There are 17 messages in this issue.
Topics in this digest:
1.1. Re: Destroying the noun/verb distinction
From: Sai
2.1. Re: another conlang promoted to natlanghood: Denden
From: Sai
2.2. Re: another conlang promoted to natlanghood: Denden
From: Daniel Bowman
3.1. Re: Fith Texts
From: R A Brown
4a. Monster sentences, was: Re: [CONLANG] Fith Texts
From: Jim Henry
4b. Re: Monster sentences, was: Re: [CONLANG] Fith Texts
From: neo gu
4c. Re: Monster sentences, was: Re: [CONLANG] Fith Texts
From: Adam Walker
4d. Re: Monster sentences, was: Re: [CONLANG] Fith Texts
From: Nikolay Ivankov
4e. Re: Monster sentences, was: Re: [CONLANG] Fith Texts
From: David Peterson
5a. Re: OT: Any romantic arabists out there?
From: Jesse Bangs
6a. Re: Indexed Predicates Sketch
From: neo gu
7a. Aeu phonology
From: Adam Walker
7b. Re: Aeu phonology
From: MorphemeAddict
7c. Re: Aeu phonology
From: Adam Walker
7d. Re: Aeu phonology
From: Dirk Elzinga
7e. Re: Aeu phonology
From: Adam Walker
7f. Re: Aeu phonology
From: Nikolay Ivankov
Messages
________________________________________________________________________
1.1. Re: Destroying the noun/verb distinction
Posted by: "Sai" [email protected]
Date: Tue Apr 24, 2012 7:09 am ((PDT))
On Thu, Apr 19, 2012 at 21:48, Puey McCleary <[email protected]> wrote:
> Why would one create a language that combines nouns and verbs (in some
> manner)?
…
> Does one create noun-verb languages for an aesthetic reason? As a
> challenge? Is it really somehow more logical? Or more alien?
Speaking about UNLWS just for myself (I expect that Alex, as the less
engelangerish and more mathyish aspect, will have other views): I'd
say rather, why would one make a language that *distinguishes* things
into "noun" and "verb"?
UNLWS is made a priori. We have certain principles we adhere to — one
of the foremost being fidelity to and maximal exploitation of the
medium — but something like treatment of word classes is not among
them. It simply was never useful to invoke this kind of distinction,
so we never did.
I'd add too that calling UNLWS predicate-oriented is a kind of
post-hoc linguist's descriptivism; we didn't have that as a goal
either, it simply fell out of what we thought was a natural way of
handling the interconnectivity of symbols. I suppose one *could* force
an analysis of UNLWS as being a verb-and-noun language, by calling
symbols "verbs" and binding points "nouns", though that strikes me as
a somewhat perverse over-clinging to theory.
FWIW, UNLWS itself is in part of my own lack of adequate imagination.
It's what I call a "node-and-connection" paradigm 2d language, as
opposed to the "massively fusional" paradigm (as described in the
hypothetical in e.g. _Story of Your Life_; cf
http://saizai.livejournal.com/657391.html). A N&C paradigm has clearly
distinguishable constituents, such that one can even (if with some
amount of coercion) analyze it in terms of "words"; a MF paradigm
language would lack that kind of separation altogether. I find the
latter possibility rather more intriguing, but in the six or so years
of thinking about it so far, I've yet to quite wrap my mind around how
I might implement it — though I retain the sense that this is a
limitation not of principle but only of my imagination.
Would a MF-paradigm language be made "to combine nouns and verbs", or
even "to remove the segregability of words"? No, it simply would never
be needed.
So I'd turn the question around: setting aside mere habituation (like
we try to do with so many other things that are unconscious L1
calquing if done naïvely without intent, like nom/acc systems), why
choose to *use* this distinction?
On Sat, Apr 21, 2012 at 15:02, Jörg Rhiemeier <[email protected]> wrote:
> A predicate is a function that takes zero or more arguments and
> outputs a truth value (which is equivalent to cutting a set out
> of a universe). Examples of predicates are:
>
> "X gives Y to Z"
…
> These can be written as:
…
> give (X,Y,Z)
…
Just as an aside: I've always hated the latter form. It's horribly
ambiguous and L1-presumptive. It's like a _quidquid Latinum_ for code,
making it look more logic-y while in fact making it *less* clearly
precise. Argh.
(At the risk of invoking NCNC, I find the same true of programming
language argument structures, whose orders are also often not clear at
all without explicitly referring to the documentation, and bad coders
[or PHP coders /oblig] often presume otherwise, to the detriment of
legibility. Object-oriented languages, eg Ruby, improve this somewhat
by letting functions have an implicit self, thus making at least
up-to-two-argument functions pretty clearly defined — and just as a
matter of culture, they tend to switch to using hashes with named
arguments when linear arguments become unwieldy, which is akin to
switching from a syntax-based to theta-role-marking system.)
On Tue, Apr 24, 2012 at 00:05, Alex Fink <[email protected]> wrote:
> OK. In UNLWS, where all open-class words are at least relations, we do have
> a fundamental mechanism of that sort. We regard it as a mood system: realis
> relations are predicated, irrealis relations are not.
