Browse Groups

• Melroch Aestan wrote: ... People might have, but Tolkien hasn t. This is my simplistic view on these matters. I don t care much for posttolkenian traditions
Message 1 of 25 , Apr 8, 2006
View Source
Melroch 'Aestan wrote:
...
> The problem with the W-tehta is that people have used
> it for _ö_ in the past, before // was invented...

People might have, but Tolkien hasn't. This is my simplistic view on
these matters. I don't care much for posttolkenian traditions in
Tolkien's work.

> I agree that your system makes more sense, with its
> systematic use of doubling to indicate "closer vowel",
> but isn't the symmetry broken if you use the doubled
> dot for /y/?

There are two kinds of symmetries: The roundedness symmetry between
unrounded and rounded front vowels; the vowel height symmetry between
close-mid and open-mid vowels. What we have attested is the use of
doubling for the roundedness symmetries (i = . y = ..). The problem
is that only one of these symmetries can be represented by simple
doubling, while the other has to be expressed by a more general way of
"modification".

My original system expresses the vowel height symmetry through
doubling and the roundedness symmetry through "modification". You're
right that nonetheless, it expresses one instance of roundedness
symmetry through doubling. However, I consider doubling just a special
case of "modification". A doubled dot is another dot modification. So
I can still consider the system to be consistent. Call this sophistry,
if you like; it's my way to save the 'modified left curl' for the use
of following _w_ sign.

But the expression of the roundedness symmetry through doubling has
the advantage that it fits better with Tolkien's attested tehtar for
_i_ and _y_. I mean, describing it as 'doubling' is more exact than
describing it as 'modification'. Neither description is wrong, but one
is more precise, so it fits better.

And now I see an obvious way to express the roundedness symmetry
through doubling but still preserve the following w tehta:

i = . y = .. u = )
e = / 2 = // o = (
E = \ 9 = \\ O = ((

It's the same trick again I used above to explain the i-y case: The
height symmetry is expressed through "modification", and by
modification I mean any kind of modification (preferrably doubling if
still available). Also, this very same system could be used for the
Swedish vowels:

i = . y = .. u = )
e = / ö = // o = (
å = ((

By the way, for my dialect I rather swap i tehta and e tehta, since it

i = \ y = \\ u = ))
I = / Y = // U = )
e = . 2 = .. o = (

It's time I get used to these new vowel systems!

> I would use / for /e/
> and \ for /E/ however -- as a mnemonic, since it is _é_
> that is /e/ and _è_ that is /E/ in French. Actually it
> would feel very strange to use your assignments when
> writing French,

You're certainly right! This shows I haven't used this system yet...
The reason why I chose / for _è_ was that I guess French è-grade
vowels occur more often than é-grade vowels and so I chose the easier
more important.

> As for the following-W-tehta I think it is often a bit
> of an embarrasment, since it gets messy when combining
> it with vowel tehtar an the nasal stroke.

I've gotten used to it, and now it's become an indispensable part of
my tengwar writing. It may get messy, but thats all attested.

> Also when
> writing French you would need a following-/H/-tehta in
> words like _nuit_: better to use Úre for /w/, Yanta
> or Anna for /j/ and Vala for /H/ across the line.

I express /H/ as if it were /wj/: Vala with two dots below in initial
position or both two dots below and the modified left curl above when
there's a preceding letter. Tehtar crowding, but it works fine.

> (Phonetic transcription again as per <http://www.theiling.de/ipa/>!)
> (BTW I wouldn't use Wilya for /w/ when using the Quessetéma
> for velars rather than for labiovelars. Wilya could OTOH
> be used for [@_^], [A_^] or /M\/ if Quessetéma is used for velars.

Or for [?].

> Also Tolkien obviously used this tehta for Old and
> Modern English /æ/ in addition to and quite possibly
> even before using it for Latin _ae_.

Did he even use it for Latin _ae_ at all? If I'm not mistaken, the
"Latin" word identified with the inscription of DTS 41 is _gladiolus_,
not *_glaediolus_. And I think it's misleading to consider the
spellings of that word to be representations of Latin sounds, since
it's rather representations of English sounds, even though the word
might be considered Latin.

---------------------------
j. 'mach' wust
http://machhezan.tripod.com
---------------------------
• ... I see your point, but while you can confuse people, you can t confuse Tolkien anymore. Besides *he* would probably just have invented new ad_hoc tehtar!
Message 1 of 25 , Apr 8, 2006
View Source
j_mach_wust skrev:
> Melroch 'Aestan wrote:
> ...
>
>>The problem with the W-tehta is that people have used
>>it for _ö_ in the past, before // was invented...
>
>
> People might have, but Tolkien hasn't. This is my simplistic view on
> these matters. I don't care much for posttolkenian traditions in
> Tolkien's work.

I see your point, but while you can confuse people,
you can't confuse Tolkien anymore. Besides *he*
would probably just have invented new ad_hoc tehtar!

[big snip]

> But the expression of the roundedness symmetry through doubling has
> the advantage that it fits better with Tolkien's attested tehtar for
> _i_ and _y_. I mean, describing it as 'doubling' is more exact than
> describing it as 'modification'. Neither description is wrong, but one
> is more precise, so it fits better.
>
> And now I see an obvious way to express the roundedness symmetry
> through doubling but still preserve the following w tehta:
>
> i = . y = .. u = )
> e = / 2 = // o = (
> E = \ 9 = \\ O = ((
>
> It's the same trick again I used above to explain the i-y case: The
> height symmetry is expressed through "modification", and by
> modification I mean any kind of modification (preferrably doubling if
> still available).

