Accent marks, French and otherwise
Don Blaheta
I'm currently in a first semester French class. According to my
teacher, the accent marks in various French words have absolutely no
relation to the pronunciation of those words, and merely serve to
distinguish written words from each other; just extra marks to be
memorized.
I find this rather hard to believe; while I'm sure simple language
evolution has changed the meaning of the marks over time, I would assume
they still have *some* relation to the pronunciation of the words they
appear in. For instance, ' seems to (at least sometimes) indicate a
stress; other than that, though, I can't figure out any obvious
interpretations.
In a more general sense, how do such markings tend to evolve in a
language? I mean, Latin itself used the Roman alphabet, completely
unmarked, yet I believe English is nearly unique among modern languages
in using a Roman alphabet devoid of diacriticals. German, for instance,
has the umlaut (¨), which developed to mark the occurence of umlaut.
Spanish has the tilde (~) to mark the palatalization of the n, and
Portuguese and Provençal use the same mark to note nasalization of
vowels. But what prompted scribes to begin using these marks? Do
languages which use other alphabets/syllabaries generally use some sort
of diacritical markings, or is that unique to the Roman alphabet langs?
(I take that back, I understand that the katakana of Japanese uses two
dots to indicate voicedness?)
Don
-=-=-=-Don Blaheta-=-=-=-bla...@quincy.edu-=-=-=-dbl...@aol.com-=-=-=-
The state law of Pennsylvania prohibits singing in the bathtub.
Peter k Chong
Well, on accent marks, the Russian language (I'm not sure about the other
Slavonic tongues) also uses them on their cyrillic characters to indicate
stress. I'm not sure why we use accents. Come to think of it, the
Mesopotamian cuneiform scripts, Chinese ideograms, Egyptian hieroglyphics
and Turkic, Gothic and Hungarian runic scripts didn't use accents either.
In any case though, I think the accents make life easier when I write in
French, German or Hungarian vs. Latin & English. (except for those damned
Latin long marks - Hey, wait a minute, Latin was marked! :-o - just not
as much as modern Roman-lettered languages... :-) )
Regards
Peter Chong
C A R
Don Blaheta (bla...@Shamino.quincy.edu) wrote:
: I'm currently in a first semester French class. According to my
: teacher, the accent marks in various French words have absolutely no
: relation to the pronunciation of those words, and merely serve to
: distinguish written words from each other; just extra marks to be
: memorized.
He is correct. In Modern French, accent marks are phonemic (that is, they
distinguish actual vowel sounds. In Spanish, they are mainly stress
marks, or in rare cases used to distinguish homophones [té = tea vs. te =
you (object)]. In Irish, they are MAINLY used to distinguish long vs.
short vowels (a vs. á).) "é" is a different sound from "è".
continued below...
: I find this rather hard to believe; while I'm sure simple language
: evolution has changed the meaning of the marks over time, I would assume
: they still have *some* relation to the pronunciation of the words they
: appear in. For instance, ' seems to (at least sometimes) indicate a
: stress; other than that, though, I can't figure out any obvious
: interpretations.
"L'accent aigu" (acute) is used solely over "e", and determines what kind
of "e" is pronounced. "L'accent grave" (grave) is used over a (I think to
separate homophones) and e (vowel quality). (Circumflex) is used over all
five standard Roman-character vowels. They indicate quality (in the case
of older speakers, [a, e, i, o, u] are shorter in length than [â, ê, î, ô,
û]. Historically, words beginning with "é + consonant" or "ê + consonant"
used to begin with "es + consonant". In general, circumflex reflects
either a disappeared "s" [être, OFr estre; château, OFr castel (?)] or
long vowel [Past Hist. "donâmes" from Lat. "donavimus" (?)].
: In a more general sense, how do such markings tend to evolve in a
: language? I mean, Latin itself used the Roman alphabet, completely
: unmarked, yet I believe English is nearly unique among modern languages
: in using a Roman alphabet devoid of diacriticals.
Not so. Dutch has very sparce marking of words with diacriticals. They
are used only to (in case of "trema") separate vowel sounds
[almost always on "e"] (tweeënde = second; België = Belgium) or distinguish
homophones [voor = fore (?) vs. vóór = for (?)]. English used to do this
with word in coo- [coöperate] or in borrowed words [café, naïve, etc...].
As far as the use or non-use of diacriticals, both English and Dutch could
get along fine without them.
: German, for instance,
: has the umlaut (¨), which developed to mark the occurence of umlaut.
: Spanish has the tilde (~) to mark the palatalization of the n,
...and acute accents as above...
: and
: Portuguese and Provençal use the same mark to note nasalization of
: vowels.
Wrong. Portuguese marks only "a" and "o" with tilde (~) for stressed
nasals [versus "am" and "om" for non-stressed nasals]. To my knowledge,
Provençal does not use tilde to mark nasality. [But Breton DOES used "ñ"
to mark nasal vowels].
: But what prompted scribes to begin using these marks? Do
: languages which use other alphabets/syllabaries generally use some sort
: of diacritical markings, or is that unique to the Roman alphabet langs?
There developed only twenty-six characters in the classical Roman alphabet
(ash, edh, and thorn are Runic in origin; eszet is a ligature). Many
languages have more sounds than characters. English is a prime example
(12-14 vowel, 20-22 consonant = 44 phonemes). Others are Czech [which
distinguishes dental d,n,t,l,r,s,z from palatal d,n,t,l,r (r-hacek),s,z
from full affricate dz-hacek, c-hacek, z-hacek, s-hacek.], Polish, much
the same, except l/l' which became [w]/[l]. Vietnamese is an extreme
case (diacriticals mark three "a", two "e", three "o" and two "u" vowels
(quality, not quantity), plus six tones). Latin had ten vowels (if one
counts long and short separately, considering those ten (of quantity)
became seven in VulgarLatin (of quality). Greek use to have three types
of accents (distinguish pitch AND stres), and two types of breaths (smooth
= none; rough = [h]). (Modern Greek has only acute accent for stress
only, and no breaths.)
