabet
correspondingly. Finns have borrowed the Swedish tradition, obviously.
Mihkel
ü > ue is a common and standard change in German, when umlauts are not available.
Albanian and Turkish should not stick their toes into German, if they want to avoid the
letter change. But since they've failed to stay away, German script will have to learn to
handle their peculiarities. Through painful experience for all of them, of course.
Mihkel
>>>>>> "Lorenzo" == Lorenzo J Lucchini <ljl...@tiscalinet.it> writes:
>
> >> "ü" is also a German letter; it is special in that it has no
> >> place in the ordered alphabet, and that there "ue" exists as a
> >> standard way to write it if "ü" is not available. The sound is
> >> like the sound of the Turkish "ü". Why should the "ue" not be
> >> used for it in a German context?
>
> Lorenzo> Most Germans I've seen writing about this don't seem to
> Lorenzo> consider 'ü' a letter - they just consider it an 'u' with
> Lorenzo> an umlaut,
>
>But there ARE root words in uninflect forms which contain a the ü.
>E.g. Gebühr, München, fühlen, etc. These ARE NOT inflected from
>*Gebuhr, *Munchen, *fuhlen. So, the ü in these cases are letters, not
>umlauts (= sound changes).
I don't think "letter" is defined in such terms.
Really, I think the only acceptable definition of "letter" is
circular, i.e. a letter is what people speaking language X consider a
letter.
>
> Lorenzo> just as I, in Italian, don't consider 'è' a
> Lorenzo> letter, but an 'e' with a grave accent. It's just a
> Lorenzo> matter of conventions...
>
>In Italian, the grave accent mark is used to indicate stress when the
>stress is not on the penultimate syllable. (Often, you only use it
>when the stress falls on the last syllable.)
Yeah but there is also the acute. 'è' and 'é' are only used on
(certain) stressed syllables, but nevertheless, they indicate two
different sounds when they're there.
Anyway, accent marks are used to indicate stress when the stress is on
the last syllable - it's not that we "often only use it" in that case.
However, I think most language authorities agree that an accent can be
optionally used on syllables other than the last of some words, in
case of a clash with another word that would be a homographe with no
accents.
Besides this (i.e. disregarding the fact that accents are only used on
the last vowel of a word), since "stressing a syllable" in Italian
simply implies making its vowel long (as far as I know), you could see
accented vowels as long vowels. Now, Greek used to have two separate
*letters* to indicate a pair of vowels that differed in length from
another pair (namely, hta and wmega vs epsilon and omikron).
>In German, there are sound change rules that convert a u to a ü, but
>there ARE words that have a ü sound in the first place, not a change
>from any u. Consider ü to be always a sound change is absurd.
Absurd or not, it has nothing to do with it being a letter or not.
Following your reasoning, one might argue that "a" and "u" are not
letters but just sound changes in English, because of things like
"drink" / "drank" / "drunk", and someone else may counter that they
are indeed letters because there ARE words that have their sounds in
the first place.
That would be a nonsense discussion, you see.
>
> Lorenzo> to a Swede, 'ö', 'ä' and 'å' *are* definitely letters on
> Lorenzo> their own right.
>
>Why?
Because they come after Z in the alphabet.
"CH" and "LL" are letters for a Spaniard. Would you have any real
arguments to convince them that they aren't?
by LjL
ljl...@tiscalinet.it
> [snip]
Drifting from the topic a little - I've been taught at school that
there are 21 letters in the alphabet, namely
A B C D E F G H I L M N O P Q R S T U V Z
This would render hundred years old "imports" like "xilofono" and
"jena" not Italian. Go figure.
One effect of this is that I still cannot write a decent K or X (the
others are easier) in cursive. And another effect is that a lot of
people confuse the names of "ipsilon"/"i greca" (Y) and "i
lunga"/"jay" (J) - when I give my email address to someone, I can
spend minutes trying to tell them which "jay" I mean. Again, go
figure.
by LjL
ljl...@tiscalinet.it
I've always considered the German umlauts as letters of their own. If they
weren't letters of their own, why would they replaced with "ae", "oe" and
"ue" when the real umlauts are not available? I don't know of any other
languages where one tries to substitute diacritics in a similar way.
Moreover, in telephone books the umlauts are also arranged like "ae", "oe"
and "ue". And most strikingly, in German the umlauts have their own names,
so ü (u umlaut) is called [y:], for instance, never "U-Umlaut" or anything
like that. Even in the phonetic alphabet, it's "Ü wie Übermut" not anything
with "u". Does the fact that the umlauts usually don't have a special
position in the German alphabet alone make them less of a letter?
According to my Langenscheidt dictionary this changed in 1994, i.e. "ch" and
"ll" are not letters if their own anymore. Nevertheless, I think your point
is completely valid.