Since back vowels tend to get more rounded the
higher they are it might be a good idea to use
(( = /o/ and ( = /O/: doubling would then indicate
relatively greater rounding. That way you can still
-- tenuously -- preserve consistency in the use of
the doubling modification.

> Also, this very same system could be used for the
> Swedish vowels:
>
> i = . y = .. u = )
> e = / ö = // o = (
> å = ((

For the mnemonic reason i said before I'd prefer
(( for _o_: this letter usually has the same sound
as English _oo_; moreover /u/ is usually derived
from historical /o:/, although few beside me would
be sensitive to that consideration!

Also should it come to expressing my native dialect
there would be three extra vowel phonemes to consider.
I would express these like this:

i /i/ = . y /y/ = .. u /u\/ = (( o /u/ = ))
e /e/ = / ö /2/ = // û /8/ = ( å /o/ = )
ä /E/ = \
â /a/ = v /ô/ /9-/ = \\ a /A/ = ^

which I feel would be a more consistent use of doubling,
by the "more closed--more rounded" criterion.

(I'm anyway ignoring the fact that long /A/ is [Q:]!
Length is not phonemic in Swedish, the rule basically
being that a stressed vowel is long if not followed
by a consonant cluster or a geminate consonant. Still
people tend to hear phonetic vowel length better than
phonetic consonant length, so the question about length
is a bit vexed both in phonology and in Tengwar writing...)

> By the way, for my dialect I rather swap i tehta and e tehta, since it
>
> i = \ y = \\ u = ))
> I = / Y = // U = )
> e = . 2 = .. o = (

Yes, that makes sense, although AFAIK the use
of . for _i_ and / for _e_ is a CJRT usage,
which you maybe would consider to speak against it.

If having to distinguish two levels of i-vowels
I'd prefer using the caret tehta for /Y/ and double
it (vertically) for /y/ -- and of course . = /I/
and .. = /i/. There would be the slight problem
that no font as yet provides for a doubled caret
tehta, be it vertically or horizontally doubled!
(At least for me that's a problem: my hands are
lame and I write atrociously by hand...)

> It's time I get used to these new vowel systems!
>
>
>>I would use / for /e/
>>and \ for /E/ however -- as a mnemonic, since it is _é_
>>that is /e/ and _è_ that is /E/ in French. Actually it
>>would feel very strange to use your assignments when
>>writing French,
>
>
> You're certainly right! This shows I haven't used this system yet...
> The reason why I chose / for _è_ was that I guess French è-grade
> vowels occur more often than é-grade vowels and so I chose the easier
> signs for è-grade. But I'd say your practical "iconic" argument is
> more important.

Yes. If one would use \ for the vowel indicated
with acute accent in the Roman orthography one
would surely write it wrong more often than not!

For French I would personally prefer a system closer
to the Roman orthography, i.e. ( for _u_, (( for _ou_,
) for all of _o, ô, au, eau_, )) for both of _eu, êu_
/ for _é_ and \ for _è_, but . for _e_ when it is /E/
and . below when it is [@] or silent, and .. for _i, î_,
v for _ai, aî, ei_.
This is probably because I'm unfamiliar and uncomfortable
with spoken French and simply am not sure about the
distribution of /2/ vs. /9/ and /o/ vs. /O/.

>>As for the following-W-tehta I think it is often a bit
>>of an embarrasment, since it gets messy when combining
>>it with vowel tehtar an the nasal stroke.
>
>
> I've gotten used to it, and now it's become an indispensable part of
> my tengwar writing. It may get messy, but thats all attested.

Yes sure, but then in a language like Icelandic, where older
/Q/ has become /9/, you almost have to use the W-tehta for
that phoneme, if you are going to use the same mode for both
the old and the new language (which anyway is possible only
because the orthography is archaizing...)

>>Also when
>>writing French you would need a following-/H/-tehta in
>>words like _nuit_: better to use Úre for /w/, Yanta
>>or Anna for /j/ and Vala for /H/ across the line.
>
>
> I express /H/ as if it were /wj/: Vala with two dots below in initial
> position or both two dots below and the modified left curl above when
> there's a preceding letter. Tehtar crowding, but it works fine.

Oh I see; yes that works too! :-) Do you have any particular
reason for choosing the analysis /wj/ over /jw/? I guess it
would matter only word-initially. My hunch is that initial /j/
is much more common than initial /w/ in French (I can only really
think of _oui_) but I guess that whatever you use for initial /w/
with a .. below looks better than Anna with a W-tehta above.
What do you use for initial /w/ BTW. I guess Úre or Vala.

>>(Phonetic transcription again as per <http://www.theiling.de/ipa/>!)
>>(BTW I wouldn't use Wilya for /w/ when using the Quessetéma
>>for velars rather than for labiovelars. Wilya could OTOH
>>be used for [@_^], [A_^] or /M\/ if Quessetéma is used for velars.
>
>
> Or for [?].

Yes, though I would rather use Halla (single rising stem)
for /?/. Certainly in Arabic Halla for Hamza and
Wilya for `Ayn!

>>Also Tolkien obviously used this tehta for Old and
>>Modern English /æ/ in addition to and quite possibly
>>even before using it for Latin _ae_.
>
>
> Did he even use it for Latin _ae_ at all? If I'm not mistaken, the
> "Latin" word identified with the inscription of DTS 41 is _gladiolus_,
> not *_glaediolus_. And I think it's misleading to consider the
> spellings of that word to be representations of Latin sounds, since
> it's rather representations of English sounds, even though the word
> might be considered Latin.