Other alphabets/syllabries with diacritics:
Arabic (marks vowels, "initial long a", glottal stop)
Amharic (glottalized vs. non-glottalized (/t./ vs. /t/),
labialization (k to kw)
Nagari and descendants [Hindi et.al. (nasalization)]
: (I take that back, I understand that the katakana of Japanese uses two
: dots to indicate voicedness?)
[k, s, t, h [from OldJ "p"]] mark voice with two short slashes on the
upper right hand corner of the syllable [becomes g, z, d, b]. (NewJ "p"
mark by a small circle in place of slashes.)
: Don
: -=-=-=-Don Blaheta-=-=-=-bla...@quincy.edu-=-=-=-dbl...@aol.com-=-=-=-
: The state law of Pennsylvania prohibits singing in the bathtub.
Just a start...hope it helps.
--
Reid
University of Maryland-College Park
pea...@wam.umd.edu
--"I love you man..."
--"..You're not gettin' my Bud Light."
Lee Sau Dan ~{@nJX6X~}
>>>>> "Don" == Don Blaheta <bla...@Shamino.quincy.edu> writes:
Don> I'm currently in a first semester French class. According to
Don> my teacher, the accent marks in various French words have
Don> absolutely no relation to the pronunciation of those words,
Don> and merely serve to distinguish written words from each
Don> other; just extra marks to be memorized.
Yes, you can think of them as extra symbols. The Latin alphabet has
only 26 letters, and only 5 (or 6, if you count "y" as a vowel) of
which are vowels. This is not sufficient for many languages. So,
they have to add accent marks to "create" several more vowel letters
while still retaining their Latin-look. This is true for French
(consider e-grave, e-acute and e-circumflex), German (a-umlaut,
o-umlaut and u-umlaut), Finnish, and especially Vietnamese.
However, some of the marks DO have relation to pronunciation. The "
in a French word (e.g. na{i"}ve) indicates that the vowels should be
pronounced separately, rather than treated as a digraph. In a tone
language such as Vietnamese and Pinyin system of Chinese, accent marks
are used to denote the tone of a syllable. [ That's why written
Vietnamese have so many accent marks. A letter in Vietnamese can
carry 2 accent marks: One for vowel quality (i.e. creating extra
letters for distinguishing the large amount of vowels) and the other
for the tone. ]
Don> I find this rather hard to believe; while I'm sure simple
Don> language evolution has changed the meaning of the marks over
Don> time, I would assume they still have *some* relation to the
Don> pronunciation of the words they appear in. For instance, '
Don> seems to (at least sometimes) indicate a stress; other than
Don> that, though, I can't figure out any obvious interpretations.
The circumflex in French sometimes indicates a historic "s" in Old
French spellings. For example, "ho^tel" and "fore^t" were spelt as
"hostel" and "forest" in Old French. [ These word were borrowed into
English with their old spellings. ] The "s" sound is lost and the
vowels before the lost "s" now carry a circumflex.
Don> Do languages which use other
Don> alphabets/syllabaries generally use some sort of diacritical
Don> markings, or is that unique to the Roman alphabet langs? (I
Don> take that back, I understand that the katakana of Japanese
Don> uses two dots to indicate voicedness?)
In Arabic, people use diacritical marks to write short vowels.
Traditionally, the Arabic script only records the consonants and long
vowels. Short vowels are not written down. So a reader have to
determine the short vowels from a piece of writing by context. Today,
the Arabic speakers still write Arabic this way. However, in
elementary text books, diacritical marks are used to write the short
vowels, so that learners can determine the short vowels easily.
Diacritical marks have the advantage that they do not distort the form
of the words.
The Japanese script *arguably* have diacritical marks for the kanas.
It has 2 diacritical marks. The first one (two dots at the upper
right corner), when added to a kana, changes the beginning consonant
of that syllable from unvoiced to voiced. This clever use of
diacritical mark effectively reduces the required amount of kanas by
half! The other mark (a small circle at the upper right corner), when
added to "ha", "hi", "hu", "he" and "ho", changes them to "pa", "pi",
"pu", "pe" and "po".
I have never seen the Korean script with diacritical marks. Well... I
don't know Korean. The Chinese ideographic script never carries
diacritical marks.
--
Lee Sau Dan
-------------------------------------------------------------------------------
URL: http://www.cs.hku.hk/~sdlee
e-mail: sd...@cs.hku.hk
Miguel Carrasquer Vidal
bla...@Shamino.quincy.edu (Don Blaheta) wrote:
>I'm currently in a first semester French class. According to my
>teacher, the accent marks in various French words have absolutely no
>relation to the pronunciation of those words, and merely serve to
>distinguish written words from each other; just extra marks to be
>memorized.
Be^tises.
>I find this rather hard to believe; while I'm sure simple language
>evolution has changed the meaning of the marks over time, I would assume
>they still have *some* relation to the pronunciation of the words they
>appear in. For instance, ' seems to (at least sometimes) indicate a
>stress; other than that, though, I can't figure out any obvious
>interpretations.
Not stress. Stress doesn't have to be marked in French, as it always
falls on the last syllable (I'm simplifying here: French _has_ no
stress might be a more adequate description.)
The accent aigu (') marks _vowel quality_: the close variety of "e".
The accent grave (`) and the circonflex (^) mark the open variety of
"e".
Of course the system is not general enough to be of real value: an
unmarked e can be either /@/ "e muet", /e/ close e, or /E/ open e,
depending on consonantal context, position in the word/syllable, or
just plain arbitrariness.
As to the diacritic signs on the other vowels (a`, a^, i^, o^, u^),
your teacher is right about `a, but the circonflex (mostly used where
there was a now lost <s> sound in Old French), does give a clue about
pronunciation: the vowel is usually long. In the case of a^, the a is
of the "back" variety /A/ (as opposed to mostly front "a" /a/ without
the circonflex). In the case of o^ the vowel is pronounced closed o
/o/, not open /O/.
>In a more general sense, how do such markings tend to evolve in a
>language? I mean, Latin itself used the Roman alphabet, completely
>unmarked, yet I believe English is nearly unique among modern languages
>in using a Roman alphabet devoid of diacriticals. German, for instance,
>has the umlaut (¨), which developed to mark the occurence of umlaut.
The umlaut developed out of a superscript <e>.