In a similar vein, Spanish doesn't consider accented letters to be separate, except
for "ñ", which *is* separate from "n". Even stranger, "ch", "ll", and "rr" are also
separate letters, meaning that, in a dictionary, words that start with "ch" come
*after* words that start with "cu". Seemingly, no language can be without a few
quirks to call its own. :)
Charles
Yes, I thought I remembered having read that the Spanish Academy got rid of
these as distinct letters.
On the contrary, the umlauted letters are nothing more than shortcuts for
writing the digraphs. When I write in English a word with two adjacent "t"s,
I cross them with one stroke. Does that make the glyph consisting of two
vertical strokes and one horizontal stroke a separate letter? Printers have
traditionally used special characters (ligatures) for digraphs and trigraphs
like "fl"and "ft". Are these distinct letters?
> I don't know of any other
> languages where one tries to substitute diacritics in a similar way.
Will Swedes use "aa" if "å" (a-ring) isn't available? Esperantists
conventionally use "x" after a consonant to indicate the haceked version of
that consonant.
> Moreover, in telephone books the umlauts are also arranged like "ae", "oe"
> and "ue".
It would seem to me that that's so people, knowing how someone's name is
pronounced, don't have to guess whether in his particular name the umlauted
sounds are spelled the old way or the new way.
>And most strikingly, in German the umlauts have their own names,
> so ü (u umlaut) is called [y:], for instance, never "U-Umlaut" or anything
> like that. Even in the phonetic alphabet, it's "Ü wie Übermut" not
anything
> with "u". Does the fact that the umlauts usually don't have a special
> position in the German alphabet alone make them less of a letter?
Yes.
Painful? Absurd. As if a native speaker couldn't reconstitute the
sound as easily as any German.
Yes. And in Danish "aa" used to be the way to write what is now "å"
(there was a spelling reform in 1948.) In Danish dictionaries, the town of
Aalborg, which retains the old spelling, is alphabetized at the end, under
"å".
Pierre
--
Pierre Jelenc | H o m e O f f i c e R e c o r d s
| * Marwood * The Cucumbers *
T h e G i g o m e t e r | * Switchblade Kittens * Pawnshop *
www.thegigometer.com | www.homeofficerecords.com
> Will Swedes use "aa" if "å" (a-ring) isn't available?
The ring is simply dropped so "a" is used for "å". The letters
"ä" and "ö" are treated likewise. On the other hand, using the
digraph "aa" for "å" is quite unproblematic in Danish (after
all, it was the norm until 1948 and it's still in common use in
certain proper names). The normal substitutes for "æ" and "ø"
are "ae" and "oe", but you would only use them when forced to by
circumstances like foreign keyboards, inadequate email systems
and so on.
--
Torsten
> Yes. And in Danish "aa" used to be the way to write what is
> now "å" (there was a spelling reform in 1948.) In Danish
> dictionaries, the town of Aalborg, which retains the old
> spelling, is alphabetized at the end, under "å".
Correct. Aalborg is not the only town using "aa". During the
past twenty years or so, the use of the digraph has been on
the increase in place names. Its use by public authorities was
approved by a circular letter from the Ministry of Education
dated March 15, 1984. It is never incorrect to use "å", however,
so "Ålborg" is also a valid spelling. The alphabetizing of "aa"
under "å" is even done with words where it isn't a digraph for
the same sound(s) as "å", meaning that a foreign place name like
"Aachen" will normally be listed under "Å". Letters that are not
part of the Danish alphabet, but are used in, for example, family
names, get similar treatment. For instance, "ü" is alphabetized
as "y", "ä" as "æ", "ö" as "ø", "ð" as "d", and "þ" as "th". In
all cases, the foreign letters come after the equivalent Danish
ones¹. The French "oe" ligature is alphabetized as if the letters
were used separately. Other diacritics are simply ignored when
alphabetizing.
¹ The name "Schytt" will be listed before the name "Schütt". This
common-letter-first rule also applies to the "å"/"aa" issue.
"Ålborg" comes before "Aalborg" (and "Aalborg" comes before
"Ålbæk").
--
Torsten
What about Norwegian?
John.
I was thinking in terms of writing of course. Writing is painful, when
you need to produce an umlaut letter, but it's not available on the
keyboard. The "umlaut" e of Albanian written as "ee" looks certainly
painful to an Albanian, who loves his literary language.
Mihkel
Of course all logical reasoning is welcome, but the alphabet shows the
traditional way of thinking. Traditionally Germans know very well that
umlaut letters were developed from ae, ue and oe digraphs, so
traditionally they are considered as not yet quite independent, though
in some sense they certainly are independent. Swedish tradition shows
that considering the umlauts as separate letters is feasible and practical.
Mihkel
>
Umlaut letters are only shortcuts for ae, oe and ue, if you know that
they are originally ae, oe and ue. The case is that they are not quite
ae, oe and ue nowadays and that's taken into good consideration in
Swedish tradition, which has arranged its alphabet so as to include
umlauts as letters in their own right. Let's agree that the German way
to settle the matter is good for Germans and the Swedish way is good for
Swedes.