I certainly consider _gladiolus_ to be "English" here.
In _daemonio_ in the same source _ae_ is curiously
Yanta with /, indicating an English pronunciation of
Latin, so yes he probably didn't use "v-dots" for
*Latin* _ae_ only for (Old) English _æ_ -- in which
case it is certainly appropriate for German, Swedish
and Finnish _ä_, and for Danish and Norwegian _æ_!

--

/BP 8^)>
--
Benct Philip Jonsson -- melroch at melroch dot se
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~__
A h-ammen ledin i phith! \ \
__ ____ ____ _____________ ____ __ __ __ / /
\ \/___ \\__ \ /___ _____/\ \\__ \\ \ \ \\ \ / /
/ / / / / \ / /Melroch\ \_/ // / / // / / /
/ /___/ /_ / /\ \ / /Roccondil\_ // /__/ // /__/ /
/_________//_/ \_\/ /Eowine __ / / \___/\_\\___/\_\
Gwaedhvenn Angeliniel\ \______/ /a/ /_h-adar Merthol naun
~~~~~~~~~Kuinondil~~~\________/~~\__/~~~Noolendur~~~~~~
|| Lenda lenda pellalenda pellatellenda kuivie aiya! ||
• ... We discussed this on Conlang list not so long ago, and concluded that any prefix or root which begins in a vowel in the orthography begins in a phonemic
Message 1 of 25 , Apr 8, 2006
View Source
hisilome skrev:

> [ Ah! I thought the glottal stop in German wasn't "phonemic" at all,
> but only occuring as an inserted element (in multi-morphemic words)
> before morphemes that begin with a vowel, as between _e_ and _a_ in
> _Beamter_. My point being, is it even necessary to represent this in
> writing at all? And--excuse me for asking what may very well be a
> stupid question--where does the glottal stop occur at the beginning
> of a word? Maybe in something like _Sieh an_? Just curious... ]

We discussed this on Conlang list not so long ago,
and concluded that any prefix or root which begins
in a vowel in the orthography begins in a phonemic
glottal stop -- even when preceded by a consonant,
thus _unabhängig_ is /'?un?aphENIC/, and _Verein_
is /fEr'?ain/ though actual realization in rapid
speach may vary. My German L1 mother anyway insisted
that _Verein_ be [fE6'?aen], not [fE'raen] as I
would have it!
(Transcription as usual CXS: <http://www.theiling.de/ipa/>)

As for Christian Thalmann's mode it is obviously
*very* orthographic. I like it for the most part.

--

/BP 8^)>
--
Benct Philip Jonsson -- melroch at melroch dot se
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~__
A h-ammen ledin i phith! \ \
__ ____ ____ _____________ ____ __ __ __ / /
\ \/___ \\__ \ /___ _____/\ \\__ \\ \ \ \\ \ / /
/ / / / / \ / /Melroch\ \_/ // / / // / / /
/ /___/ /_ / /\ \ / /Roccondil\_ // /__/ // /__/ /
/_________//_/ \_\/ /Eowine __ / / \___/\_\\___/\_\
Gwaedhvenn Angeliniel\ \______/ /a/ /_h-adar Merthol naun
~~~~~~~~~Kuinondil~~~\________/~~\__/~~~Noolendur~~~~~~
|| Lenda lenda pellalenda pellatellenda kuivie aiya! ||
• ... [ Well, first of all thanks to you (and Melroch!) for your detailed explanations on Latin _ae_ (and sound changes in the Latin vowel system in general).
Message 1 of 25 , Apr 8, 2006
View Source
--- In elfscript@yahoogroups.com, "j_mach_wust" <j_mach_wust@...>
wrote:
>

[ Well, first of all thanks to you (and Melroch!) for your detailed
explanations on Latin _ae_ (and sound changes in the Latin vowel
system in general).
Quite apart from that, there's so much in the discussion between you
guys in the last few posts that it'll take me a while to digest it
all. :)

But about the German mode: ]

> > Have you ever bothered to create an entire mode for High
> > ("Standard") German? There are at least one or two out there,
>like the one Danny mentioned, but I'd be curious to see yours
> > (if it exists).
>
> I sure have, and there used to be an online description, but not
>any more.
>
> Let's see... Phonemic mode vowels:
>
> i: . ü: .. u: )
> e: / ö: \ o: (
>
> ä: v a: ^

[ Exactly what I had expected after the last few posts! Except maybe
for _ö_, but you allow for the possibility of using the double acute
(in analogy to the spelling of _ü_) further down.

I'd certainly prefer to have distinct spellings for diphthongs (au,
ai/ei, eu/äu) and umlauts (ä, ö, ü). Christian Thalmann's mode, which
Danny mentioned, spells both kinds of sounds with combinations of
tehta plus uure/yanta, I guess in analogy to the spelling of Quenya
dipthongs... ]

> Vowel length is expressed by placing the vowels on a long carrier (I
> also use a long carrier in the case of ä, even though that's
>redundant).

[ That's because _ä_ is always long, I gather. ]

> The vowel tehtar are placed on the following consonants, but schwa
is expressed by a single dot under the preceding consonant since many
> words end with schwa. Consequently, syllabic /l m n r/ are spelled
> without preceding schwa unless it is clearly pronounced, in words
>like _kennen, stellen_ (of course, these are no syllabic consonants
>then).
>
> _ö_ might as well be represented with a double acute //.
>
> Dipththongs are represented with vala and anna combinations, of
course.
>
>
> Consonants:
> Plain general use with extended tyeller for affricates (though I'd
> admit I prefer calma for /tS/).

