There are 16 messages in this issue.

Topics in this digest:

1a. Re: Diversity in conlang families    
    From: David McCann
1b. Re: Diversity in conlang families    
    From: John Vertical

2a. Re: Font/word-processing question    
    From: David McCann
2b. Re: Font/word-processing question    
    From: p...@phillipdriscoll.com
2c. Re: Font/word-processing question    
    From: Calculator Ftvb
2d. Re: Font/word-processing question    
    From: Calculator Ftvb

3a. Re: Verbs-of-motion-centric language    
    From: Garth Wallace

4a. Re: what is the meaning of "vs" in music?    
    From: John Vertical

5a. OT: Techno [was Re: what is the meaning of "vs" in music?]    
    From: Garth Wallace
5b. Re: OT: Techno [was Re: what is the meaning of "vs" in music?]    
    From: Logan Kearsley

6a. Re: Toward an understanding of serial verbs    
    From: Leland Kusmer

7. 30-Day Conlang: Day 24    
    From: Gary Shannon

8. Latin/Chinese altlang    
    From: Dale McCreery

9a. Fully featured language    
    From: Ben Collier
9b. Re: Fully featured language    
    From: Samuel Stutter
9c. Re: Fully featured language    
    From: Daniel Nielsen


Messages
________________________________________________________________________
1a. Re: Diversity in conlang families
    Posted by: "David McCann" da...@polymathy.plus.com 
    Date: Wed Nov 24, 2010 8:28 am ((PST))

On Tue, 2010-11-23 at 18:37 -0500, Roman Rausch wrote:

> Now that you mention it - my languages were all SOV until now, but I also
> really like VSO. I'm unsure, however, how changes in word order come about
> at all. Is there a recommendable source to read up on it?

The second chapter of The Romance Languages (ed. M. Harris & N. Vincent)
discusses the change from SOV to SVO (and VS) in Latin and Romance.
There's a discussion of the similar changes in Greek, with a different
explanation, that I think is in Greek: a history, by Geoffrey Horrocks.





Messages in this topic (4)
________________________________________________________________________
1b. Re: Diversity in conlang families
    Posted by: "John Vertical" johnverti...@hotmail.com 
    Date: Wed Nov 24, 2010 11:23 am ((PST))

On Mon, 22 Nov 2010 16:39:27 +0100, Jörg Rhiemeier wrote:
>Hallo!
>
>To those who build families of related diachronic conlangs:
>
>How much diversity do you build into your families?  Do you just use
>the same morphosyntactic structure and build languages differing only
>in terms of sound changes, such that you could translate between your
>conlangs almost morpheme by morpheme (natlang example: Vedic and
>Avestan)?  Or do you throw in as much typological diversity as seems
>plausible to you, such that almost every major language type is
>represented in the family?  Or do you follow a middle road between
>these extremes?

Well, nominally I've more families going on than I can keep track of, and I
do intend to cover both Sprachbund areas and hotspots of diversification, so
most of these, I suppose? However, in practice these projects are mostly
sound change excercises languishing with little grammar and prectially no
syntax…

At least East Persian' is tending towards monosyllabicity while West
Persian' has been developing an initial mutation system, so that's some
degree of morphophonological variableness.

John Vertical





Messages in this topic (4)
________________________________________________________________________
________________________________________________________________________
2a. Re: Font/word-processing question
    Posted by: "David McCann" da...@polymathy.plus.com 
    Date: Wed Nov 24, 2010 8:33 am ((PST))

On Tue, 2010-11-23 at 18:46 -0500, MorphemeAddict wrote:
> Write it left to right and top to bottom, and then turn it upside down.

There's a almost real-world precedent here. Mongolian is written
downwards, but it's derived, via Sogdian, from Syriac, which is
right-to-left. It seems Sogdian scribes wrote downwards (presumably to
avoid smudging) and turn the text to read.





Messages in this topic (14)
________________________________________________________________________
2b. Re: Font/word-processing question
    Posted by: "p...@phillipdriscoll.com" p...@phillipdriscoll.com 
    Date: Wed Nov 24, 2010 8:43 am ((PST))

David McCann wrote: > MorphemeAddict wrote:
  > > Write it left to right and top to bottom, and then turn it upside down. 
  >
  > There's a almost real-world precedent here. Mongolian is written
  > downwards, but it's derived, via Sogdian, from Syriac, which is
  > right-to-left. It seems Sogdian scribes wrote downwards (presumably to
  > avoid smudging) and turn the text to read. 
 