Mihkel
First i have to admit that this thread is very intriguing. Being
a native German speaker, I have to deal with umlauts every day,
and I never even bothered to think about whether they constitute
separate letters or not.
When I first saw the question "are umlauts letters or not" - my
gut reaction was "of course they are letters - they look
different and they sound different, too".
Then I checked the Duden, and to my surprise I found the words
with umlauts not sorted according to the "umlaut u" = "ue"
convention, no, they were sorted just as the letter "u" alone.
Now this seemed to me even more puzzling...
I checked some sources on the origin of the typographical signs,
and I stumbled over the explanation, that the umlaut originally
were indeed combinations of a+e, o+e and u+e, and the dots
evolved from the convention of the "e" being written over its
predecessing vowel.
Now I wonder where the *sounds* of the umlauts come from, as in
past times, the "e" following a vowel usually denoted
lengthening, as it is still seen in names, such as the town
Straelen (not written with umlaut!), which is in fact pronounced
"Strahlen"...
Thanks a lot for a lot of food for thought :-))
Clear Ether!
Stayka
A fairly scientific article on umlaut sounds:
http://www.cogs.susx.ac.uk/lab/nlp/polylex/polynode53.html
Mihkel
Thx :-))
Bb, Stayka
> The alphabetizing of "aa"
> under "å" is even done with words where it isn't a digraph for
> the same sound(s) as "å", meaning that a foreign place name like
> "Aachen" will normally be listed under "Å".
... with the exception of compound words, in which each a is on its
own side of the compound border, such as "ekstraarbejde". These words
are alphabetized according to "aa" rather than "å".
> ... with the exception of compound words, in which each a is on its
> own side of the compound border, such as "ekstraarbejde". These words
> are alphabetized according to "aa" rather than "å".
True. Thanks.
--
Torsten
Does anyone really write double e for ë? I was under the impression
that in e-mail addresses and foreign keyboards they just omit the
diacritics.
The same is true for Finnish names, that are still often written with
ae for <ä> and oe for <ö>, especially in skiing competitions held in
German-speaking area. The most painful ones were Maeaettae for
<Määttä> and Haemaelaeinen for <Hämäläinen>...
Toni
--
# Replace .invalid with .fi for personal mail only #
"Jos aivastan, koko seinä kaikuu." (Tommi Liimatta)
> The same is true for Finnish names, that are still often written with
> ae for <ä> and oe for <ö>, especially in skiing competitions held in
> German-speaking area. The most painful ones were Maeaettae for
> <Määttä> and Haemaelaeinen for <Hämäläinen>...
Google in de/fi/sv also treats "ö" and "oe" alike.
<http://google.com/search?hl=fi&ie=ISO-8859-1&q=%E5ngstr%F6m>
<http://google.com/search?hl=fi&ie=ISO-8859-1&q=aangstroem>
have the same number of hits.
In English, they are not considered equivalent:
<http://google.com/search?hl=en&ie=ISO-8859-1&q=%E5ngstr%F6m>
<http://google.com/search?hl=en&ie=ISO-8859-1&q=aangstroem>
--
Top posting.
What's the most irritating thing on Usenet?
>Harlan Messinger wrote:
>
>> Will Swedes use "aa" if "å" (a-ring) isn't available?
>
>The ring is simply dropped so "a" is used for "å". The letters
>"ä" and "ö" are treated likewise.
I've seen other arrangements, like aa/ae/oe and ao/ae/oe, besides the
{ | } that hasn't yet disappeared completely from the Internet.
> [snip]
by LjL
ljl...@tiscalinet.it
Ok I stand corrected, my dictionary must be older.
(I personally think it was a good thing to get rid of them in any case
- whether they're philosophically letters or not, the fact remains
that they're letters for a computer, and coincidence has it that
respectively a C followed by an H and a L followed by another L look
the exact same as those two ex-letters. Ergo, it's a mess)
by LjL
ljl...@tiscalinet.it
> I've seen other arrangements, like aa/ae/oe and ao/ae/oe,
Or <o">, <a">, etc.
> besides the { | } that hasn't yet disappeared completely from
> the Internet.
Sometimes those texts were supposed to be in ISO 646-{DK,NO,SE}.
In these encodings, they are not a work-around, but just an
artifact of viewing the text as if it was using US-ASCII. Before
the advent of MIME, there was no reliable way to automatically
determine at the receiving end what those codes represented.
What is really funny is that you nowadays sometimes see [\]{|}
in postings that claim to be using ISO 8859-1. Especially in the
swnet.* and se.* hierarchies. Old habits die hard.
--
Torsten
I see; I only knew those characters were used instead of ö ä å, didn't
know the reason had to do with an older encoding standard.
I'm lucky that, with Italian, one only has to change accents into
apostrophes, and that's really painless since accents only occour at
the end of words.
Now to exchange some tidbit, I'm using a Swedish keymap to write
almost everything including Italian. The Italian keymap has ò, à and è
where you have respectively ö, ä and å; the key for è also does é
using Shift.