[ Ah, yes, as in the English modes. But then it wouldn't mesh so
nicely with the spellings of _pf_ and _z_ (/ts/), which I gather
would be extended parma/tinco in your mode, correct? And _tS_ would
in fact be extended calma (and, in the fricative tyelle, _sch_ (/S/)
harma and _ch_ (/x/) hwesta, I presume). ]

> Óre is used for vocalized _r_. Since I prefer a transcription where
> vocalisation of _r_ occurs only after long vowels (and of course in
> syllabic _r_), I don't mark vowel length before óre, so I transcribe
> _wir_ /vi:r/ [vi:6] with single dot on óre, but no long carrier, and
> _wirr_ /vIr/ [vIR] with single dot on rómen.

[ I think this roughly corresponds to Thalmann's mode, though he
doesn't think it absolutely necessary to make a distinction between
short and long vowels to begin with, if I understand correctly.
Otherwise, he also applies the "r-rule", i.e. vocalic _r_ as oore,
consonantal _r_ as roomen. He quotes the examples _Verein_/_bereit_.

For double _r_ (as in _wirr_) he proposes to use arda (and alda for
_ll_, aaze nuquerna for _ss_), but I don't really see the need for
these unattested (at least in the case of _rr, ll_) spellings. For
_ll_, for example, why not use the tilde inside lambe. ]

> Vilya is used for glottal stop, but I don't mark it at the beginning
> of a word. ... Now that I think about it, halla might also be
> considered for that purpose...

[ Ah! I thought the glottal stop in German wasn't "phonemic" at all,
but only occuring as an inserted element (in multi-morphemic words)
before morphemes that begin with a vowel, as between _e_ and _a_ in
_Beamter_. My point being, is it even necessary to represent this in
writing at all? And--excuse me for asking what may very well be a
stupid question--where does the glottal stop occur at the beginning
of a word? Maybe in something like _Sieh an_? Just curious... ]

> Orthographic mode:
> Mostly the same.
>
> I usually express both the doubling of vowel letters and the
> combination of vowel letter + 'lengthening _h_' by the long carrier,
> though the latter might as well be represented by plain hyarmen
> (indeed, this is more advantageous for words like _Rehe_).

[ Indeed! Though confusion probably wouldn't be possible in this
example, since I don't think there's a word called _ree_--but there
may be other instances where problems would arise? ]

> I've started to express _ie_ by single dot on stemless calma.
>
> The dot below is only used for final _e_.
>
> _y_ is represented by the breve accent.

[ This would be an orthographic spelling indeed, then, as in
Tolkien's orthographic English modes that also show this usage
(outside diphthongal combinations such as in _day_ which are spelled
with anna). I believe _y_ is quite a rare letter in German, but can
yet stand for a variety of sounds (_Ypsilon, Yak, Yvonne_). ]

> I prefer to distinguish _v_ and _w_, using ampa for the former, vala
> for the latter (or the sign for following _w_).

[ Again, orthographic indeed (and also based on English modes), since
German _w_ really corresponds to English _v_, while German _v_ is
either the same as English _v_ OR the same as _f_ (as in _Verein_),
while English _w_ doesn't even occur in German. ]

> Silme and esse are used for _s_ and _ss/ß_, not for /s/ and /z/.

[ Though it would be nice, at least in phonetic spelling, to
distinguish /s/ (_fest_) and /z/ (_Sand_), and in phonetic spelling
no extra sign would then be needed for _ss/ß_, since that's
always /s/, right? ;) ]

> Double consonant letters are expressed by the bar below, though I
> sometimes use alda for _ll_ (also because most computer fonts don't
> allow for a bar to be placed below alda).

[ You mean, they don't allow for a bar below lambe, is it? Hm, I
think that was also one of the reasons why Thalmann proposed alda for
double _l_.
I'm out of my depth here, but how come a frequently attested spelling
such as a tilde inside lambe (as it is, for example in Namaarie--I
don't think Tolkien himself ever actually put the tilde _under_
lambe) is not available in most computer fonts as you say?
Remarkable. ]

Hisilome
• ... I d rather agree with Hisilome on this point. German is, like English or Swedish, a language that does not have one single standard form and pronunciation,
Message 1 of 25 , Apr 9, 2006
View Source
Melroch 'Aestan wrote:
>
> hisilome skrev:
>
> > [ Ah! I thought the glottal stop in German wasn't "phonemic" at
> > all, but only occuring as an inserted element (in multi-morphemic
> > words) before morphemes that begin with a vowel, as between _e_
> > and _a_ in _Beamter_. My point being, is it even necessary to
> > represent this in writing at all? And--excuse me for asking what
> > may very well be a stupid question--where does the glottal stop
> > occur at the beginning of a word? Maybe in something like _Sieh
> > an_? Just curious... ]
>
> We discussed this on Conlang list not so long ago,
> and concluded that any prefix or root which begins
> in a vowel in the orthography begins in a phonemic
> glottal stop -- even when preceded by a consonant,
> thus _unabh�ngig_ is /'?un?aphENIC/, and _Verein_
> is /fEr'?ain/ though actual realization in rapid
> speach may vary. My German L1 mother anyway insisted
> that _Verein_ be [fE6'?aen], not [fE'raen] as I
> would have it!
> (Transcription as usual CXS: <http://www.theiling.de/ipa/>)

I'd rather agree with Hisilome on this point. German is, like English
or Swedish, a language that does not have one single standard form and
pronunciation, but different standards varying regionally (unlike
English or Swedish, this has only been recognized by linguistics very
recently, mainly over the past ten years). The glottal stop seems to
be typical for the standard German of Germany, while in Austria and
Switzerland it is only seldom realized and many speakers don't use it
at all (in standard German, that is, not in the dialect). Therefore,
I'd rather not consider it a phoneme.