Nor are things that are partial-parse quoted; nor are things that are
deep inside an irrealis structure… I think at some point, especially
for the latter, it's more licit to interpret it as some sort of
logophor-type multiple-realities system where things are evaluated
within the context of their local reality but that reality might not
be the same one as other parts of the whole.
- Sai
Messages in this topic (33)
________________________________________________________________________
________________________________________________________________________
2.1. Re: another conlang promoted to natlanghood: Denden
Posted by: "Sai" [email protected]
Date: Tue Apr 24, 2012 7:21 am ((PDT))
As a hypothetical for amusement:
Supposing that more linguistics were done confusing conlangs for
natlangs, how might linguistic theories change as a result? (Pick
confused-as-nat conlangs as you prefer to make for an amusing
discussion of the results.)
- Sai
Messages in this topic (41)
________________________________________________________________________
2.2. Re: another conlang promoted to natlanghood: Denden
Posted by: "Daniel Bowman" [email protected]
Date: Tue Apr 24, 2012 7:32 am ((PDT))
Angosey: evidence has been found for the interaction of Bantu and Korean
speakers during the colonial period of the British Empire. Further
evidence has been found for the introduction of click consonants when
previously these consonants were thought to be quite rare and invariant.
Possible admixtures of Basque are also proposed.
...any linguistic or historical theories using my conlang Angosey as
evidence would be a complete and utter disaster for the researchers.
2012/4/24 Sai <[email protected]>
> As a hypothetical for amusement:
>
> Supposing that more linguistics were done confusing conlangs for
> natlangs, how might linguistic theories change as a result? (Pick
> confused-as-nat conlangs as you prefer to make for an amusing
> discussion of the results.)
>
> - Sai
>
Messages in this topic (41)
________________________________________________________________________
________________________________________________________________________
3.1. Re: Fith Texts
Posted by: "R A Brown" [email protected]
Date: Tue Apr 24, 2012 8:53 am ((PDT))
On 24/04/2012 10:46, And Rosta wrote:
[snip]
>
> R A Brown, On 24/04/2012 07:08:
>> But Logan's email now makes me wonder if Jeffrey's aim
>> was not so much what I have detailed above, but rather
>> to produce a speakable version of FORTH!. If this is
>> so, then And is right after all that the 'stack
>> conjunctions' are an essential part of the language and
>> not incidental.
>
> I had only said that it was only the stack conjunctions
> that made Fith weird/counternatural, not that they were
> an essential part of the language.
Sorry - misquoted. You are right.
> Now that my understanding of Fith has changed (thanks to
> you), they seem even more incidental.
I think they probably are or would have been modified if
Jeffrey had seriously developed the language.
> Even if Jeffrey's primary aim was to make something
> FORTH-like, I think the thing that makes Fith unlike
> (human) language is that the parser operates on
> semantically-evaluated items.
Yep - on that we seem both agreed ;)
Now to get back to my other projects .....
--
Ray
==================================
http://www.carolandray.plus.com
==================================
Nid rhy hen neb i ddysgu.
There's none too old to learn.
[WELSH PROVERB]
Messages in this topic (95)
________________________________________________________________________
________________________________________________________________________
4a. Monster sentences, was: Re: [CONLANG] Fith Texts
Posted by: "Jim Henry" [email protected]
Date: Tue Apr 24, 2012 10:58 am ((PDT))
On 4/23/12, R A Brown <[email protected]> wrote:
> I recall many years ago when I was researching
> for my M.Litt degree, coming across a sentence in a German
> book that went on and on and on. There was - and I don't
> exaggerate - a full half-page of text before I met the
> full-stop (period). I was completely lost and the only way
> I could make sense of the thing was to take the long period
> apart and construct what, in effect, was a rough & ready
> parse tree.
I ran across a few page-long sentences in the work of early conlanger
Sir Thomas Urquhart; I'm not sure offhand if there are any extremely
long and complex sentences in his Logopantecteision, available for
free online. The ones I specially remember are in Ekskybalauron,
which overlaps with a lot of the conlangy content of Logopantecteision
but differs in the autobiographical and historical anecdotes. This is
the longest of the ones I quoted in my review of Ekskybalauron a few
years ago; there are much longer ones where that came from:
>>
Then is it, when having couched an Alphabet materiative of all the
words the mouth of man, with its whole implements, is able to
pronounce, and bringing all these words within the systeme of a
Language, which, by reason of its logopandochie, may deserved be
intituled, The Universal Tongue, that nothing will better merit the
labour of a Grammatical Arithmetician, then, after due enumeration,
hinc inde, to appariate the words of the Universall Language with the
things of the Universe.
<<
St. Paul's letters have some monster sentences in them, which tend to
be broken up into several shorter sentences in English translation.
Sometime in the next few days I'll try to find some examples, and
maybe try to translate one of them into gzb as single long sentences.
I remember finding some fairly long sentences in the short stories of
Lord Dunsany, back when I was transcribing them as etexts.