This should give a clue: no, if you do Shift+ò or Shift+à, you don't
get the uppercase versions - you get c-cedilla and the degree symbol.
Fine if it weren't for the fact that every decent standard says that
uppercase text *should* be accented in Italian... a very common
occurrance is that of the verb "è" (is), which is often found at the
beginning of a sentence, after a full stop.
Thus... the Italian keymap does not allow writing correct Italian.
Duh. The Swedish keyboard, on the other hand, has a dead key for ` and
´: you can write better Italian with a Swedish keyboard than an
Italian one.
Not to mention the fact that the Swedish keymap also has dead keys for
¨, ^, and ~, which the Italian lacks (and the first two are arguably
Italian diacritics, even though they aren't used in nowaday's texts).
by LjL
ljl...@tiscalinet.it
>
> The same is true for Finnish names, that are still often written with
> ae for <ä> and oe for <ö>, especially in skiing competitions held in
> German-speaking area. The most painful ones were Maeaettae for
> <Määttä> and Haemaelaeinen for <Hämäläinen>...
I am not sure but i think that the German rule that requires replacing
Umlauts by vowel-plus-e is not restricted to German words. It is generally
applied to names, even foreign ones.
Joachim
> On the contrary, the umlauted letters are nothing more than shortcuts for
> writing the digraphs.
Historically, yes. But currently, no.
Joachim
How do you figure "currently no" when, even today, as soon as umlauts
aren't available, the spelling shifts right back to original?
--
Harlan Messinger
Remove the first dot from my e-mail address.
Veuillez ôter le premier point de mon adresse de courriel.
> Joachim Pense <joachi...@t-online.de> wrote:
>
>>Harlan Messinger wrote <b33cbp$1fg8bc$1...@ID-114100.news.dfncis.de>:
>>
>>
>>> On the contrary, the umlauted letters are nothing more than shortcuts
>>> for writing the digraphs.
>>
>>Historically, yes. But currently, no.
>
> How do you figure "currently no" when, even today, as soon as umlauts
> aren't available, the spelling shifts right back to original?
>
Because it is considered an error not to use the umlaut-letters if they
*are* available (in particular in handwriting, of course.) If the umlauts
are not available, these are the "official" replacements, and it is not
acceptable to just omit the umlauts (like accents), because you would
replace one letter by a different one then.
Joachim
>Joachim Pense <joachi...@t-online.de> wrote:
>>Harlan Messinger wrote <b33cbp$1fg8bc$1...@ID-114100.news.dfncis.de>:
>>> On the contrary, the umlauted letters are nothing more than shortcuts for
>>> writing the digraphs.
>>Historically, yes. But currently, no.
>How do you figure "currently no" when, even today, as soon as umlauts
>aren't available, the spelling shifts right back to original?
Presumably because the digraphs are considered a poor man's
substitute for the real thing. In any case, they don't shift
back to the original, as in most cases there is no original.
Early practice varies enormously, from failing to indicate the
mutations at all, to using assorted superscripts, to using
various digraphs, and not necessarily those used today when
umlauts aren't available.
Brian
In Swiss practice, it's not necessarily an error except for
handwriting; when using capitals digraphs are generally preferred in
many areas (and acceptable in handwriting).
For capitals, digraphs were standard in the 19th century, before the 1903
Orthography "reform", where this usage was dismissed.
Joachim
>> But there ARE root words in uninflect forms which contain a
>> the ü. E.g. Gebühr, München, fühlen, etc. These ARE NOT
>> inflected from *Gebuhr, *Munchen, *fuhlen. So, the ü in these
>> cases are letters, not umlauts (= sound changes).
Lorenzo> I don't think "letter" is defined in such terms. Really,
Lorenzo> I think the only acceptable definition of "letter" is
Lorenzo> circular, i.e. a letter is what people speaking language
Lorenzo> X consider a letter.
Replace "speaking" with "writing".
>> In German, there are sound change rules that convert a u to a
>> ü, but there ARE words that have a ü sound in the first place,
>> not a change from any u. Consider ü to be always a sound
>> change is absurd.
Lorenzo> Absurd or not, it has nothing to do with it being a
Lorenzo> letter or not. Following your reasoning, one might argue
Lorenzo> that "a" and "u" are not letters but just sound changes
Lorenzo> in English, because of things like "drink" / "drank" /
Lorenzo> "drunk", and someone else may counter that they are
Lorenzo> indeed letters because there ARE words that have their
Lorenzo> sounds in the first place. That would be a nonsense
Lorenzo> discussion, you see.
Yeah. English has just one true vowel: the schwa! :D
Lorenzo> Because they come after Z in the alphabet. "CH" and "LL"
Lorenzo> are letters for a Spaniard. Would you have any real
Lorenzo> arguments to convince them that they aren't?
I think one thing we argue around here is the word "umlaut". I
interpret it as "sound change sign", whereas you interpret it simply
as just another "diacritical mark".