>Spanish has the tilde (~) to mark the palatalization of the n, and
>Portuguese and Provençal use the same mark to note nasalization of
>vowels.
The tilde developed out of a superscript <n> (Spanish <n~> is
therefore a way of writing <nn>: Latin annum -> Spanish an~o).
I don't believe ~ is used in Provenc,al, though. Speaking of <c,>,
the cedilla used to be a subscript z (c, < cz).
> But what prompted scribes to begin using these marks?
Well, if you had to copy entire books by hand, you'd start using some
abbreviations too. The Medieval scribes did not invent shorthand, but
they did make it easy on themselves by using cursive scripts and lots
of abbreviations. Vowel+n/m usually became vowel with a little wiggle
on top (the tilde), &c. [<= another example of a monkish
abbreviation].
>Do languages which use other alphabets/syllabaries generally use some sort
>of diacritical markings, or is that unique to the Roman alphabet langs?
>(I take that back, I understand that the katakana of Japanese uses two
>dots to indicate voicedness?)
Arabic is known for its dots, as is Hebrew. Cyrillic also uses a few
diacritics (and the Old Church Slavonic manuscripts are full of
abbreviations of the kind I mentioned above). Devanagari (the Indian
script used to write Sanskrit) abbreviates s and m after a vowel by
the use of "dots" (visarga and anusva^ra). I think the practice is
almost universal. English and ASCII are the exception rather than the
rule.
==
Miguel Carrasquer Vidal ~ ~
Amsterdam _____________ ~ ~
m...@pi.net |_____________|||
========================== Ce .sig n'est pas une .cig
Mark Rosenfelder
In article <50ttop$j...@iolo.quincy.edu>,
Don Blaheta <bla...@Shamino.quincy.edu> wrote:
>I'm currently in a first semester French class. According to my
>teacher, the accent marks in various French words have absolutely no
>relation to the pronunciation of those words, and merely serve to
>distinguish written words from each other; just extra marks to be
>memorized.
Your teacher is crazy. Here's a brief tour of the accents. I have to
put them after the letters to make them generally readable.
e' and e` are distinct sounds: e' is a more open sound, more or less like
English 'break'; e` is more closed, like in English 'met'. An e without
an accent very often becomes an indistinct sound (a schwa or a rounded oe)
or is dropped (e.g. _premier_ = [pr@mje], je suis = [Sui]).
A circumflex indicates a vowel that many though not all French speakers
lengthen: compare patte [pat] with pa^te [pa:t], bette [bet] with be^te
[be:t], cote [kOt] with co^te [ko:t].
A diaresis indicates a vowel that is pronounced separately, as opposed to
forming a diphthong with the previous vowel: compare nai"ve [naiv] with
glaive [glE:v], Moi"se [mOi:z] with moise [mwa:z].
Grave accents over a and u are indeed used only to distinguish homophones:
ou vs. ou`, a vs. a`, la vs. la`, etc. (Perhaps your teacher was referring
*only* to these words?)
Finally, c with cedilla is consistently used to indicate a soft c (pronounced
[s]) where one might otherwise have expected [k]: lec,on [l@so~].
Poll Dubh
In article <50vdp1$n...@news.inforamp.net>,
Peter k Chong <pet...@inforamp.net> wrote:
>Well, on accent marks, the Russian language (I'm not sure about the other
>Slavonic tongues) also uses them on their cyrillic characters to indicate
>stress.
So does Spanish, and to a lesser extent Italian. (That is, the accent in
Italian is never written on an unstressed syllable, but it's more often
optional than in Spanish---and less often so than in Russian. :-( )
Russian can also have a breve sign on the i, and a trema on the e.
The latter implies a stressed vowel, the former precludes stress (or
indeed pure vowelness) and is the only one that should never be omitted.
> I'm not sure why we use accents.
It depends on the language. In some cases the accent developed out of a
digraph in which the second letter came to be written above the first.
I think this is the case of the French circumflex (superscripted "s"),
the Czech etc. haczek (superscripted "z"), the German Umlaut (superscripted
"e"), the Spanish tilda (superscripted "n"---or was it "h"?), and many
others. Where the accent no longer corresponds to a difference in
pronunciation, it is usually kept either to disambiguate the spelling or
out of sheer conservatism (witness the situation in Greek until recently).
In other cases accents provide a way of extending the set of available
vowels or consonants within the basic framework of the Latin alphabet.
Examples in this category include Hungarian, Irish, Esperanto, Wolof, ...
> Come to think of it, the
>Mesopotamian cuneiform scripts, Chinese ideograms, Egyptian hieroglyphics
>and Turkic, Gothic and Hungarian runic scripts didn't use accents either.
I'm not quite sure how you can make that statement about Chinese characters.
[Aside: how old is Devanagari? That most certainly uses accents...]
>In any case though, I think the accents make life easier when I write in
>French, German or Hungarian vs. Latin & English. (except for those damned
>Latin long marks - Hey, wait a minute, Latin was marked! :-o - just not
Ah, but since when? I have yet to see a stone inscription with a macron.
Greek accents and breathings only started being written when it became
necessary to give reading cues to non-natives.
Ralph T. Edwards
In article <510b0l$p...@halley.pi.net>, m...@pi.net (Miguel Carrasquer
Vidal) wrote:
In the case of o^ the vowel is pronounced closed o
> /o/, not open /O/.
>
Yes this the standard dictionary pronunciation (and for au as well). My
question is, what fraction of French speakers actually use closed /o/ in
closed syllables? Seems to me a lot use /O/ whenever the syllable ends in
a pronounced consonant.
--
R.T.Edwards r...@elmo.lz.att.com 908 576-3031
Miguel Carrasquer Vidal
r...@elmo.lz.att.com (Ralph T. Edwards) wrote:
>In article <510b0l$p...@halley.pi.net>, m...@pi.net (Miguel Carrasquer
>Vidal) wrote:
>In the case of o^ the vowel is pronounced closed o
>> /o/, not open /O/.
>>
>Yes this the standard dictionary pronunciation (and for au as well). My
>question is, what fraction of French speakers actually use closed /o/ in
>closed syllables? Seems to me a lot use /O/ whenever the syllable ends in
>a pronounced consonant.