Steven Brust writing as Paarfi of Roundwood tends to use long, complex
sentences in the narration, and less often in the dialogue; Sylvia
Sotomayor posted a fairly tangled sentence from _The Phoenix Guards_
here a couple of years ago as a translation challenge, and I
translated it into gjâ-zym-byn.
ObConlang: have you any monster sentences in your conlang corpus,
either translations or original writing? Are monster sentences
admired or excoriated by the literary stylists of your conculture?
--
Jim Henry
http://www.pobox.com/~jimhenry/
Messages in this topic (5)
________________________________________________________________________
4b. Re: Monster sentences, was: Re: [CONLANG] Fith Texts
Posted by: "neo gu" [email protected]
Date: Tue Apr 24, 2012 11:31 am ((PDT))
On Tue, 24 Apr 2012 13:58:35 -0400, Jim Henry <[email protected]> wrote:
>On 4/23/12, R A Brown <[email protected]> wrote:
>> I recall many years ago when I was researching
>> for my M.Litt degree, coming across a sentence in a German
>> book that went on and on and on. There was - and I don't
>> exaggerate - a full half-page of text before I met the
>> full-stop (period). I was completely lost and the only way
>> I could make sense of the thing was to take the long period
>> apart and construct what, in effect, was a rough & ready
>> parse tree.
>
>I ran across a few page-long sentences in the work of early conlanger
>Sir Thomas Urquhart; I'm not sure offhand if there are any extremely
>long and complex sentences in his Logopantecteision, available for
>free online. The ones I specially remember are in Ekskybalauron,
>which overlaps with a lot of the conlangy content of Logopantecteision
>but differs in the autobiographical and historical anecdotes. This is
>the longest of the ones I quoted in my review of Ekskybalauron a few
>years ago; there are much longer ones where that came from:
>
>
>
>ObConlang: have you any monster sentences in your conlang corpus,
>either translations or original writing? Are monster sentences
>admired or excoriated by the literary stylists of your conculture?
>
>--
>Jim Henry
>http://www.pobox.com/~jimhenry/
If they were speaking Apr20, they might run out of indexes!
--
neogu
Messages in this topic (5)
________________________________________________________________________
4c. Re: Monster sentences, was: Re: [CONLANG] Fith Texts
Posted by: "Adam Walker" [email protected]
Date: Tue Apr 24, 2012 12:02 pm ((PDT))
On Tue, Apr 24, 2012 at 12:58 PM, Jim Henry <[email protected]> wrote:
>
> ObConlang: have you any monster sentences in your conlang corpus,
> either translations or original writing? Are monster sentences
> admired or excoriated by the literary stylists of your conculture?
>
> --
> Jim Henry
> http://www.pobox.com/~jimhenry/
>
Paul does, indeed, have some monstrously long sentences. Jane Austin can
provide some examples as can Hawthorne, and I seem to recall coming across
a monster or two in Dickens. I once wrote an email on another list which
included the following overlong sentences. (I have edited the text for
typos, and changed the names to protect the innocent.)
For the original germ of his wierdness, which runs as
true and as deep as does any river of wild and
mountainous country, rushing savagely against the rock
which it scores deeper and deeper year after year or
the vein of ore bearing precious metal deep into the
bowels of the earth as temptation to the prospector
leading him to his death therein, he can lay no charge
at our feet, for that germ of wierdness is his native
inheritance, the gift of his genes and his Lord. But,
though we hold no blame for that germ, for its
full-flowering, its over-respelndant shining forth in
such splendour as would dazzle the minds of lesser
beings, for this we must, through the long days of our
association with him and the effusive encouragement of
his eccentricities through both our words and our
deeds, truly, in some small part, bare a blame much like
that which he has laid at our feet.
In this I feel no great remorse, nor do I suspect
will Sue, for we both are likewise odd of spirit and
eccentric of character, and, having seen in young
Bob one of like bent, have done our all to grace the
world with yet another after our own sort, or at the
very least have offerd him no inducements to garb
himself with conformity and having so done to plot his
course the more smoothly through the doldrums of
American culture as it is; for in this one, I, at
least, have hope of changes being wroght upon this
dreary and conformist world from which the world
might, if God be gracious, never recover its former
indolent state.
Senetences such as these are both a joy and a horror to come across.
I don't believe anything resembling this yet exists in the C-a corpus, but
once I get around to translating a bit more of Paul, I'm sure some of his
monsters will survive in tact enough to frighten small children.
Adam
Messages in this topic (5)
________________________________________________________________________
4d. Re: Monster sentences, was: Re: [CONLANG] Fith Texts
Posted by: "Nikolay Ivankov" [email protected]
Date: Tue Apr 24, 2012 11:52 pm ((PDT))
Speaking of which, Tolstoy is known as the one composing the longest
sentences in Russian literature. I have no idea about English translation,
but 2/3 page long sentences are common in War and Peace, especially when he
begins with his philosophical stuff there. Really made my some of my school
friends start experimenting with monster-sentences in Russian. The Chekhov
words: "Brevity is a sister of the talent." - were in part a trolling of
Tolstoy.