 
About fifteen years ago I worked with a man from China. 
He had a program called China Star which allowed him to
use Chinese characters in Windows (3.1!). Among other
things, it could display glyphs rotated 90 or 270 degrees,
so one could use MS Word left-to-right, then after
printing, they would appear top-to-bottom. 
 
--Ph. D. 
 





Messages in this topic (14)
________________________________________________________________________
2c. Re: Font/word-processing question
    Posted by: "Calculator Ftvb" i...@futuramerlin.com 
    Date: Wed Nov 24, 2010 1:21 pm ((PST))

This? http://www.njstar.com/cms/njstar-chinese-word-processor

On 24 November 2010 11:40, <p...@phillipdriscoll.com> wrote:

> David McCann wrote: > MorphemeAddict wrote:
>  > > Write it left to right and top to bottom, and then turn it upside
> down.  >
>  > There's a almost real-world precedent here. Mongolian is written
>  > downwards, but it's derived, via Sogdian, from Syriac, which is
>  > right-to-left. It seems Sogdian scribes wrote downwards (presumably to
>  > avoid smudging) and turn the text to read.
>
> About fifteen years ago I worked with a man from China. He had a program
> called China Star which allowed him to
> use Chinese characters in Windows (3.1!). Among other
> things, it could display glyphs rotated 90 or 270 degrees,
> so one could use MS Word left-to-right, then after
> printing, they would appear top-to-bottom.
> --Ph. D.
>





Messages in this topic (14)
________________________________________________________________________
2d. Re: Font/word-processing question
    Posted by: "Calculator Ftvb" i...@futuramerlin.com 
    Date: Wed Nov 24, 2010 1:30 pm ((PST))

Ideas for a conlang word-processor : I think a useful piece of software
might be one that would allow any text direction whatsoever, and include the
ability to create fonts and encodings, so a person could construct
customised fonts, keyboard layouts, and character encodings for their
languages. Also helpful in such a thing would be the ability to have more
than 64k characters in a font (an inconvenient limit I've encountered when
trying to assemble pan-Unicode fonts using FontForge). Also, it might be
helpful to include support for stuff like one encounters in music such as
characters relying on semantically-significant background content like a
staff. Also potentially useful would be side-stacked diacritics, again like
in music (accidentals). Also useful might be the ability to have characters
that could stretch based on context (again, like slurs and ties in music).

Hmmm. I see a very-difficult-to-write program idea in the previous
paragraph... :-P

—Calculator Ftvb

On 24 November 2010 16:17, Calculator Ftvb <i...@futuramerlin.com> wrote:

> This? http://www.njstar.com/cms/njstar-chinese-word-processor
>
> On 24 November 2010 11:40, <p...@phillipdriscoll.com> wrote:
>
>> David McCann wrote: > MorphemeAddict wrote:
>>  > > Write it left to right and top to bottom, and then turn it upside
>> down.  >
>>  > There's a almost real-world precedent here. Mongolian is written
>>  > downwards, but it's derived, via Sogdian, from Syriac, which is
>>  > right-to-left. It seems Sogdian scribes wrote downwards (presumably to
>>  > avoid smudging) and turn the text to read.
>>
>> About fifteen years ago I worked with a man from China. He had a program
>> called China Star which allowed him to
>> use Chinese characters in Windows (3.1!). Among other
>> things, it could display glyphs rotated 90 or 270 degrees,
>> so one could use MS Word left-to-right, then after
>> printing, they would appear top-to-bottom.
>> --Ph. D.
>>
>
>





Messages in this topic (14)
________________________________________________________________________
________________________________________________________________________
3a. Re: Verbs-of-motion-centric language
    Posted by: "Garth Wallace" gwa...@gmail.com 
    Date: Wed Nov 24, 2010 8:54 am ((PST))

On Tue, Nov 23, 2010 at 8:29 PM, Amanda Babcock Furrow
<la...@quandary.org> wrote:
>
> Heck, a lot of things can be paraphrased in terms of going, entering,
> exiting, being located at, and even more in terms of putting, taking,
> sending etc.  Writing?  "Put words/thoughts/pen to paper."  Cutting?
> "Removed pieces with a knife."