I think that in a word like "rose", the standard "Parisian"
pronunciation is indeed /ro:z/. This was not the case in older
French, and in many modern dialects. Dutch, for instance,
has borrowed these words with an /O:/ sound, which isn't even
a real Dutch phoneme, as it is _only_ used in these borrowings
from French (roze /rO:z@/, contro^le /kOntrO:l@/, zone /zO:n@/).
Miguel Carrasquer Vidal
pea...@wam.umd.edu (C A R) wrote:
>Don Blaheta (bla...@Shamino.quincy.edu) wrote:
>: I'm currently in a first semester French class. According to my
>: teacher, the accent marks in various French words have absolutely no
>: relation to the pronunciation of those words, and merely serve to
>: distinguish written words from each other; just extra marks to be
>: memorized.
>He is correct. In Modern French, accent marks are phonemic (that is, they
>distinguish actual vowel sounds.
And thus, he is _not_ correct.
[good points snipped]
>Not so. Dutch has very sparce marking of words with diacriticals. They
>are used only to (in case of "trema") separate vowel sounds
>[almost always on "e"] (tweeënde = second; België = Belgium) or distinguish
You probably mean tweee"ndertig "32". "Second" is plain tweede.
Another nice one is zeee"gel, sea-urchin.
>homophones [voor = fore (?) vs. vóór = for (?)]. English used to do this
I hate to say this, but it's the other way round.
>with word in coo- [coöperate] or in borrowed words [café, naïve, etc...].
>As far as the use or non-use of diacriticals, both English and Dutch could
>get along fine without them.
There are also some aigus, graves and circonflexes used in borrowings
from French. I think there is in Holland a Society of People with
Accented Names [it had, I believe, a funny name with a diacritic in
it], who have united to force banks and other companies to finally
start using the right software to address them properly.
[many other good points snipped]
Miguel Carrasquer Vidal
Peter k Chong <pet...@inforamp.net> wrote:
>Well, on accent marks, the Russian language (I'm not sure about the other
>Slavonic tongues) also uses them on their cyrillic characters to indicate
>stress.
It certainly doesn't. That's just in introductory textbooks to aid
the beginning student. Pick up a real Russian book and you're on your
own, unfortunately.
>I'm not sure why we use accents. Come to think of it, the
>Mesopotamian cuneiform scripts, Chinese ideograms, Egyptian hieroglyphics
>and Turkic, Gothic and Hungarian runic scripts didn't use accents either.
>In any case though, I think the accents make life easier when I write in
>French, German or Hungarian vs. Latin & English. (except for those damned
>Latin long marks - Hey, wait a minute, Latin was marked! :-o - just not
>as much as modern Roman-lettered languages... :-) )
Nope. Latin long marks are only used in modern textbooks (and
etymological dictionaries). You won't find any real Roman inscription
using them.
You won't find any Greek inscriptions with accents and spiriti (is
that the plural?) either, but at least that practice started with the
Alexandrian and Byzantine clerks, so it's sufficiently genuine.
Rudy Vonk
Ralph T. Edwards wrote:
> =
=
> Yes this the standard dictionary pronunciation (and for au as well). M=
y
> question is, what fraction of French speakers actually use closed /o/ i=
n
> closed syllables? Seems to me a lot use /O/ whenever the syllable ends=
in
> a pronounced consonant.
Oh yes? Have you really ever heard "C=F4tes de Beaune" pronounced like
(appr.) "cot de bon" instead of "coat de bone"? It wouldn't even taste
the same!
-- =
___________________________________________
Rudy Vonk <mo...@asturnet.es> (business)
Oviedo, Spain <roo...@asturnet.es> (private)
Don't believe everything you hear or anything you say.
(from "Twenty Past Midnight")
_________________________________________________
Rudy Vonk
Poll Dubh wrote:
>
> So does Spanish, and to a lesser extent Italian. (That is, the accent in
> Italian is never written on an unstressed syllable, but it's more often
> optional than in Spanish---and less often so than in Russian. :-( )
It is never optional in Spanish or Italian. In Italian, it is used only
on the final syllable, but there is no choice involved. The word has it
or it doesn't
> Russian can also have a breve sign on the i, and a trema on the e.
> The latter implies a stressed vowel, the former precludes stress (or
> indeed pure vowelness) and is the only one that should never be omitted.
The former is the semivowel "y", the latter is a completely different
vowel "yo". It is stressed all right, and the trema is normally omitted.
> ...the Spanish tilda (superscripted "n"---or was it "h"?)
actually, it developed from "gn"
It gets worse after this. Why not make the distinction between
diacritical marks in general and accents in particular? Surely the
"accents" you refer to in devanagari cannot be described as "accents"
any more than the vowel marks in Hebrew or Arabic?
--
Ralph T. Edwards
> Ralph T. Edwards wrote:
> > =
>
> =
>
> > Yes this the standard dictionary pronunciation (and for au as well). M=
> y
> > question is, what fraction of French speakers actually use closed /o/ i=
> n
> > closed syllables? Seems to me a lot use /O/ whenever the syllable ends=
> in
> > a pronounced consonant.
>
> Oh yes? Have you really ever heard "C=F4tes de Beaune" pronounced like
> (appr.) "cot de bon" instead of "coat de bone"? It wouldn't even taste
> the same!
> -- =
>
Well if I heard it pronounced coat de bone, I'd know it was being said by
an English speaker.:-) The real question is not whether its "cot de bon"
but whether _every_ French speaker distinguishes it from cotte de bonne,
which definitely would taste different.
Stephen Rowland
m...@pi.net (Miguel Carrasquer Vidal) wrote:
>Nope. Latin long marks are only used in modern textbooks (and
>etymological dictionaries). You won't find any real Roman
inscription
>using them.
Some Roman inscriptions *did* mark long vowels, with a mark like
an acute or circumflex accent called an apex (or, in the case of
the letter I, with an extra-height letter extending above the top
of the other letters). But most inscriptions don't use the apex
at all, and those that do don't usually mark all long vowels.
>You won't find any Greek inscriptions with accents and spiriti
(is
>that the plural?) either
<Hesitantly> Spiritus? <Confidently> Breathings!