On Tue, Apr 24, 2012 at 7:58 PM, Jim Henry <[email protected]> wrote:
> On 4/23/12, R A Brown <[email protected]> wrote:
> > I recall many years ago when I was researching
> > for my M.Litt degree, coming across a sentence in a German
> > book that went on and on and on. There was - and I don't
> > exaggerate - a full half-page of text before I met the
> > full-stop (period). I was completely lost and the only way
> > I could make sense of the thing was to take the long period
> > apart and construct what, in effect, was a rough & ready
> > parse tree.
>
> I ran across a few page-long sentences in the work of early conlanger
> Sir Thomas Urquhart; I'm not sure offhand if there are any extremely
> long and complex sentences in his Logopantecteision, available for
> free online. The ones I specially remember are in Ekskybalauron,
> which overlaps with a lot of the conlangy content of Logopantecteision
> but differs in the autobiographical and historical anecdotes. This is
> the longest of the ones I quoted in my review of Ekskybalauron a few
> years ago; there are much longer ones where that came from:
>
> >>
> Then is it, when having couched an Alphabet materiative of all the
> words the mouth of man, with its whole implements, is able to
> pronounce, and bringing all these words within the systeme of a
> Language, which, by reason of its logopandochie, may deserved be
> intituled, The Universal Tongue, that nothing will better merit the
> labour of a Grammatical Arithmetician, then, after due enumeration,
> hinc inde, to appariate the words of the Universall Language with the
> things of the Universe.
> <<
>
> St. Paul's letters have some monster sentences in them, which tend to
> be broken up into several shorter sentences in English translation.
> Sometime in the next few days I'll try to find some examples, and
> maybe try to translate one of them into gzb as single long sentences.
>
> I remember finding some fairly long sentences in the short stories of
> Lord Dunsany, back when I was transcribing them as etexts.
>
> Steven Brust writing as Paarfi of Roundwood tends to use long, complex
> sentences in the narration, and less often in the dialogue; Sylvia
> Sotomayor posted a fairly tangled sentence from _The Phoenix Guards_
> here a couple of years ago as a translation challenge, and I
> translated it into gjâ-zym-byn.
>
> ObConlang: have you any monster sentences in your conlang corpus,
> either translations or original writing? Are monster sentences
> admired or excoriated by the literary stylists of your conculture?
>
> --
> Jim Henry
> http://www.pobox.com/~jimhenry/
Messages in this topic (5)
________________________________________________________________________
4e. Re: Monster sentences, was: Re: [CONLANG] Fith Texts
Posted by: "David Peterson" [email protected]
Date: Wed Apr 25, 2012 12:02 am ((PDT))
On Apr 24, 2012, at 11:52 PM, Nikolay Ivankov wrote:
> Speaking of which, Tolstoy is known as the one composing the longest
> sentences in Russian literature. I have no idea about English translation,
> but 2/3 page long sentences are common in War and Peace, especially when he
> begins with his philosophical stuff there. Really made my some of my school
> friends start experimenting with monster-sentences in Russian. The Chekhov
> words: "Brevity is a sister of the talent." - were in part a trolling of
> Tolstoy.
One of the longer sentences in English before people started catching on and
doing silly things (a million word "novels" consisting of the word "the"
repeated a million times, etc.) was in Faulkner's Absalom, Absalom! The
sentence is, I believe, 26 pages long and straddles two chapters.
David Peterson
LCS President
[email protected]
www.conlang.org
Messages in this topic (5)
________________________________________________________________________
________________________________________________________________________
5a. Re: OT: Any romantic arabists out there?
Posted by: "Jesse Bangs" [email protected]
Date: Tue Apr 24, 2012 11:03 am ((PDT))
>
>
> Arabic, however, is in a different situation. Because Islam is so
> influential, all Arabic speakers at least have an awareness of Qur'anaic/
> Classical Arabic. The péóple, the masses, also have knowledge of MSA
> because of the News, schooling, the Internet and the international nature
> of modern society. The Modern Arabic dialects do have innovations, I'm not
> denying this, but due to the unifying qualities of Qur'anaic (Classical)
> Arabic and MSA - the Arabic dialects' divergence will, at the very least,
> be much slower than that of the Romance languages.
>
>
Is this really so different from Latin, though? The average European
peasant certainly couldn't read or write Latin in the Middle Ages, but I
expect that their passive understanding was more than nothing. Especially
since many prayers and liturgical rites were memorized in Latin... which
again is comparable to Classical Arabic and the modern dialects.
For an additional point of comparison, how much passive comprehension does
a modern Russian or Bulgarian have of Church Slavonic? And how does this
intelligibility relate to the mutual intelligibility of Russian and
Bulgarian?
(I don't know. I'm actually asking.)
>
> 2012/4/22 Peter Collier <[email protected]>
>
> > With Arabic today we have a dachsprache, MSA, and a continuum of
> > not-universally-intelligible regional dialects. If we were going to
> compare
> > this current diversity in Arabic to the historical development of the
> > Romance languages in Europe; whenabouts, roughly speaking, might the
> Arabic
> > of today be placed on a timeline of the Romance languages?