--
Lee Sau Dan 李守敦(Big5) ~{@nJX6X~}(HZ)
E-mail: dan...@informatik.uni-freiburg.de
Home page: http://www.informatik.uni-freiburg.de/~danlee
>> According to my Langenscheidt dictionary this changed in 1994,
>> i.e. "ch" and "ll" are not letters if their own
>> anymore. Nevertheless, I think your point is completely valid.
Lorenzo> Ok I stand corrected, my dictionary must be older.
Pay attention to the date of the reform. It's less than 10 years old.
Don't expect everyone and every dictionary to be updated that quickly.
I also have a Spanish-English dictionary that lists "ch" after "ci",
"ll" after "lo", etc. Well... I mean, they're in separate 'chapters'.
Lorenzo> (I personally think it was a good thing to get rid of
Lorenzo> them in any case - whether they're philosophically
Lorenzo> letters or not, the fact remains that they're letters for
Lorenzo> a computer, and coincidence has it that respectively a C
Lorenzo> followed by an H and a L followed by another L look the
Lorenzo> exact same as those two ex-letters. Ergo, it's a mess)
Internationalization software libraries have been around and
standardized for around 2 decades. They can handle these "simple" and
common situations (collating sequence) well. (Isn't sorting words in
an Arabic dictionary even harder?)
Again, that's a philosophical question: Should we change our habits to
fit _inferior_ tools, or improve our tools to suit our habits?
Harlan> On the contrary, the umlauted letters are nothing more
Harlan> than shortcuts for writing the digraphs. When I write in
Harlan> English a word with two adjacent "t"s, I cross them with
Harlan> one stroke. Does that make the glyph consisting of two
Harlan> vertical strokes and one horizontal stroke a separate
Harlan> letter? Printers have traditionally used special
Harlan> characters (ligatures) for digraphs and trigraphs like
Harlan> "fl"and "ft". Are these distinct letters?
Since German was in the story, I would like to raise out the "sz"
ligature, which eventually evolved into the nowadays sharp s (ß). Is
ß just a ligature, or a letter on its own? (If it is the latter, then
I can't see why ä, ö and ü shouldn't be letters on their own. The ß
lacks some features of a proper letter, whereas the umlauted letters
don't.)
Harlan> Will Swedes use "aa" if "å" (a-ring) isn't available?
Harlan> Esperantists conventionally use "x" after a consonant to
Harlan> indicate the haceked version of that consonant.
There are many conventions for Esperanto. Zamenhof recommended using
a post-"h", e.g. "ch" for "ĉ". Some people prefer to use a pre-"^"
mark. The post-"x" convention is established quite lately, perhaps
not before we have digital computers. It was invented so that sorting
routines for pure ASCII do not need rewriting just to sort Esperanto
word lists.
BTW, Esperanto uses a "breve" on the "u", i.e. ŭ, not a haĉek.
>> Moreover, in telephone books the umlauts are also arranged like
>> "ae", "oe" and "ue".
Harlan> It would seem to me that that's so people, knowing how
Harlan> someone's name is pronounced, don't have to guess whether
Harlan> in his particular name the umlauted sounds are spelled the
Harlan> old way or the new way.
How, how about Spanish's "ch" and "ll". Let's forget the new reform
for the moment. Are these diagraphs? Ligatures? Single letters?
What do they do if the ~ accent or ñ are not available?
Stayka> I checked some sources on the origin of the typographical
Stayka> signs, and I stumbled over the explanation, that the
Stayka> umlaut originally were indeed combinations of a+e, o+e and
Stayka> u+e, and the dots evolved from the convention of the "e"
Stayka> being written over its predecessing vowel.
You can sometimes see this writing style in some monuments, esp. when
it's in the Gothic typeface! :)
Stayka> Now I wonder where the *sounds* of the umlauts come from,
Stayka> as in past times,
What I find more interesting is why "ei" is /aj/ and "eu" is /oy/. Do
you know the answer? Hint: pay attention to certain more
'conservative' dialects.
Stayka> the "e" following a vowel usually denoted lengthening,
Not the "h"?
Stayka> as it is still seen in names, such as the town Straelen
Stayka> (not written with umlaut!), which is in fact pronounced
Stayka> "Strahlen"...
How should I pronounce "Goethe", then?
Harlan> Yes, I thought I remembered having read that the Spanish
Harlan> Academy got rid of these as distinct letters.
Has the ñ remained a letter?
>>>>>> "Lorenzo" == Lorenzo J Lucchini <ljl...@tiscalinet.it> writes:
>
> [snip]
>
> Lorenzo> (I personally think it was a good thing to get rid of
> Lorenzo> them in any case - whether they're philosophically
> Lorenzo> letters or not, the fact remains that they're letters for
> Lorenzo> a computer, and coincidence has it that respectively a C
> Lorenzo> followed by an H and a L followed by another L look the
> Lorenzo> exact same as those two ex-letters. Ergo, it's a mess)
>
>Internationalization software libraries have been around and
>standardized for around 2 decades. They can handle these "simple" and
>common situations (collating sequence) well. (Isn't sorting words in
>an Arabic dictionary even harder?)