--
Stephen Rowland
10014...@compuserve.com
Erland Sommarskog
Thus quoth sing...@oort.ap.sissa.it (Poll Dubh):
>Russian can also have a breve sign on the i, and a trema on the e.
>The latter implies a stressed vowel, the former precludes stress (or
>indeed pure vowelness) and is the only one that should never be omitted.
True, indeed this character is indeed called i-kratkoje in Russian,
but from many circumstances it appears to be a letter on its own.
You don't find an entry for in the dictionary, but that is due to
the spelling rules of Russian. In Bulgarian there are a few words
starting with this letter, but my dictionary was to small to include
them. I suppose Ivan could tell us how they are treated in Bulgarian.
In any case, what is accent and what is not, is not an obvious case.
If you would ask me if Swedish uses any accents, I would tell you
that acute accent is used on "e" in a couple of words, mainly French
loans, to mark stress. In names a few more accents can appear, but
that's that. Hm, well, we have W as well, that is some of an accented
V so to speak. Then you would maybe look at me in disbelief and ask
about A-ring, and dotted A and O. But then I would tell you that they
are no more accented letters than Q or E is. Sure, they originate from
other letters in the Latin alphabet, but so does J, U and W. So when
talking about Swedish ÅÄÖ are three more letters, not accented A:s or
O:s.
--
Erland Sommarskog, Stockholm, som...@algonet.se
F=F6r =F6vrigt anser jag att QP b=F6r f=F6rst=F6ras.
B=65sid=65s, I think QP should b=65 d=65stroy=65d.
M. Murray
Erland Sommarskog (som...@algonet.se) wrote:
[snip]
: In any case, what is accent and what is not, is not an obvious case.
: If you would ask me if Swedish uses any accents, I would tell you
: that acute accent is used on "e" in a couple of words, mainly French
: loans, to mark stress. In names a few more accents can appear, but
: that's that. Hm, well, we have W as well, that is some of an accented
: V so to speak. Then you would maybe look at me in disbelief and ask
: about A-ring, and dotted A and O. But then I would tell you that they
: are no more accented letters than Q or E is. Sure, they originate from
: other letters in the Latin alphabet, but so does J, U and W. So when
: talking about Swedish ÅÄÖ are three more letters, not accented A:s or
: O:s.
All I would say is that your idea of what constitues an accented letter
isn't the same as mine. Your (arbitrary) attitude is the same as the Spanish
idea that ch and ll are single letters, so they put them separately in
dictionaries. My idea of common sense says that ll is two letters (after
all, it IS undeniably two l's), and that Ö is an accented O (it IS an O
with two dots on it). If you don't agree with my idea of common sense so be
it.
--
Martin Murray :: School of Chemistry, Bristol University, BS8 1TS, England
Neil Coffey
In article <rte-120996...@135.25.40.118>, Ralph T. Edwards
There are a lot of cases where the choice of o is ideolectal, but I think
such hesitation and variation exists in words such as '_au_xiliare',
'teleph_o_ner' rather than 'côte'/'cotte', 'votre/vôtre/ distinctions.
--
Neil Coffey
ne...@ccsware.demon.co.uk
cof...@teaching.physics.ox.ac.uk
B.Philip.Jonsson
In article <51hak7$3...@epimetheus.algonet.se>, som...@algonet.se (Erland
Sommarskog) wrote:
> Then you would maybe look at me in disbelief and ask
> about A-ring, and dotted A and O. But then I would tell you that they
> are no more accented letters than Q or E is. Sure, they originate from
> other letters in the Latin alphabet, but so does J, U and W. So when
> talking about Swedish ÅÄÖ are three more letters, not accented A:s or
> O:s.
OK agree from a purely Swedish-alphabetization-rules point of view, but
really, we must distinguish this from the fact that _graphically_ these
Swedish letters ARE indeed letters with accent marks -- or diacritics, as
they're properly called --. In some languages such letters are
alphabetized with their base letter, in others they are given a separate
alphabetic entry, rather like W in Swedish and English respectively.
Spanish even treats the digraph <ll> as a separate letter from <l>!
I think the trouble with diacritics is that since English uses no
diacritics anglophones seem to think they are somehow an unnecessary
frills. In fact diacritics were introduced to distinguishe sounds that had
no letters of their own in the alphabet of the phonologically not very
rich Latin language, which the Roman Church imposed on all its subjects in
the Middle Ages, regarding it as sacrosanct. Only in the religiously
rather more independent early England did scribes expand the Roman
alphabet with new letters, taken from runic script.
In fact, since English orthography uses only five letters to express about
20 vowel sounds it could need some dacritics: is Donovan Bailey the
fästest, the fæstest or even the fástest man in the world, I wonder? He's
definitely NOT the fâstest /feist^st/, anyway, since there is no such
word...
--
B.Philip.Jonsson
Colin Fine
In article <510b0l$p...@halley.pi.net>, Miguel Carrasquer Vidal
<m...@pi.net> writes
>bla...@Shamino.quincy.edu (Don Blaheta) wrote:
>>Do languages which use other alphabets/syllabaries generally use some sort
>>of diacritical markings, or is that unique to the Roman alphabet langs?
>>(I take that back, I understand that the katakana of Japanese uses two
>>dots to indicate voicedness?)
>
>Arabic is known for its dots, as is Hebrew. Cyrillic also uses a few
>diacritics (and the Old Church Slavonic manuscripts are full of
>abbreviations of the kind I mentioned above). Devanagari (the Indian
>script used to write Sanskrit) abbreviates s and m after a vowel by
>the use of "dots" (visarga and anusva^ra). I think the practice is
>almost universal. English and ASCII are the exception rather than the
>rule.
>
In many languages, the letters-with-diacritics are regarded as separate
letters in their own right. This means that there's little difference
between a-with-two-dots being a letter in Finnish, and i-with-one-dot
being a letter in English. It's true that we don't use the dot to
distinguish two letters (unlike Turkish), but English still has (what
might as well be called) diacritics.
Then Welsh (for example) has letters 'ch' 'll' etc. as well!