> >
> >
> >
> > How conservative/innovative are the Arabic dialects? If we projected the
> > Arabic dialects forward the same distance in time (i.e. gave them the
> same
> > amount of time to develop and diverge as the Romance languages have had
> up
> > to know), then all other things being equal might we expect to see a
> > similar
> > degree of separateness, or more, or less?
> >
> >
> >
> > P.
> >
>
--
JS Bangs
[email protected]
http://jsbangs.wordpress.com
"Be kind, for everyone you meet is fighting a great battle" -Philo of
Alexandria
Messages in this topic (15)
________________________________________________________________________
________________________________________________________________________
6a. Re: Indexed Predicates Sketch
Posted by: "neo gu" [email protected]
Date: Tue Apr 24, 2012 3:43 pm ((PDT))
I'm not sure yet if I'll actually use it but this is how "action
nominalization" should work.
Action nominalization makes a verb take on noun characteristics, to the extent
possible. In Apr20, such nominalization can be applied to any predicate. The
S-arg becomes a way to reference the situation or event or characteristic
designated. The P-arg retains one of the verb's original argument, interpreted
as a possessor. Because there are 2 original arguments, there are 2 action
nominalizers: PIP and PIA. PIP stands for "Possessor Is Patient" (where
patient really means original P-arg) and PIA stands for "Possessor Is Agent"
(where agent really means original S-arg). If the word is univalent, only PIA
can be used.
Some examples (ignoring tense marking*):
woman-i Dj-saw-Di town-k Dk-destroy-PIP-j.
"The woman saw the town's destruction."
Uj-like-1 Fluffy-i Di-orange-PIA-j.
"I like Fluffy being orange."
Note that the argument not retained is inaccessible.
* The tense marking needs more work.
Messages in this topic (12)
________________________________________________________________________
________________________________________________________________________
7a. Aeu phonology
Posted by: "Adam Walker" [email protected]
Date: Tue Apr 24, 2012 6:29 pm ((PDT))
I just got into a conversation on one of the FB conlang groups that made me
realize that I have no idea how to define a phoneme in Aeu, one of my
sketchier langs. The problem I have is figuring out where to draw the line
between phoneme and suprasegmental.
There are 9 unrounded vowels:
/i/ /ɨ/ /ɯ/
/e/ /ə/ /ɤ/
/æ/ /ɐ/ /ʌ/
I am certain these 9 are all phonemes.
Each of the above can be rounded as:
/y/ /ʉ/ /u/
/ø/ /ɵ/ /o/
/ɶ/ /ɞ/ /ɔ/
I also confidently assert that all of those are phonemes in Aeu.
That makes 18 phonemes.
Every one of those can be short, half-long, long or over-long. (I know at
least three of theese will be in the language, but I might not decide to go
for all four. I've waffled back and forth over the years.) Length is
lexically important. A A; A: and A:: are all different lexical items.
Should I count length as phonemic? It normally is in cases like this.
That would make anywhere from 54 to 72 phonemes.
I'm fairly confident up to that point, but then we get to the "voicings."
I'm not using phonations, because, as I understand it, nasalization is
normally not considered a phonation, but in Aeu, it forms a class with the
phonations, since a vowel of any rounding or length (or tone though we
haven't gotten to that yet) may be either naslized or not, but a vowel that
is creaky voice, may not be nasalize nor may it be breathy or rough. A
vowel may be plain, nasalized, creaky, breathy or rough, but it may not,
under any circumstances, be more than one of these simultaneously. Now
nasalization is often concidered phonemic, but what of the other "voicings"
in my list. Should/can I count breathy-/e/ as a distinct phoneme from
creaky-/e/. They are just as lexically important as the roundedness or the
length. This would mean that Aeu has anywhere from 270 to 360 phonemes.
Now we come to the 11 tones, and here I am really unsure how to describe
Aeu. Each of the tones may apply to any of the vowels, round or unround,
at any length, with any voicing (though there are bound to be holes here
iand there in the actual vocabulary where a given *possible* syllable is
not an *actual* word), and again tone is as lexically important in Aeu as
it is in Chinese, so should I count the tones as phonemes? In that case
Aeu has between 2970 and 3960 phonemes and essentially every possible
syllable is a unique phoneme.
Or should I say Aeu has 9 vowel phonemes, 2 rounding phonemes, 3 or 4
length phonemes, 5 voicing phonemes and 11 tonal phonemes and every
syllable requires exactly one phoneme from each category such that eveery
syllable is 5 phonemes.
Or should I be looking at some hybrid?
Or some entirely different analysis.
Adam
Messages in this topic (6)
________________________________________________________________________
7b. Re: Aeu phonology
Posted by: "MorphemeAddict" [email protected]
Date: Tue Apr 24, 2012 6:34 pm ((PDT))
On Tue, Apr 24, 2012 at 9:28 PM, Adam Walker <[email protected]> wrote:
> I just got into a conversation on one of the FB conlang groups that made me
> realize that I have no idea how to define a phoneme in Aeu, one of my
> sketchier langs. The problem I have is figuring out where to draw the line
> between phoneme and suprasegmental.