Hmm, but cutting doesn't necessarily remove pieces. Maybe a verb for a
group splitting in two directions?

> One of the less likely semantic fields
> to be expressed in terms of location might be speech acts, but that
> seems to me more because of what we would judge as basic and
> fundamental rather than due to any problem phrasing it as "put words
> in their ears".

Looking at it the other way around, I wonder how you could
metaphorically extend a verb for the act of speaking to movement.

> I'm about to be the mother of an infant again, so I don't really plan
> to pursue this idea, but it seems an interesting constraint.  A rather
> different set of basic true verbs from that we occasionally see of "do,
> be, have"!  Has anybody tried it?

Congratulations!





Messages in this topic (2)
________________________________________________________________________
________________________________________________________________________
4a. Re: what is the meaning of "vs" in music?
    Posted by: "John Vertical" johnverti...@hotmail.com 
    Date: Wed Nov 24, 2010 11:18 am ((PST))

On Mon, 22 Nov 2010 21:25:04 -0500, Daniel Bowman wrote:
>I was wondering what the "vs" means when used for the artists of a song.
>For example:  "Globe vs. Push-Dreams from Above" or "Darude vs. Paul
>Okenfold-Some Techno Song"

Bit of a tangent here, but Darude and Oakenfold play trance, not techno.
Sorry, bit of a pet peeve of mine. :P

Cf. eg: http://rateyourmusic.com/list/Miklak/this_is_not_techno___

John Vertical





Messages in this topic (7)
________________________________________________________________________
________________________________________________________________________
5a. OT: Techno [was Re: what is the meaning of "vs" in music?]
    Posted by: "Garth Wallace" gwa...@gmail.com 
    Date: Wed Nov 24, 2010 1:23 pm ((PST))

On Wed, Nov 24, 2010 at 11:13 AM, John Vertical
<johnverti...@hotmail.com> wrote:
> On Mon, 22 Nov 2010 21:25:04 -0500, Daniel Bowman wrote:
>>I was wondering what the "vs" means when used for the artists of a song.
>>For example:  "Globe vs. Push-Dreams from Above" or "Darude vs. Paul
>>Okenfold-Some Techno Song"
>
> Bit of a tangent here, but Darude and Oakenfold play trance, not techno.
> Sorry, bit of a pet peeve of mine. :P
>
> Cf. eg: http://rateyourmusic.com/list/Miklak/this_is_not_techno___

Basshunter? I thought he was classed as Jumpstyle, which derived from
hardcore techno and was still considered a subgenre thereof?

(Also, Crazy Frog should be excluded from all lists of anything, on
general principle)





Messages in this topic (2)
________________________________________________________________________
5b. Re: OT: Techno [was Re: what is the meaning of "vs" in music?]
    Posted by: "Logan Kearsley" chronosur...@gmail.com 
    Date: Wed Nov 24, 2010 2:17 pm ((PST))

On Wed, Nov 24, 2010 at 2:20 PM, Garth Wallace <gwa...@gmail.com> wrote:
> On Wed, Nov 24, 2010 at 11:13 AM, John Vertical
> <johnverti...@hotmail.com> wrote:
>> On Mon, 22 Nov 2010 21:25:04 -0500, Daniel Bowman wrote:
>>>I was wondering what the "vs" means when used for the artists of a song.
>>>For example:  "Globe vs. Push-Dreams from Above" or "Darude vs. Paul
>>>Okenfold-Some Techno Song"
>>
>> Bit of a tangent here, but Darude and Oakenfold play trance, not techno.
>> Sorry, bit of a pet peeve of mine. :P
>>
>> Cf. eg: http://rateyourmusic.com/list/Miklak/this_is_not_techno___
>
> Basshunter? I thought he was classed as Jumpstyle, which derived from
> hardcore techno and was still considered a subgenre thereof?
>
> (Also, Crazy Frog should be excluded from all lists of anything, on
> general principle)

I disagree with a lot of entries on that list. I find the huge number
of super-specific and mutually exclusive categories that Electronica
gets divided into to be rather ridiculous. If you can't classify it
unambiguously in five seconds of listening, then your classification
system is probably not useful outside of some sort of academic
analysis setting, and you should probably fall back on more inclusive
groupings.

Turning this into a discussion of just what the lexical space diagram
of all of the electronic music genre names looks like (or some other
similar topic) could even get this back on-topic as a conlanging
discussion.