--
-----------------------------------------------------------------------
| Colin Fine 66 High Ash, Shipley, W Yorks. BD18 1NE, UK |
| Tel: 01274 592696/0976 436109 e-mail: co...@kindness.demon.co.uk |
| "We're all in a box and the instructions for getting out |
| are on the outside" -K.B.Brown |
-----------------------------------------------------------------------
Keith C. Ivey
co...@zeus.bris.ac.uk (M. Murray) wrote:
>All I would say is that your idea of what constitues an
>accented letter isn't the same as mine. Your (arbitrary)
>attitude is the same as the Spanish idea that ch and ll
>are single letters, so they put them separately in
>dictionaries. My idea of common sense says that ll is
>two letters (after all, it IS undeniably two l's), and
>that Ö is an accented O (it IS an O with two dots on it).
>If you don't agree with my idea of common sense so be
>it.
Although I share your feelings (and I posted about it in a
similar thread that started in alt.usage.english a while back),
this "common sense" is not so simple to define.
Why doesn't your idea of common sense say that "G" is undeniably
"C" with an added crossbar, or that "j" is undeniably "i" with a
tail, or that "U" is undeniably a rounded "V", or that "W" is
undeniably two "V"s? In some sense all of these are as true as
your claim about Swedish "Ö".
[posted and mailed]
Keith C. Ivey <kci...@cpcug.org> Washington, DC
Untangling the Web <http://www.eei-alex.com/eye/utw/>
Magnus Olsson
In article <DxvvJ...@fsa.bris.ac.uk>, M. Murray <co...@zeus.bris.ac.uk> wrote:
>Erland Sommarskog (som...@algonet.se) wrote:
>
>[snip]
>
>: In any case, what is accent and what is not, is not an obvious case.
>: If you would ask me if Swedish uses any accents, I would tell you
>: that acute accent is used on "e" in a couple of words, mainly French
>: loans, to mark stress. In names a few more accents can appear, but
>: that's that. Hm, well, we have W as well, that is some of an accented
>: about A-ring, and dotted A and O. But then I would tell you that they
>: are no more accented letters than Q or E is. Sure, they originate from
>: other letters in the Latin alphabet, but so does J, U and W. So when
>: talking about Swedish ÅÄÖ are three more letters, not accented A:s or
>: O:s.
>
>isn't the same as mine.
Just to makes things perfectly clear: this is not a matter of personal
opinion (yours or Erland's) but of official definition. Swedish
lexicographers treat the "dotted o" as a separate letter, placing it
last in the alpahabet. German lexicographers treat it as an accented
o, equivalent to the letter combination "oe".
> Your (arbitrary) attitude is the same as the Spanish
>idea that ch and ll are single letters, so they put them separately in
>dictionaries.
Exactly. But who's to say the Spanish are wrong?
> My idea of common sense says that ll is two letters (after
>all, it IS undeniably two l's), and that Ö is an accented O (it IS an O
>with two dots on it). If you don't agree with my idea of common sense so be
>it.
You're being more than a little bit ethnocentric and superior here,
aren't you? *Your* idea of "ocmmon sense" is what you've been taught
in your English-speaking school. Please realize that that doesn't make
it a universal truth.
You consider the letter "w" a separate letter, and call it "double-U".
We Swedes call it (in translation) "double-V" and regard it as a variant
of "v". Swedish libraries place books by Vinge after those by Wilde -
is that wrong? Not if you're speaking Swedish!
If you could ask an ancient Roman - they, after all, invented our
alphabet - he would regard "g" as a variant form of "c", and "u" as a
variant form of "v". Are you implying that there was anything wrong
with the ancient Romans' common sense?
--
Magnus Olsson (m...@df.lth.se)
Magnus Olsson
Reposting article removed by rogue canceller.
Andras Malatinszky
>>all, it IS undeniably two l's), and that Ö is an accented O (it IS an O
>>with two dots on it). If you don't agree with my idea of common sense so
be
>>it.
-- with a third horizontal bar.
Poll Dubh
In article <527frg$u...@news2.h1.usa.pipeline.com>,
Not quite. The relationship you describe between E and F does not hold for
lower case (e and f), nor for script capital E and F. The description of Ö
as O with a trema, by contrast, does hold across script styles[*]. It is also
rooted in the origin of Ö, whereas if you look at the origin of E you'll see
that the vertical stroke is a later development (absent in the lower case
version "e"). [Or perhaps I should say that the right angles at the bottom
and top left have come later.]
The distinction between a letter with a diacritic and a different letter
is very much in the eye of the beholder. This thread cannot possibly
achieve a consensus without a set of very precise definitions which most
of us may not be pedantic enough to be really interested in.
[*] The trema need not always be two dots. Some styles may use two long
strokes, or a Fraktur e, or... One has to avoid tempting identifications
like that of German ö with French or English ö (o with diaeresis). The
ISO 10646 (Unicode) standard attempts to deal with this sort of issue,
I believe.
Lee Sau Dan ~{@nJX6X~}
>>>>> "Poll" == Poll Dubh <sing...@oort.ap.sissa.it> writes:
Poll> Not quite. The relationship you describe between E and F
Poll> does not hold for lower case (e and f), nor for script
Poll> capital E and F. The description of Ö as O with a trema, by
Poll> contrast, does hold across script styles[*]. It is also
Poll> rooted in the origin of Ö, whereas if you look at the origin
Poll> of E you'll see that the vertical stroke is a later
Poll> development (absent in the lower case version "e"). [Or
Poll> perhaps I should say that the right angles at the bottom and
Poll> top left have come later.]
What do you think about the letters "J" and "U", then?
Originally, Latin didn't have "J" and "U". Their functions were
carried by the letters "I" and "V". So, in some ancient texts, you'll
find "V" and "I" appearing at odd places, where you would expect a "U"
or "J". Later, people invented "J" to separate the two uses
(consonant/vowel) of "I". The case is similar for the letter "U".
So, shall we consider "J" as another "I", and "U" as nothing but
another "V"?
Poll Dubh
In article <7fg2467...@wisdom.cs.hku.hk>,
Lee Sau Dan ~{@nJX6X~} <sd...@cs.hku.hk> wrote:
>So, shall we consider "J" as another "I", and "U" as nothing but
>another "V"?