>
> There are 9 unrounded vowels:
>
> /i/ /ɨ/ /ɯ/
> /e/ /ə/ /ɤ/
> /æ/ /ɐ/ /ʌ/
>
> I am certain these 9 are all phonemes.
> Each of the above can be rounded as:
>
> /y/ /ʉ/ /u/
> /ø/ /ɵ/ /o/
> /ɶ/ /ɞ/ /ɔ/
>
> I also confidently assert that all of those are phonemes in Aeu.
> That makes 18 phonemes.
>
> Every one of those can be short, half-long, long or over-long. (I know at
> least three of theese will be in the language, but I might not decide to go
> for all four. I've waffled back and forth over the years.) Length is
> lexically important. A A; A: and A:: are all different lexical items.
> Should I count length as phonemic? It normally is in cases like this.
> That would make anywhere from 54 to 72 phonemes.
>
> I'm fairly confident up to that point, but then we get to the "voicings."
> I'm not using phonations, because, as I understand it, nasalization is
> normally not considered a phonation, but in Aeu, it forms a class with the
> phonations, since a vowel of any rounding or length (or tone though we
> haven't gotten to that yet) may be either naslized or not, but a vowel that
> is creaky voice, may not be nasalize nor may it be breathy or rough. A
> vowel may be plain, nasalized, creaky, breathy or rough, but it may not,
> under any circumstances, be more than one of these simultaneously. Now
> nasalization is often concidered phonemic, but what of the other "voicings"
> in my list. Should/can I count breathy-/e/ as a distinct phoneme from
> creaky-/e/. They are just as lexically important as the roundedness or the
> length. This would mean that Aeu has anywhere from 270 to 360 phonemes.
>
> Now we come to the 11 tones, and here I am really unsure how to describe
> Aeu. Each of the tones may apply to any of the vowels, round or unround,
> at any length, with any voicing (though there are bound to be holes here
> iand there in the actual vocabulary where a given *possible* syllable is
> not an *actual* word), and again tone is as lexically important in Aeu as
> it is in Chinese, so should I count the tones as phonemes? In that case
> Aeu has between 2970 and 3960 phonemes and essentially every possible
> syllable is a unique phoneme.
>
> Or should I say Aeu has 9 vowel phonemes, 2 rounding phonemes, 3 or 4
> length phonemes, 5 voicing phonemes and 11 tonal phonemes and every
> syllable requires exactly one phoneme from each category such that eveery
> syllable is 5 phonemes.
>
I don't have an answer, but I prefer the latter. It's a lot more concise,
maybe 'efficient', even elegant.
stevo
>
> Or should I be looking at some hybrid?
>
> Or some entirely different analysis.
>
> Adam
>
Messages in this topic (6)
________________________________________________________________________
7c. Re: Aeu phonology
Posted by: "Adam Walker" [email protected]
Date: Tue Apr 24, 2012 6:39 pm ((PDT))
On Tue, Apr 24, 2012 at 8:33 PM, MorphemeAddict <[email protected]> wrote:
> On Tue, Apr 24, 2012 at 9:28 PM, Adam Walker <[email protected]> wrote:
> >
> > Or should I say Aeu has 9 vowel phonemes, 2 rounding phonemes, 3 or 4
> > length phonemes, 5 voicing phonemes and 11 tonal phonemes and every
> > syllable requires exactly one phoneme from each category such that eveery
> > syllable is 5 phonemes.
> >
>
> I don't have an answer, but I prefer the latter. It's a lot more concise,
> maybe 'efficient', even elegant.
>
> stevo
>
>
>
It is undeniably more concise. The first description is a monstrosity. I
just don't know if the concise description is defensible on any other
grounds than concision and elegance. But then, that alone may be enough.
I will wait to hear from Dirk and any other RL phonologists before making
any further attempt at describing this lang.
Adam
Messages in this topic (6)
________________________________________________________________________
7d. Re: Aeu phonology
Posted by: "Dirk Elzinga" [email protected]
Date: Tue Apr 24, 2012 7:30 pm ((PDT))
The question you need to ask yourself is whether it is really the vowel
that serves as a distinctive unit. In this case, I would say no. The
distinctive units in your system are whatever properties or features you
assign to distinguish one vowel quality from another. In your scheme as
described you have the following features: 1) vowel height, with 3 values,
2) vowel fronting, with 3 values, 3) rounding, with 2 values, 4) length,
with 3 (or 4) values, 5) phonation, with 5 values, and 6) tone, with 11
values.[1] This constitutes a 6-dimensional space within which vowels vary
from one another. A vowel segment in this system is simply the intersection
of feature values specified for a particular vowel slot. The notion of
'segment' really isn't useful anymore, except as a shorthand for this
intersection of properties. The term 'phoneme' probably isn't helpful at
this point, either.