---------------------------

Hisilome wrote:
> [ Exactly what I had expected after the last few posts! Except maybe
> for _ö_, but you allow for the possibility of using the double acute
> (in analogy to the spelling of _ü_) further down.

After the last posts with Benct, I'd also recommend the doubled acute,
not the grave accent.

> _ä_ is always long, I gather. ]

Phonemic _ä_ is; orthographic _ä_ may be long or short, but the short
_ä_ is homophonous with short _e_.

> and _ch_ (/x/) hwesta, I presume). ]

Certainly. I forgot to mention I sometimes use hyarmen in the phonemic
mode for palatal _ch_. I usually don't put the two dots below, since
it can hardly ever be confused with _h_ which occurs only initially.

> For double _r_ (as in _wirr_) he proposes to use arda (and alda for
> _ll_, aaze nuquerna for _ss_), but I don't really see the need for
> these unattested (at least in the case of _rr, ll_) spellings. For
> _ll_, for example, why not use the tilde inside lambe. ]

Alda is attested for _ll_, though only in full writing, but in
'general use' style full writing. I don't remember whether it was in
one of the King's Letters or in the Mazarbul inscriptions. Arda,
however, is not attested for _rr_.

> > I usually express both the doubling of vowel letters and the
> > combination of vowel letter + 'lengthening _h_' by the long
> > carrier, though the latter might as well be represented by plain
> > hyarmen (indeed, this is more advantageous for words like _Rehe_).
>
> [ Indeed! Though confusion probably wouldn't be possible in this
> example, since I don't think there's a word called _ree_--but there
> may be other instances where problems would arise? ]

It's not because of confusion that hyarmen would be more advantageous
in _Rehe_; just try to spell the _eh_ with a long carrier and then put
a dot for the final _-e_ below! An isolated short carrier with a dot
below would be possible, but I dislike it (and it is not attested). In
a phonemic mode, I sometimes use an ephentetic anna in such cases (or
vala after _u, o_), which is as far as I know a similar solution to be
found as well in Arabic orthography.

> [ Though it would be nice, at least in phonetic spelling, to
> distinguish /s/ (_fest_) and /z/ (_Sand_), and in phonetic spelling
> no extra sign would then be needed for _ss/ß_, since that's
> always /s/, right? ;) ]

Exactly, and so I do in phonemic modes (even though there are
varieties of standard German which in this respect are more similar to
the spelling and distinguish /s/ and /ss/, not /z/ and /s/).

> > Double consonant letters are expressed by the bar below, though I
> > sometimes use alda for _ll_ (also because most computer fonts
> > don't allow for a bar to be placed below alda).
>
> [ You mean, they don't allow for a bar below lambe, is it? Hm, I
> think that was also one of the reasons why Thalmann proposed alda
> for double _l_.
> I'm out of my depth here, but how come a frequently attested
> spelling such as a tilde inside lambe (as it is, for example in
> Namaarie--I don't think Tolkien himself ever actually put the tilde
> _under_ lambe) is not available in most computer fonts as you say?
> Remarkable. ]

I'm sorry; of course the computer fonts provide a special bar to be
placed inside lambe. They don't have a bar that could be placed below
alda, but that was totally besides the point, since the possible
spellings are either lambe with bar (widely attested) or alda
(scarcely attested).

---------------------------
j. 'mach' wust
http://machhezan.tripod.com
---------------------------
• ... That also sounds reasonable. ... That s a pretty system! What is /9-/? ... Is consonant gemination considered to be mere consonant length in Swedish
Message 1 of 25 , Apr 9, 2006
View Source
Melroch 'Aestan wrote:
> Since back vowels tend to get more rounded the
> higher they are it might be a good idea to use
> (( = /o/ and ( = /O/: doubling would then indicate
> relatively greater rounding. That way you can still
> -- tenuously -- preserve consistency in the use of
> the doubling modification.

That also sounds reasonable.

> Also should it come to expressing my native dialect
> there would be three extra vowel phonemes to consider.
> I would express these like this:
>
> i /i/ = . y /y/ = .. u /u\/ = (( o /u/ = ))
> e /e/ = / ö /2/ = // û /8/ = ( å /o/ = )
> ä /E/ = \
> â /a/ = v /ô/ /9-/ = \\ a /A/ = ^

That's a pretty system! What is /9-/?

> (I'm anyway ignoring the fact that long /A/ is [Q:]!
> Length is not phonemic in Swedish, the rule basically
> being that a stressed vowel is long if not followed
> by a consonant cluster or a geminate consonant. Still
> people tend to hear phonetic vowel length better than
> phonetic consonant length, so the question about length
> is a bit vexed both in phonology and in Tengwar writing...)

Is consonant gemination considered to be mere consonant length in
Swedish phonology? Traditionally, the Alemannic consonant length is
considered to be a secondary feature of the fortis-lenis opposition,
but it is debated whether there are other features to that opposition.

> AFAIK the use
> of . for _i_ and / for _e_ is a CJRT usage,

It's attested in DTS 10, after all one of the two major English tehtar
mode samples.

> If having to distinguish two levels of i-vowels
> I'd prefer using the caret tehta for /Y/ and double
> it (vertically) for /y/ -- and of course . = /I/
> and .. = /i/.