In some contexts, certainly. I think I'm still allowed to write "jeri"
for "ieri", even though it will look hopelessly old-fashioned nowadays.
(Earlier this century it would have been perfectly normal and expected.)
Likewise, in the same language, W is really still only VV (except in foreign
loanwords).
But your comment actually bore no relevance to my original point, which
was to counter the claim that calling Ö "an O with two dots above" is
exactly analogous to calling E "an F with an extra stroke at the bottom".
I repeat: the former holds across styles of script (for some suitable concept
of "two dots"), the latter doesn't.
Anyway, the Italian name for "J" is "i lunga". The French name for "W" is
"double v", the English one "double u". So if I ignore for a moment your
unnecessarily polemic "nothing but" (I *had* said that these distinctions
were mostly in the eye of the beholder), my answer to your question is yes.
Of course all this is from a graphical rather than functional point of view.
At the functional level, you can call digraphs or trigraphs separate letters,
and alphabetise them accordingly. W and the Russian ery (no relation with
the Italian word for "yesterday" I mentioned earlier) are two examples. Try
also Spanish ll, ch, ...
Colin Fine
Andras Malatinszky
On Sep 24, 1996 16:14:36 in article <Re: Accent marks, French and
otherwise>, 'sing...@oort.ap.sissa.it (Poll Dubh)' wrote:
>In article <527frg$u...@news2.h1.usa.pipeline.com>,
>Andras Malatinszky <fai...@usa.pipeline.com> wrote:
>>In article <DxvvJ...@fsa.bris.ac.uk>, M. Murray <co...@zeus.bris.ac.uk>
>>wrote:
>>>> My idea of common sense says that ll is two letters (after
>>>>with two dots on it). If you don't agree with my idea of common sense
so
>>be
>>>>it.
>
>>Excuse me, but this is about as silly as saying that an E is basically an
F
>>-- with a third horizontal bar.
>
>lower case (e and f), nor for script capital E and F. The description of Ö
>as O with a trema, by contrast, does hold across script styles[*]. It is
also
>rooted in the origin of Ö, whereas if you look at the origin of E you'll
see
>that the vertical stroke is a later development (absent in the lower case
>version "e"). [Or perhaps I should say that the right angles at the bottom
>and top left have come later.]
>
>The distinction between a letter with a diacritic and a different letter
>is very much in the eye of the beholder. This thread cannot possibly
>achieve a consensus without a set of very precise definitions which most
>of us may not be pedantic enough to be really interested in.
At first I was going to continue the argument but then I read this
preceding sentence and realised that, indeed, this whole issue is probably
not in the forefront of people's interest.
>[*] The trema need not always be two dots. Some styles may use two long
>strokes,
This is totally unrelated to the original subject, but may be interesting:
in Hungarian an o with two dots is not the same as an o with two long
strokes. The word "toke" means 'capital' (as in accounting) when the o has
two long strokes and means 'his balls' (in the rude sense) when it has the
two dots on top.
Avi Jacobson
I have been following this thread since it started, and cannot resist
presenting my own opinion:
It is the native readers and writers of a language who determine, with
regard to that language only, whether a particular orthographic symbol
which is historically derived from two discrete symbols (including the
case of a "base" symbol and a diacritic) should be considered a "single
letter", "two letters", or "a letter plus a diacritic".
It has correctly been noted here that W is clearly derived from VV (or
UU, since U and V are historically the same letter), that I and J were
once calligraphic variations of the same letter -- as, indeed, were U
and V. This is totally irrelevant to readers of most languages which
use these letters, since they clearly consider I, J, V, U, and W as
separate letters.
For the Spanish reader, "ll" is a single letter (and may someday evolve
typographically into a connected version, as have "ss" in German and
"ij" in Dutch). In languages where o and ö are used as separate
letters, they are considered to be such. In English, on the other hand,
the ö sometimes seen in "coöperate" (God, it's been a long time since an
issue of The New Yorker made its way to this corner of the Middle
East!), is considered by readers of English to be nothing more than "an
o with a trema (or, to the less educated, two dots) on top".
Russian bI (called [jerÏ] by old Russians, [Ï] by young Russians, and
[sIktI wVn] by several British students of Russian I have met) once
consisted of "two separate letters" -- a "soft sign" or [mjaxkij znak]
plus a vowel letter which has now disappeared from the Russian alphabet,
though it remains in other Slavic languages' variations of the Cyrillic
alphabet, such as Ukrainian.
Arabic, which has no "v" or "p" sound, uses an "f" with three dots over
it for "v" (Arabic speakers consider it to be a "strange version" of f,
or f with a diacritic), and a "b" with three dots under it for "p".
Farsi uses the same "p" but considers it a bona fide independent
consonant. It does _not_ use the "funny" v; rather, it uses the Arabic
w for that sound, since there is no [w] sound in Farsi.
Yiddish, basically a Germanic language, uses a modified form of the
Hebrew alphabet. Vowels, written as a combination of Hebrew guttural
consonant signs (aleph, ayin) and Hebrew vowel diacritics, are
considered independent letters: [o] is kometz-aleph (a Hebrew
glottal-stop consonant combined with a Hebrew long-a vowel diacritic),
while [a] is pasech-aleph (the same Hebrew glottal-stop consonant
combined with a Hebrew short-a vowel diacritic).
My point is that historical origins and pure logic (whatever that may
be) have little to do with the issue: like distinctions between colors,
distinctions between letters are a cultural issue and have no absolute
answers.
--
Avi Jacobson, email: avi_...@netvision.net.il | When an idea is
Home Page (Israel): | wanting, a word
http://www.netvision.net.il/php/avi_jaco | can always be found
Mirror Home Page (U.S.): | to take its place.
http://www.geocities.com/Paris/4034 | -- Goethe
Erland Sommarskog
>lower case (e and f), nor for script capital E and F. The description of Ö
>as O with a trema, by contrast, does hold across script styles[*].
Oh, it you want another example look at O as a C with an extra circle
arc. That holds across styles.