Yes, I guess so.
But on the other hand, it's my impression that it was quite trivial
and painless for the Spanish to get rid of CH and LL as separate
letters, since they don't look any different from the letter
combinations C+H and L+L.
There might be some mess during the transition phase, but I guess
there was already some mess before the reform, mainly because of
computers and foreign languages.
>Again, that's a philosophical question: Should we change our habits to
>fit _inferior_ tools, or improve our tools to suit our habits?
I think mostly the latter, but when it's reasonable to do, the former
may give advantages, too.
Take a very simple example: before starting to use a computer (well,
there really wasn't a "before" for me practically, but anyway), I used
to write with a pen.
Now, do I want my computer to use my handwriting as input instead of a
keyboard? Damn NO!!! (except in those cases where a keyboard is too
big, like palmtops). Why should I teach my computer to "learn my
habit" of writing with a pen, when I can write much faster with this
keyboard-thing that was invented to "change my habits to fit inferior
tools"?
Note also that the "old" Spanish alphabet wasn't particularly
consistent, all tools aside: CH as a letter looked exactly like letter
C + letter H.
You can be sure this kind of things *will* cause confusion in some
situation, sooner or later.
by LjL
ljl...@tiscalinet.it
> [snip]
>
>Since German was in the story, I would like to raise out the "sz"
>ligature, which eventually evolved into the nowadays sharp s (ß). Is
>ß just a ligature, or a letter on its own? (If it is the latter, then
>I can't see why ä, ö and ü shouldn't be letters on their own. The ß
>lacks some features of a proper letter, whereas the umlauted letters
>don't.)
I don't think it's considered a letter, at least in a dictionary
listing.
> [snip]
>
>How, how about Spanish's "ch" and "ll". Let's forget the new reform
>for the moment. Are these diagraphs? Ligatures? Single letters?
>What do they do if the ~ accent or ñ are not available?
Sometimes they write N, sometimes they write ~ or something else.
And most of the time, if they are on Usenet, they is someone who
bitches about their decision :-)
by LjL
ljl...@tiscalinet.it
>>>>>> "Lorenzo" == Lorenzo J Lucchini <ljl...@tiscalinet.it> writes:
>
> [snip]
>
> Lorenzo> Because they come after Z in the alphabet. "CH" and "LL"
> Lorenzo> are letters for a Spaniard. Would you have any real
> Lorenzo> arguments to convince them that they aren't?
>
>I think one thing we argue around here is the word "umlaut". I
>interpret it as "sound change sign", whereas you interpret it simply
>as just another "diacritical mark".
Well it's what it is.
It might change the sound, but so do other marks in other languages.
In addition, you don't necessarily need an umlaut to "change the
sound" of a vowel.
What is it? It's two dots you put over a vowel. That's it.
Then a language may consider it more practical to see it as a mark,
another may feel it's better not to give it any special status, and
instead make vowels-with-the-dots separate letters.
Who gives a darn, as long as there is a reasonable standard in the
scope of a language.
by LjL
ljl...@tiscalinet.it
>>>>>> "Harlan" == Harlan Messinger <h.mes...@comcast.net> writes:
>
> Harlan> Yes, I thought I remembered having read that the Spanish
> Harlan> Academy got rid of these as distinct letters.
>
>Has the ñ remained a letter?
I don't know, but I guess people would complain about this being
"downgraded" to an N+diacritic much more than about LL and CH.
by LjL
ljl...@tiscalinet.it
> On 24 Feb 2003 17:54:21 +0100, Lee Sau Dan
> <dan...@informatik.uni-freiburg.de> wrote:
>
>> [snip]
>>
>>Since German was in the story, I would like to raise out the "sz"
>>ligature, which eventually evolved into the nowadays sharp s (ß). Is
>>ß just a ligature, or a letter on its own? (If it is the latter, then
>>I can't see why ä, ö and ü shouldn't be letters on their own. The ß
>>lacks some features of a proper letter, whereas the umlauted letters
>>don't.)
>
> I don't think it's considered a letter, at least in a dictionary
> listing.
It is considered a letter, though one missing some properties other letters
have (alphabet position, capital form). In the orthography standardisation
papers from around 1900 it was even considered a temporary situation that ß
had no capital letter; they installed a rule when to use SS and SZ in
all-caps-writing and expressed their hopes that this it soon would be made
obsolete by a capital ß.
Anyway, it is certainly not considered a ligature anymore.
(It *is* considered a pain in the neck by many, though...)
Joachim
> Since German was in the story, I would like to raise out the "sz"
> ligature, which eventually evolved into the nowadays sharp s (ß).
Isn't <ß> a ligature between <long s> and <s>, rather than <s> and
<z> (cf. <http://sivanataraja.free.fr/temp/ss.gif>) ?
It was s (in fraktur font) and z (in the same font). If written
closely together it looked very similar to aGreek beta, which is
why this sign later replaced the simple ligature.
Clear Ether!