As for whether there are any advantages to this sort of description, I
would say there definitely are. Don't discount concision and elegance as
defensible criteria for choosing one description over another. I would much
prefer to see a concise, elegant description in terms of intersecting
features or properties over one that obtusely clings to segmental
representations.
Dirk
[1] Most phonologists now agree that tone is actually a property of the
syllable or mora, rather than of the vowel. This opens up the possibility
that other properties of yours might also be syllable-level properties
(phonation is a good candidate for this sort of thing). Phonological
description and analysis becomes a lot more challenging (and fun!) when you
can assign features to different tiers of representation.
On Tue, Apr 24, 2012 at 7:39 PM, Adam Walker <[email protected]> wrote:
> On Tue, Apr 24, 2012 at 8:33 PM, MorphemeAddict <[email protected]> wrote:
>
> > On Tue, Apr 24, 2012 at 9:28 PM, Adam Walker <[email protected]>
> wrote:
> > >
> > > Or should I say Aeu has 9 vowel phonemes, 2 rounding phonemes, 3 or 4
> > > length phonemes, 5 voicing phonemes and 11 tonal phonemes and every
> > > syllable requires exactly one phoneme from each category such that
> eveery
> > > syllable is 5 phonemes.
> > >
> >
> > I don't have an answer, but I prefer the latter. It's a lot more concise,
> > maybe 'efficient', even elegant.
> >
> > stevo
> >
> >
> >
> It is undeniably more concise. The first description is a monstrosity. I
> just don't know if the concise description is defensible on any other
> grounds than concision and elegance. But then, that alone may be enough.
> I will wait to hear from Dirk and any other RL phonologists before making
> any further attempt at describing this lang.
>
> Adam
>
Messages in this topic (6)
________________________________________________________________________
7e. Re: Aeu phonology
Posted by: "Adam Walker" [email protected]
Date: Tue Apr 24, 2012 7:55 pm ((PDT))
Thank you, Dirk, that helps significantly. So I think I will try
working with this scheme. However since Aeu has no consonants and no
dipthongs, I am not quite sure how tiers would come into play. A vowel
and a syllable are synonymous in this lang. Adam
On 4/24/12, Dirk Elzinga <[email protected]> wrote:
> The question you need to ask yourself is whether it is really the vowel
> that serves as a distinctive unit. In this case, I would say no. The
> distinctive units in your system are whatever properties or features you
> assign to distinguish one vowel quality from another. In your scheme as
> described you have the following features: 1) vowel height, with 3 values,
> 2) vowel fronting, with 3 values, 3) rounding, with 2 values, 4) length,
> with 3 (or 4) values, 5) phonation, with 5 values, and 6) tone, with 11
> values.[1] This constitutes a 6-dimensional space within which vowels vary
> from one another. A vowel segment in this system is simply the intersection
> of feature values specified for a particular vowel slot. The notion of
> 'segment' really isn't useful anymore, except as a shorthand for this
> intersection of properties. The term 'phoneme' probably isn't helpful at
> this point, either.
>
> As for whether there are any advantages to this sort of description, I
> would say there definitely are. Don't discount concision and elegance as
> defensible criteria for choosing one description over another. I would much
> prefer to see a concise, elegant description in terms of intersecting
> features or properties over one that obtusely clings to segmental
> representations.
>
> Dirk
>
> [1] Most phonologists now agree that tone is actually a property of the
> syllable or mora, rather than of the vowel. This opens up the possibility
> that other properties of yours might also be syllable-level properties
> (phonation is a good candidate for this sort of thing). Phonological
> description and analysis becomes a lot more challenging (and fun!) when you
> can assign features to different tiers of representation.
>
>
> On Tue, Apr 24, 2012 at 7:39 PM, Adam Walker <[email protected]> wrote:
>
>> On Tue, Apr 24, 2012 at 8:33 PM, MorphemeAddict <[email protected]> wrote:
>>
>> > On Tue, Apr 24, 2012 at 9:28 PM, Adam Walker <[email protected]>
>> wrote:
>> > >
>> > > Or should I say Aeu has 9 vowel phonemes, 2 rounding phonemes, 3 or 4
>> > > length phonemes, 5 voicing phonemes and 11 tonal phonemes and every
>> > > syllable requires exactly one phoneme from each category such that
>> eveery
>> > > syllable is 5 phonemes.
>> > >
>> >
>> > I don't have an answer, but I prefer the latter. It's a lot more
>> > concise,
>> > maybe 'efficient', even elegant.
>> >
>> > stevo
>> >
>> >
>> >
>> It is undeniably more concise. The first description is a monstrosity. I
>> just don't know if the concise description is defensible on any other
>> grounds than concision and elegance. But then, that alone may be enough.
>> I will wait to hear from Dirk and any other RL phonologists before making
>> any further attempt at describing this lang.