You can also consider a Latvian letter which in it is uppcase form is a
"G" with a comma under, but which in its lowercase form is a "g" with
a comma above it. OK, I don't know whether this character is considered
a letter of its own in Latvian, but I would assume it is.
Simply, consistency over script styles is neither a sufficient or
necessary condition for diacriticness.
>The distinction between a letter with a diacritic and a different letter
>is very much in the eye of the beholder. This thread cannot possibly
>achieve a consensus without a set of very precise definitions which most
>of us may not be pedantic enough to be really interested in.
We can not achieve a consensus, period. At least not an multi-lingual
level, since these issues are inherently defined at the language level.
The only frame within this discussion could really make sense, is
among the speakers of a certain language. For instance, sometimes
we have long threads over in swnet.svenska (the Swedish newsgroup
about the Swedish language) on whether W is really a letter or not
in Swedish. It actually got as far that one guy called the Swedish
Academy and was told that W is indeed a letter. But it wasn't in
the alphabet I learnt in school, and most dictionaries and directories
co-sort V and W. On the other hand, in crosswords V and W are not
interchangeable where E and É are.
Coby (Jacob) Lubliner
In article <324D62...@netvision.net.il>,
Avi Jacobson <avi_...@netvision.net.il> wrote:
>It is the native readers and writers of a language who determine, with
>regard to that language only, whether a particular orthographic symbol
>which is historically derived from two discrete symbols (including the
>case of a "base" symbol and a diacritic) should be considered a "single
>letter", "two letters", or "a letter plus a diacritic".
Well, not quite. It is the language authorities -- academies,
lexicographers, editors and the like -- who do so.
>For the Spanish reader, "ll" is a single letter (and may someday evolve
>typographically into a connected version, as have "ss" in German and
>"ij" in Dutch).
Probably not. The best dictionary of Spanish (Maria Moliner)
already treats "ll" and "ch" as two-letter combinations (and
alphabetizes entries accordingly), and the Royal Spanish Academy has
announced its intention to take the same route in the next edition
of its dictionary. Besides, the use of computers would seem to
discourage the creation of new characters.
Coby
Ochlocrat
Erland Sommarskog wrote:
>
> The only frame within this discussion could really make sense, is
> among the speakers of a certain language. For instance, sometimes
> we have long threads over in swnet.svenska (the Swedish newsgroup
> about the Swedish language) on whether W is really a letter or not
> in Swedish. It actually got as far that one guy called the Swedish
> Academy and was told that W is indeed a letter. But it wasn't in
> the alphabet I learnt in school, and most dictionaries and directories
> co-sort V and W. On the other hand, in crosswords V and W are not
> interchangeable where E and É are.
Continuing this discussion but on a slightly different tangent...
In Croatian, the combination "lj" and "nj" are considered
a single letter. Not only do they have a separate section
in the dictionary (K, L, LJ, M, N, NJ, O...), but they occupy
a single square in crossword puzzles and the Croatian version
of Wheel of Fortune (!), and in words that are written vertically
(eg, on a placard), LJ and NJ are still written together horizontally.
Of course, in the Serbian Cyrillic alphabet the corresponding
letters are unambiguously single-letter glyphs.
Are all the above conditions true for Spanish LL and CH as well?
Avi Jacobson
Coby (Jacob) Lubliner wrote:
>
> In article <324D62...@netvision.net.il>,
> Avi Jacobson <avi_...@netvision.net.il> wrote:
>
> >It is the native readers and writers of a language who determine, with
> >regard to that language only, whether a particular orthographic symbol
> >which is historically derived from two discrete symbols (including the
> >case of a "base" symbol and a diacritic) should be considered a "single
> >letter", "two letters", or "a letter plus a diacritic".
>
> Well, not quite. It is the language authorities -- academies,
> lexicographers, editors and the like -- who do so.
Coby, I wish I had a sheqel for every edict of the Academy of the Hebrew
Language that has been totally ignored by practically all Israelis --
educated or not.
>
> >For the Spanish reader, "ll" is a single letter (and may someday evolve
> >typographically into a connected version, as have "ss" in German and
> >"ij" in Dutch).
>
> Probably not. The best dictionary of Spanish (Maria Moliner)
[Opinion daemon automatically adds:] ...in Coby's opinion,...
> already treats "ll" and "ch" as two-letter combinations (and
> alphabetizes entries accordingly), and the Royal Spanish Academy has
> announced its intention to take the same route in the next edition
> of its dictionary. Besides, the use of computers would seem to
> discourage the creation of new characters.
Why, Coby? The use of typewriters certainly didn't! Orthographies have
changed since that sorely-missed piece of hardware was invented, and the
typewriter manufacturers simply cast new dies.
Miguel Carrasquer Vidal
Ochlocrat <oc...@netcom.ca> wrote:
>In Croatian, the combination "lj" and "nj" are considered
>a single letter. Not only do they have a separate section
>in the dictionary (K, L, LJ, M, N, NJ, O...), but they occupy
>a single square in crossword puzzles and the Croatian version
>of Wheel of Fortune (!), and in words that are written vertically
>(eg, on a placard), LJ and NJ are still written together horizontally.
>Of course, in the Serbian Cyrillic alphabet the corresponding
>letters are unambiguously single-letter glyphs.
>Are all the above conditions true for Spanish LL and CH as well?
No. I'm not sure about Wheel of Fortune, but in crossword puzzles
LL and CH occupy two squares. On the other hand, Spanish Scrabble
does provide separate chips for CH and LL (but using C+H and L+L is
also acceptable, at least in my family).
Robert Underhill
In article <52i2ls$r...@news2.h1.usa.pipeline.com>,
fai...@usa.pipeline.com(Andras Malatinszky) wrote:
>
> This is totally unrelated to the original subject, but may be interesting:
> in Hungarian an o with two dots is not the same as an o with two long
> strokes. The word "toke" means 'capital' (as in accounting) when the o has
> two long strokes and means 'his balls' (in the rude sense) when it has the
> two dots on top.
This has come up before. In Hungarian a vowel with two dots is an
umlauted vowel. A vowel with one stroke (acute accent) is a long
vowel. A vowel with two strokes (acute accents) is a long umlauted
vowel.
--
Bob Underhill
runde...@mail.sdsu.edu