Stayka
> It was s (in fraktur font) and z (in the same font). If written
> closely together it looked very similar to aGreek beta, which is
> why this sign later replaced the simple ligature.
Thanks.
That a few never used (and it hasn't killed them).
You can not state something general about it. (They are certainly not
ligatures, because that would imply joining in appearance, the Dutch
'ij' could come in all three categories.)
But let's take the Dutch 'ij' as an example, because that one is more
problematical than the others. There are regular heated discussions
in the Dutch newsgroups on whether that is a single letter or a digraph.
You might wish to have a single way to sort it, but there is none. In
Dutch reference works you will find three ways were it is placed.
In dictionaries you will find it between 'ii and 'ik', as if it was a
digraph. In some encyclopedias you will find it between 'x' and 'y',
as if it was a separate letter. In other sources you will find it
as if it was 'y', so a substitute for 'y' (or the reverse, as it
actually has been in the course of time). When capitalised, both
parts should be capitalised (encouraging the idea of a single letter).
So we have the town 'IJmuiden', not 'Ijmuiden'. And these two capital
letter again can be joined in a ligature. So what is it? It is all.
It is a digraph, a ligature and also a single letter. There is
*linguistically* no difference.
Similarly, in German some reference works sort the umlauted letters
as if the umlaut was not present (Duden?), others as if it is
equivalent to the base letter followed by an 'e' (phonebook).
The first suggests just an accented letter, the latter suggests a
digraph.
Now in Swedish some of the "umlauted" letters have a special place at
the end of the alphabeth. But not all. The remainder is sorted as
if there is no umlaut. But in Swedish, 'w' is of course just an
accented version of 'v', so sorted as if it actually is a 'v'.
And in older German texts you may find that the uppercase 'I' is
replaced by a 'J'. So apparently they did not feel a distinction.
Whether something is a single letter or not is more a cultural question
than a linguistic question. (Consider my frustration when my book on
Maltese told me that 'g' followed by 'h-crossbar' was a single letter
sorted between 'p' and 'q'. I have one other book on Maltese; a
dictionary. In that the combination was sorted as if it was a digraph.)
--
dik t. winter, cwi, kruislaan 413, 1098 sj amsterdam, nederland, +31205924131
home: bovenover 215, 1025 jn amsterdam, nederland; http://www.cwi.nl/~dik/
> [About Dutch <ij>] When capitalised, both parts should be
> capitalised (encouraging the idea of a single letter). So we
> have the town 'IJmuiden', not 'Ijmuiden'.
Contrast this with the Danish <aa>. It is alphabetized as <å>,
but capitalized as <Aa> (except of course when writing in
all-caps). Now, despite being capitalized as if it was two
separate letters, it has its own name; the same one as <å>.
When a Danish speaker wants to distinguish between the two in
speech, it is usually the <å>, not the <aa>, that gets special
treatment by using the name "bolle-å" ('ringed å'). Before the
official introduction of <å>, <aa> was alphabetized before <a>.
That is, effectively as <a>. Nevertheless, it was also considered
a letter in its own right with its own name back then. When the
<å> was introduced in 1948, it simply took the place of the
<aa> in most dictionaries and was therefore placed first in the
alphabet. A few years later, it was moved to the end, taking the
tradititional <aa> with it.
> Now in Swedish some of the "umlauted" letters have a special
> place at the end of the alphabeth. But not all. The remainder
> is sorted as if there is no umlaut. But in Swedish, 'w' is of
> course just an accented version of 'v', so sorted as if it
> actually is a 'v'.
This was also the traditional way of treating <w> in Danish.
The 1955 official normative orthographic dictionary used a 28
letter alphabet without <w>. The 1986 edition expanded this to 29
letters with <w> following <v>. This doesn't mean that <w> didn't
exist in Danish prior to 1986, only that it was alphabetized as
<v> and that the dictionary didn't inlude it in its definition of
the Danish alphabet. The Danish alphabet song ends with "... S T
U V X Y Z Æ Ø Å, 28 skal der stå". Must sound a bit odd in the
ears of today's children.
--
Torsten
>Dik T. Winter wrote:
>
> [snip]
>
>The Danish alphabet song ends with "... S T
>U V X Y Z Æ Ø Å, 28 skal der stå". Must sound a bit odd in the
>ears of today's children.
Makes me wonder if they happen to be teaching the "full" alphabet now
in Italy. I should ask someone who's at elementary... after all,
dictionaries make no mention of the fact that the alphabet is made of
21 letters, they list all 26 in the "canonical" ordering, and always
have, as far as I know.
I think it would make a lot of sense to get rid of that weird teaching
habit... even *not* counting foreign words, the Italian alphabet has
*more* than 21 letters, and it makes no sense not to count them
anyway.