I tend to just refer to everything as generic "Electronica" to avoid
upsetting anybody.

-l.





Messages in this topic (2)
________________________________________________________________________
________________________________________________________________________
6a. Re: Toward an understanding of serial verbs
    Posted by: "Leland Kusmer" lelandp...@thypyramids.com 
    Date: Wed Nov 24, 2010 5:02 pm ((PST))

>
>
> >
> So I won't be terribly naive if I do the same thing.  That's good.
>

Nope! Not at all. See above, though, re: bizarre other things that happen in
SVCs (and similar constructions).

>
> What grammatical functions are served by serial verbs, then?  It seems
> they're indicating something aspecty in some of your examples.
>

I mean, mostly in Akan SVCs (and related constructions) serve stylistic
rather than grammatical purpose – that is, you can express any of the above
examples completely sans SVCs, but the result is considerably longer for
having overt coordination, pronouns to reintroduce the subject, etc. When I
was doing fieldwork on the language, I'd often give prompts that had, in the
past or with other speakers, elicited an SVC, and get some other
construction as a result. As an example, these two sentences are basically
in free alternation, so far as I could tell:

1)
Kofi twaa       dua no tOO       fom.
Kofi  cut.past  tree the fall.past ground.
"Kofi cut the tree down."

2)
Kofi twaa dua no na    3-tOO       fom.
                         and   3s-fall.past
"Kofi cut the tree and it fell to the ground."

While I don't know this for certain, I'd expect there to be discourse
function, really – (2) probably foregrounds the bit about the tree hitting
the ground, as I tried to suggest in my gloss. Common action sequences are
more likely to be expressed via SVCs than uncommon ones, etc.

Generally speaking, true SVCs in Akan seem to be primarily about consecutive
actions – there's an isomorphy in which words spoken sooner denote actions
which occurred sooner. Again, though, note the weirdness in the past tense
mentioned above.


> Of *course* I want to know more about weird and awesome do-insertion in
> Akan!


Mmkay, here goes – this is mostly recounting research done by my
professor[0]. I'll give the evidence first, and then give an overview of the
High Chomskian analysis, if you care.[1]

The data:

/y3/ is the verb "to do" or "to make" – about as semantically bland as the
English verb, and used in a similar range of situations. (For instance, when
questioning what action someone performed, it's the default verb: <Hwain na
Kofi y33?> "What did Kofi do?".)

Various scholars have noted that /y3/ [2] shows up in a surprising place –
after past tense intransitive verbs:

3) Kofi saa            y3.
    Kofi dance.past do.

Most scholars have thus tried to say that /y3/, in those situations, is not
actually the verb "to do", but really just an allomorph of the past-tense
suffix (see note [2] for why this occasionally looks more true). However, it
turns out that this is not the only place it turns up. For instance, the 3rd
person object pronoun is null in Twi, so we get:

4a) Kofi bO          __.
           hit  3s.acc
"Kofi hits him/her/it."

4b) Kofi bOO       __          y3.
            hit.past   3s.acc    do.
"Kofi hit him/her/it."

Similarly, if we topicalize the object, we get it – basically, the rule
seems to be, insert "do" in the past tense whenever the verb would occur at
the very end of the sentence. But this prompts a question: What if we put in
adverbs at the end? This seems to work, at first:

5) Kofi saa (*y3) nt3m (*y3).
                        quickly
"Kofi danced quickly."

So putting in a manner adverb at the end makes do-insertion ungrammatical.
However, then we notice this weirdness:

6) Kofi saa *(y3) amparampara.
                         truly
"Truly, Kofi danced."

Putting a sentential adverb at the end doesn't change the do-insertion!
Bizarre!




The High Chomskian Analysis, in brief:

What differentiates manner adverbs from sentential adverbs, syntactically?
In the usual framework, we'd say that manner adverbs adjoin to the verb
phrase, while sentential adverbs adjoin to the whole clause – that is, at
TP. If we then allow that Akan shows V-to-T movement in the past tense,
where the verb actually leaves the verb phrase to land in T in order to get
that lengthened vowel, then the generalization is: Do-insertion in Akan
occurs whenever the verb phrase would be left entirely empty after all
movement has occurred. Kandybowicz justifies this by dealing with the
syntax-phonology interface – basically, the language needs to have a
prosodic phrase aligned with the verb phrase, but has trouble doing this if
there's nothing in the verb phrase, so it inserts /y3/.