I prefer to keep the attested symmetry between unrounded front vowel =
. and corresponding rounded front vowel = ..

> in a language like Icelandic, where older
> /Q/ has become /9/, you almost have to use the W-tehta for
> that phoneme, if you are going to use the same mode for both
> the old and the new language (which anyway is possible only
> because the orthography is archaizing...)

Wouldn't the usual (and attested) distinction between a phonemic and
an orthographic mode work for Icelandic as well (the latter allowing a
fair representation of Old Icelandic) and make the use of the modified
left curl dispensable?

> Do you have any particular
> reason for choosing the analysis /wj/ over /jw/? I guess it
> would matter only word-initially. My hunch is that initial /j/
> is much more common than initial /w/ in French (I can only really
> think of _oui_) but I guess that whatever you use for initial /w/
> with a .. below looks better than Anna with a W-tehta above.

That's the reason.

> What do you use for initial /w/ BTW. I guess Úre or Vala.

Úre is not attested for initial /w/ in tehtar modes, but only as a
"reading direction inverter" (as in Quenya). So I'd certainly use vala.

---------------------------
j. 'mach' wust
http://machhezan.tripod.com
---------------------------
• ... Well, traditionally it s identified as an open rounded front vowel -- [9] or even [& ] -- but in my pronunciation it certainly is a low mid rounded
Message 1 of 25 , Apr 9, 2006
View Source
j_mach_wust skrev:

>>Also should it come to expressing my native dialect
>>there would be three extra vowel phonemes to consider.
>>I would express these like this:
>>
>>i /i/ = . y /y/ = .. u /u\/ = (( o /u/ = ))
>>e /e/ = / ö /2/ = // û /8/ = ( å /o/ = )
>>ä /E/ = \
>>â /a/ = v /ô/ /9-/ = \\ a /A/ = ^
>
>
> That's a pretty system! What is /9-/?

Well, traditionally it's identified as an
open rounded front vowel -- [9] or even [&\] --
but in my pronunciation it certainly is a
low mid rounded *central* vowel [3\], so
I notated it as a retracted front vowel.
I also put it one row too low for some reason...

>>(I'm anyway ignoring the fact that long /A/ is [Q:]!
>>Length is not phonemic in Swedish, the rule basically
>>being that a stressed vowel is long if not followed
>>by a consonant cluster or a geminate consonant. Still
>>people tend to hear phonetic vowel length better than
>>phonetic consonant length, so the question about length
>>is a bit vexed both in phonology and in Tengwar writing...)
>
>
> Is consonant gemination considered to be mere consonant length in
> Swedish phonology? Traditionally, the Alemannic consonant length is
> considered to be a secondary feature of the fortis-lenis opposition,
> but it is debated whether there are other features to that opposition.

Yes consonant gemination is just consonant length in Swedish.

>>AFAIK the use
>>of . for _i_ and / for _e_ is a CJRT usage,
>
>
> It's attested in DTS 10, after all one of the two major English tehtar
> mode samples.

Ah, OK.

>>in a language like Icelandic, where older
>>/Q/ has become /9/, you almost have to use the W-tehta for
>>that phoneme, if you are going to use the same mode for both
>>the old and the new language (which anyway is possible only
>>because the orthography is archaizing...)
>
>
> Wouldn't the usual (and attested) distinction between a phonemic and
> an orthographic mode work for Icelandic as well (the latter allowing a
> fair representation of Old Icelandic) and make the use of the modified
> left curl dispensable?

Sure, but the point is that you would want to be able
to use the same orthographic mode both for the old and
the new language. It would feel very weird otherwise.
Anyway my usage is to represent _v_ with Vala in all
positions. I guess anyone who wants can use \ for _ö_
and W-tehta for _v_ after consonants, but I don't.

>>Do you have any particular
>>reason for choosing the analysis /wj/ over /jw/? I guess it
>>would matter only word-initially. My hunch is that initial /j/
>>is much more common than initial /w/ in French (I can only really
>>think of _oui_) but I guess that whatever you use for initial /w/
>>with a .. below looks better than Anna with a W-tehta above.
>
>
> That's the reason.
>
>
>>What do you use for initial /w/ BTW. I guess Úre or Vala.
>
>
> Úre is not attested for initial /w/ in tehtar modes, but only as a
> "reading direction inverter" (as in Quenya). So I'd certainly use vala.
>

I thought so.