And I wouldn't have to repeat the alphabet in *English* in my mind to
remember where J, K, W, X and Y are, if I'd been taught the whole
thing.
by LjL
ljl...@tiscalinet.it
: Internationalization software libraries have been around and
: standardized for around 2 decades. They can handle these "simple" and
: common situations (collating sequence) well. (Isn't sorting words in
: an Arabic dictionary even harder?)
if arranged by "roots" yes. otherwise rather straightforward. purely
alphabetical dictionaries exist and it is the norm for languages other
than arabic using the arabic script.
Yusuf> Lee Sau Dan <dan...@informatik.uni-freiburg.de> wrote: :
Yusuf> Internationalization software libraries have been around
Yusuf> and : standardized for around 2 decades. They can handle
Yusuf> these "simple" and : common situations (collating sequence)
Yusuf> well. (Isn't sorting words in : an Arabic dictionary even
Yusuf> harder?)
Yusuf> if arranged by "roots" yes. otherwise rather
Yusuf> straightforward. purely alphabetical dictionaries exist and
Yusuf> it is the norm for languages other than arabic using the
Yusuf> arabic script.
How do you handle the diacritical marks when you sort the Arabic
words? Is that trivial with the computer *encoded* representation of
the strings?
How about the "l" in "al-" or "il-" which is assimilated to a
following Sun-letter?
Arabic letters have no diacritical marks. The dots are as integral to
the letters as the dot of i and j.
> How about the "l" in "al-" or "il-" which is assimilated to a
> following Sun-letter?
Never in writing, and only in informal transcriptions.
--
Peter T. Daniels gram...@att.net
: Yusuf> Lee Sau Dan <dan...@informatik.uni-freiburg.de> wrote: :
: Yusuf> Internationalization software libraries have been around
: Yusuf> and : standardized for around 2 decades. They can handle
: Yusuf> these "simple" and : common situations (collating sequence)
: Yusuf> well. (Isn't sorting words in : an Arabic dictionary even
: Yusuf> harder?)
: Yusuf> if arranged by "roots" yes. otherwise rather
: Yusuf> straightforward. purely alphabetical dictionaries exist and
: Yusuf> it is the norm for languages other than arabic using the
: Yusuf> arabic script.
: How do you handle the diacritical marks when you sort the Arabic
if you mean the vowel signs? words with the same consonantal skeleton but
differing only in the short vowels are usually ararnged in the order
of fatHa /a/, Damma /u/, kasra /i/ (corresponding to alif, waw, ya). at
least so in a persian dictionary.
: words? Is that trivial with the computer *encoded* representation of
: the strings?
: How about the "l" in "al-" or "il-" which is assimilated to a
: following Sun-letter?
either not included in the alphabetization or treated in its garphic
value.
:> How about the "l" in "al-" or "il-" which is assimilated to a
:> following Sun-letter?
: Never in writing, and only in informal transcriptions.
turksih romanizations usually romanize according to sound, and include
case endings, at least in context.
: --
: Peter T. Daniels gram...@att.net
Yep.
>> How do you handle the diacritical marks when you sort the
>> Arabic words? Is that trivial with the computer *encoded*
>> representation of the strings?
Peter> Arabic letters have no diacritical marks. The dots are as
Peter> integral to the letters as the dot of i and j.
How about the "w"-like shape ("shadda"?) used to denote doubled
letter?
>> How about the "l" in "al-" or "il-" which is assimilated to a
>> following Sun-letter?
Peter> Never in writing, and only in informal transcriptions.
Don't they mark it with that "w"-like shape?
Harlan> Yes, I thought I remembered having read that the Spanish
Harlan> Academy got rid of these as distinct letters.
>> Has the ñ remained a letter?
the> Yep.
Maybe, they should have replaced it with "nn" and treated it in the
same manner as "rr" and "ll". :)
I disagree on the basis that "ñ" takes up only one "space" while "ll", "rr" and "ch"
each take up two.
Charles
It's the mark most likely to be used (after hamza) in an otherwise
unpointed text, but it too is optional. It's morphophonemic, so not
exactly a diacritic on a letter; presumably if a dictionary lemma
carried a shadda, it would come after a lemma that didn't.
Why would a dictionary entry begin with al-? English dictionaries don't
put "a" or especially "the" before the noun entries (and in poetry
indexes, for instance, the ones that ignore the article are a lot easier
to use than the ones that put all lines starting with "the" under <T>).
> Lee Sau Dan wrote:
>
> > Harlan> Yes, I thought I remembered having read that the Spanish
> > Harlan> Academy got rid of these as distinct letters.
> >
>> Has the ñ remained a letter?
> >
> Yep.
> >
> > Maybe, they should have replaced it with "nn" and treated it in the
> > same manner as "rr" and "ll". :)
> >
> I disagree on the basis that "ñ" takes up only one "space" while "ll",
"rr" and "ch"
> each take up two.
According to http://www.evertype.com/alphabets/spanish.pdf
"ch" and "ll" have been split up, but not "rr". This digraph remains a
separate letter, coming after "r". Is this correct? Seems weird.
John.
Seems both weird and inconsistant. But it it back up my credo that every
language must have at least one strange quirk to call its own. :)
Charles