>>
>> Adam
>>
>
Messages in this topic (6)
________________________________________________________________________
7f. Re: Aeu phonology
Posted by: "Nikolay Ivankov" [email protected]
Date: Wed Apr 25, 2012 4:56 am ((PDT))
My Janjarin project is not that radical about the vowels: there are 4
consonants, and used to be more in proto-Janjarin. I threat nasalization
and voicing as separate things: IIRC I was told here that it can be compose
with breathy and creaky voices, and I can actually pronounce them. Since
the Yanja people have a bit different anatomy - in particular thier tongue
is of such a form that they just cannot pronounce other consonants than
bilabial, uvular and glottal - I may assume that further differences may
make such kind of voicings absolutely non-problematic for them
I also consider closedness of the vowels as another phonation feature.
Thus, I have 4 basic vowels:
a, o, ɨ, u
which can be:
1) closed (y/n) yielding e, ø, i, y when yes.
2) nasalized (y/n)
3) plain/breathy/creaky
As in Aeu, there are 11 tones in Janjarin. However, only the relative tone
plays a role, and it may rise and fall for no more that 3 tones in relation
to the previous vowel, so that the actual number of tones is 7. The 11
tones comes out from the polysyntheistic character of the language. The
phonation does not come alone either. In the writing is is represented by
the characters n, h and r for nasalization, breathy and creaky voice
respectively, the clusters hn, rn allowed. Unless there is no w or j
preceding or following the these characters, the vowels or -phthongs around
n, r, hn and rn become phonated. Also, according to phonological roots, a
open vowel surrounded by two closed also becomes closed. In proto-Janjarin,
there used to be the consonants like я, ŋ, h and possibly some other
variants of r, n and h sounds, that have later been accumulated in
phonations.
To make things worse, the vowels accumulate into -phthongs, and there is a
liaison (if I use this term correctly) of the unstressed vowels: so, for
instance, if in as separate word parts one of the -phthongs is ends on a
and the other starts with e, then they combine to become æ, and similarly
with others.
Yep, a shameless self-promotion.
On Wed, Apr 25, 2012 at 3:28 AM, Adam Walker <[email protected]> wrote:
> I just got into a conversation on one of the FB conlang groups that made me
> realize that I have no idea how to define a phoneme in Aeu, one of my
> sketchier langs. The problem I have is figuring out where to draw the line
> between phoneme and suprasegmental.
>
> There are 9 unrounded vowels:
>
> /i/ /ɨ/ /ɯ/
> /e/ /ə/ /ɤ/
> /æ/ /ɐ/ /ʌ/
>
> I am certain these 9 are all phonemes.
> Each of the above can be rounded as:
>
> /y/ /ʉ/ /u/
> /ø/ /ɵ/ /o/
> /ɶ/ /ɞ/ /ɔ/
>
> I also confidently assert that all of those are phonemes in Aeu.
> That makes 18 phonemes.
>
> Every one of those can be short, half-long, long or over-long. (I know at
> least three of theese will be in the language, but I might not decide to go
> for all four. I've waffled back and forth over the years.) Length is
> lexically important. A A; A: and A:: are all different lexical items.
> Should I count length as phonemic? It normally is in cases like this.
> That would make anywhere from 54 to 72 phonemes.
>
> I'm fairly confident up to that point, but then we get to the "voicings."
> I'm not using phonations, because, as I understand it, nasalization is
> normally not considered a phonation, but in Aeu, it forms a class with the
> phonations, since a vowel of any rounding or length (or tone though we
> haven't gotten to that yet) may be either naslized or not, but a vowel that
> is creaky voice, may not be nasalize nor may it be breathy or rough. A
> vowel may be plain, nasalized, creaky, breathy or rough, but it may not,
> under any circumstances, be more than one of these simultaneously. Now
> nasalization is often concidered phonemic, but what of the other "voicings"
> in my list. Should/can I count breathy-/e/ as a distinct phoneme from
> creaky-/e/. They are just as lexically important as the roundedness or the
> length. This would mean that Aeu has anywhere from 270 to 360 phonemes.
>
> Now we come to the 11 tones, and here I am really unsure how to describe
> Aeu. Each of the tones may apply to any of the vowels, round or unround,
> at any length, with any voicing (though there are bound to be holes here
> iand there in the actual vocabulary where a given *possible* syllable is
> not an *actual* word), and again tone is as lexically important in Aeu as
> it is in Chinese, so should I count the tones as phonemes? In that case
> Aeu has between 2970 and 3960 phonemes and essentially every possible
> syllable is a unique phoneme.
>
> Or should I say Aeu has 9 vowel phonemes, 2 rounding phonemes, 3 or 4
> length phonemes, 5 voicing phonemes and 11 tonal phonemes and every
> syllable requires exactly one phoneme from each category such that eveery
> syllable is 5 phonemes.
>
> Or should I be looking at some hybrid?
>
> Or some entirely different analysis.
>
> Adam
>
Messages in this topic (6)
------------------------------------------------------------------------
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/conlang/
<*> Your email settings:
Digest Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/conlang/join
(Yahoo! ID required)
<*> To change settings via email:
[email protected]
[email protected]
<*> To unsubscribe from this group, send an email to:
[email protected]
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
------------------------------------------------------------------------