--

/BP 8^)>
--
Benct Philip Jonsson -- melroch at melroch dot se
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~__
A h-ammen ledin i phith! \ \
__ ____ ____ _____________ ____ __ __ __ / /
\ \/___ \\__ \ /___ _____/\ \\__ \\ \ \ \\ \ / /
/ / / / / \ / /Melroch\ \_/ // / / // / / /
/ /___/ /_ / /\ \ / /Roccondil\_ // /__/ // /__/ /
/_________//_/ \_\/ /Eowine __ / / \___/\_\\___/\_\
Gwaedhvenn Angeliniel\ \______/ /a/ /_h-adar Merthol naun
~~~~~~~~~Kuinondil~~~\________/~~\__/~~~Noolendur~~~~~~
|| Lenda lenda pellalenda pellatellenda kuivie aiya! ||
• ... [ I think your mother s realization corresponds more closely to Thalmann s, who also has the glottal stop in _Verein_. Thalmann says that _Verein_ is
Message 1 of 25 , Apr 9, 2006
View Source
Melroch 'Aestan wrote:

> > hisilome skrev:
> >
>Ah! I thought the glottal stop in German wasn't "phonemic" at
>all, but only occuring as an inserted element (in multi-morphemic
>words) before morphemes that begin with a vowel, as between _e_ and
>_a_ in _Beamter_. My point being, is it even necessary to
>represent this in writing at all?

>We discussed this on Conlang list not so long ago,
>and concluded that any prefix or root which begins
>in a vowel in the orthography begins in a phonemic
>glottal stop -- even when preceded by a consonant...
>My German L1 mother anyway insisted that _Verein_ be [fE6'?aen],
>not [fE'raen] as I would have it!

and j_mach_wust replied:

>The glottal stop seems to be typical for the standard German of
>Germany, while in Austria and Switzerland it is only seldom realized
>and many speakers don't use it at all (in standard German, that is,
>not in the dialect). Therefore, I'd rather not consider it a phoneme.

[ I think your mother's realization corresponds more closely to
Thalmann's, who also has the glottal stop in _Verein_. Thalmann says
that _Verein_ is _Ver-ein_, but _bereit_ is _be-reit_, i.e. in the
latter case the _r_ is syllable-initial and thus pronounced
differently.
While I think that _bereit_ is practically never realized as _ber-
eit_, in the case of _Verein_, the realization _Ve-rein_ is also
possible (and I think this would correspond to what 'Mach' means when
he says that in some variants of High [Standard] German [and I
believe not just in Austria and Switzerland, but also in some
regions/speakers in Germany, espcially in the south] the glottal stop
would not be realized).
Thalmann doesn't introduce a separate sign for the glottal stop.
Where _r_ is involved, he simply uses oore for syllable-final _r_
(regardless of whether it's followed by a vowel/glottal stop or a
consonant), and roomen for syllable-initial/medial _r_, which is
always pre-vocalic (syllable-medial for example in _Christian_). ]

Melroch 'Aestan wrote:

>As for Christian Thalmann's mode it is obviously
>*very* orthographic. I like it for the most part.

[ Yes, I agree it is! And I don't think there's a problem with that
(as you may remember, I'm actually quite partial to orthographic
modes, no matter what the language). The only thing I'm really not so
happy with (as I said) is that this mode uses basically the same
method for representing the diphthongs and the umlauts. I think since
umlauts are simple vowels, they should be spelled as such (and not
like diphthongs) even in orthographic writing, which is why I prefer
the solutions with inverted A-tehta/double acute/double over-dot for
_ä, ö, ü_. ]

----------------------------------------------------------------------

j_mach_wust wrote:

> > and _ch_ (/x/) hwesta, I presume).
>
>Certainly. I forgot to mention I sometimes use hyarmen in the
>phonemic mode for palatal _ch_. I usually don't put the two dots
>below, since it can hardly ever be confused with _h_ which occurs
>only initially.

[ Hm. How about words like _gehabt, gehoben, verhören_, where the _h_
is certainly pronounced (and not a sign of vowel lengthening)? But
even then, confusion would probably rarely arise... ]

> Alda is attested for _ll_, though only in full writing, but in
> 'general use' style full writing. I don't remember whether it was in
> one of the King's Letters or in the Mazarbul inscriptions.

[ Ah, you're of course right: both DTS 13 (Mazarbul) in _Dimrill,
shall_ (and, mistakenly, in the first occurrence of _hold_) and all
three copies of the King's Letter have it (in _all_). Hm, so only
arda for _rr_ is unattested...could warm up to this. Although in
your proposal one wouldn't need it, since _rr_ occurs only after
short vowels (=not vocalized) and is then spelled with roomen (as
opposed to oore) anyway. ]

> > > I usually express both the doubling of vowel letters and the
> > > combination of vowel letter + 'lengthening _h_' by the long
> > > carrier, though the latter might as well be represented by plain
> > > hyarmen (indeed, this is more advantageous for words like
_Rehe_).
> >
> > Indeed! Though confusion probably wouldn't be possible in this
> > example, since I don't think there's a word called _ree_...

>It's not because of confusion that hyarmen would be more
>advantageous in _Rehe_; just try to spell the _eh_ with a long
>carrier and then put a dot for the final _-e_ below! An isolated
>short carrier with a dot below would be possible, but I dislike it
(and it is not attested). In a phonemic mode, I sometimes use an
>ephentetic anna in such cases (or vala after _u, o_), which is as
>far as I know a similar solution to be found as well in Arabic
>orthography.

[ So, anna to lengthen _a, e, i_, and vala to lengthen _u, o_, do I
understand you correctly here? Is that attested somewhere in
Tolkien's modes (don't know any Arabic I'm afraid :()? Also, why call
it "epenthetic"? Isn't that sort of an "added" sound in spoken
language, while here we're dealing with an added _letter_ to express
vowel length in writing? I'm no expert, mind you, but curious. ]

> > Though it would be nice, at least in phonetic spelling, to
> > distinguish /s/ (_fest_) and /z/ (_Sand_), and in phonetic
spelling no extra sign would then be needed for _ss/ß_, since that's
> > always /s/, right?
>
>Exactly, and so I do in phonemic modes (even though there are
>varieties of standard German which in this respect are more similar
>to the spelling and distinguish /s/ and /ss/, not /z/ and /s/).

[ That's interesting! Even though this is drifting OT, I'd be curious
to know which varieties these are (geographically speaking)? And what
would this mean concretely: that one would pronounce the _s_ in
_Sand_ and _fest_ identically (and this would have to be an unvoiced
_s_ in both cases, right?), while discerning, say, _dass_
(conjunction, i.e. _daß_) and _das_ (article) in pronuncation? Would
the article then have a voiced _s_ (/z/)...? Sorry if I'm a bit slow
here. :) ]

Hisilome
• ... Especially since even a basic knowledge of German will be sufficient to recognize that _ge_ and _ver_ are typical prefixes. I d also spell _ge_ with a dot
Message 1 of 25 , Apr 9, 2006
View Source
Hisilome wrote:
> >I sometimes use hyarmen in the
> >phonemic mode for palatal _ch_. I usually don't put the two dots
> >below, since it can hardly ever be confused with _h_ which occurs
> >only initially.
>
> [ Hm. How about words like _gehabt, gehoben, verhï¿½ren_, where the
> _h_ is certainly pronounced (and not a sign of vowel lengthening)?
> But even then, confusion would probably rarely arise... ]

Especially since even a basic knowledge of German will be sufficient
to recognize that _ge_ and _ver_ are typical prefixes. I'd also spell
_ge_ with a dot below ungwe and _ver_ with formen + Ã³re, so it'd be
even more obvious that it's prefixes.

> so only
> arda for _rr_ is unattested...could warm up to this. Although in
> your proposal one wouldn't need it, since _rr_ occurs only after
> short vowels (=not vocalized) and is then spelled with roomen (as
> opposed to oore) anyway. ]

In the orthographic proposal of mine, however, a doubled _rr_ would be
necessary, but I haven't ever thought much about it. I think I
wouldn't use arda because it isn't attested at all. Maybe I'd use Ã³re
with a bar below, since a bar below rÃ³men would be awkward to me and
since a doubled _rr_ in one way is always at the end of a syllable
(but also at the beginning of the next syllable).

> [ So, anna to lengthen _a, e, i_, and vala to lengthen _u, o_, do I
> understand you correctly here? Is that attested somewhere in
> Tolkien's modes (don't know any Arabic I'm afraid :()? Also, why
> call it "epenthetic"? Isn't that sort of an "added" sound in spoken
> language, while here we're dealing with an added _letter_ to express
> vowel length in writing? I'm no expert, mind you, but curious. ]

I wouldn't use anna and vala to lengthen the vowels. I'd lengthen them
in the normal way with the long carrier. I'd only use vala/anna as a
kind of "carrier" for a following schwa, because I'd rather not write
a short carrier with a dot below. And I've called it "epenthetic"
because the reason why I'd use these letters is the assumption that
_Rehe_ /re:@/ is really pronounced [re:j@], with a epenthetic [j] in
the hiatus.

> >(even though there are
> >varieties of standard German which in this respect are more similar
> >to the spelling and distinguish /s/ and /ss/, not /z/ and /s/).
>
> [ That's interesting! Even though this is drifting OT, I'd be
> curious to know which varieties these are (geographically speaking)?
> And what would this mean concretely: that one would pronounce the
> _s_ in _Sand_ and _fest_ identically (and this would have to be an
> unvoiced _s_ in both cases, right?), while discerning, say, _dass_
> (conjunction, i.e. _daï¿½_) and _das_ (article) in pronuncation? Would
> the article then have a voiced _s_ (/z/)...? Sorry if I'm a bit slow
> here. :) ]

At least all of Austro-Bavarian and Alemannic-Swabian dialects are
said not to have any voiced obstruents at all (except for /v/, but
that's a special case), and neither have the varieties of standard
German in the same regions, that is to say, in much of Southern
Germany and in all of Switzerland and Austria (except for a border
region with Slovenia).

Since most of these regions, as far as I know, have terminal devoicing
(note that "devoicing" must not be understood literally in this case,
since there is no voice in the first place, but the term is still
used; it means just that the opposition is neutralized at the ends of
words), the opposition between _s_ and _ss_ exists only within a word
between voiced sounds, for instance in _reisen_ vs. _reissen_.

It is debated what makes the difference. The traditional point of view
is that the main feature of that opposition is a fortis-lenis
distinction, though others say the main feature is a length
distinction. If I remember correctly, the length difference can be
measured, but the force difference can't, as far as I know. The
traditional way of transcribing the opposition is with [s] vs. [z_0]
(voiceless [z]). This may seem strange to someone who supposes that
it's the voice that differentiates [s] from [z], and that a voiceless
[z] would be the same as an [s]; in the traditional use in German
linguistics, however, there's more to that differentiation than just
voice, but as I said, the nature of that "more" is debated.

I don't know whether the initial opposition heard in Germany between
_Sex_ /sEks/ 'sex' (an English loanword) and _sechs_ /zEks/ 'six' is
also made in areas that don't have voiced [z]. I'd say it would be
perfectly possible, but I could also figure that this initial
distinction is only made in areas that have voiced [z] since in
Switzerland, at least, it is not made. At least in Switzerland, on the
other hand, there is no terminal devoicing, so pairs like _Reis_
'rice' and _reiss_ 'rip (imperative singular)' are not homophonous.
Nonetheless, 'das' and 'dass' are perfectly homophonous, just like
they are etymologically identical. Their differenciation is just
made-up (surely an invention of cruel teachers to annoy their students).

---------------------------
j. 'mach' wust
http://machhezan.tripod.com
---------------------------
Your message has been successfully submitted and would be delivered to recipients shortly.
• Changes have not been saved
Press OK to abandon changes or Cancel to continue editing
• Your browser is not supported
Kindly note that Groups does not support 7.0 or earlier versions of Internet Explorer. We recommend upgrading to the latest Internet Explorer, Google Chrome, or Firefox. If you are using IE 9 or later, make sure you turn off Compatibility View.