ObConlang: Does anyone have conlangs that do anything weird with
do-insertion (or similar phenomenon of putting in low-content words in
certain situations)? I'll admit to not having done anything with it, but
definitely want to work it into one of my projects somewhere after dealing
with Twi.


-Leland

[0] Jason Kandybowicz, http://www.swarthmore.edu/SocSci/jkandyb1/
[1] Given that Kandybowicz is my thesis advisor, I'm currently required to
work in this framework, but personally find it highly suspect. It works
pretty nicely for these results, though.
[2] This is for Asante Twi. In many other dialects it's become a suffix
-i/-e (form depends on ATR harmony), but follows the same distribution.





Messages in this topic (5)
________________________________________________________________________
________________________________________________________________________
7. 30-Day Conlang: Day 24
    Posted by: "Gary Shannon" fizi...@gmail.com 
    Date: Wed Nov 24, 2010 6:16 pm ((PST))

As of day 24 the McGuffey translation now has 77 sentences translated,
plus a new pronoun/tense marker reference chart. There is also a
review vocabulary and a quiz on the first sentences.

http://fiziwig.com/conlang/txtana_mcguffey.html

--gary





Messages in this topic (1)
________________________________________________________________________
________________________________________________________________________
8. Latin/Chinese altlang
    Posted by: "Dale McCreery" mccre...@uvic.ca 
    Date: Wed Nov 24, 2010 10:23 pm ((PST))

I remember reading on the list about some speculation about Roman soldiers
in China and an altlang?  Here's an article on a village in China that is
believed (by some) to be descended from Roman Soldiers.  This seems
relevant to the interests of this list despite no obvious reference to
language or conlanging.

http://www.telegraph.co.uk/news/worldnews/asia/china/8154490/Chinese-villagers-descended-from-Roman-soldiers.html?sms_ss=reddit&at_xt=4cedc8fb39e0e671,0





Messages in this topic (1)
________________________________________________________________________
________________________________________________________________________
9a. Fully featured language
    Posted by: "Ben Collier" bmcoll...@gmail.com 
    Date: Thu Nov 25, 2010 5:25 am ((PST))

Hello all,

I was wondering if there had been an attempt to create a language
which incorporated as many linguistic features, tenses, declensions
etc. as possible, to enable much greater a range of both specificities
and subtleties in speech.

Has anything like that bee tried?

Thanks

Ben





Messages in this topic (3)
________________________________________________________________________
9b. Re: Fully featured language
    Posted by: "Samuel Stutter" sam.stut...@student.manchester.ac.uk 
    Date: Thu Nov 25, 2010 6:02 am ((PST))

My obsessive work on Nauspayr is pretty like that, stemming from a youthful 
hijacking of a linguistics reference book and inserting every feature possible. 
Frankly, it's killing me. The never ending project calls for 6 deictic 
positions, 22 moods, 26 tenses, 28 aspects, 16 persons, 22 cases, 4 noun 
classes, 5 plurals,  and 4 suffix conjunctions and 4 stem-changing 
conjugations. The fact the language is quite fusional is just another hateful 
element. I love the language - it's nice and technical, but developing new 
affixes is hell, not to mention translations from English which require 
substantial use of reference material ("now what's the 1st plural inclusive 
optative Hesternal past 3rd stem changing affix for the irregular verb 'to 
be'?").

My advice, if you value your social life, don't :D



On 25 Nov 2010, at 13:09, Ben Collier <bmcoll...@gmail.com> wrote:

> Hello all,
> 
> I was wondering if there had been an attempt to create a language
> which incorporated as many linguistic features, tenses, declensions
> etc. as possible, to enable much greater a range of both specificities
> and subtleties in speech.
> 
> Has anything like that bee tried?
> 
> Thanks
> 
> Ben





Messages in this topic (3)
________________________________________________________________________
9c. Re: Fully featured language
    Posted by: "Daniel Nielsen" niel...@uah.edu 
    Date: Thu Nov 25, 2010 6:26 am ((PST))

Ben,

Try googling Logopandekteision.





Messages in this topic (3)





------------------------------------------------------------------------
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:
    conlang-nor...@yahoogroups.com 
    conlang-fullfeatu...@yahoogroups.com

<*> To unsubscribe from this group, send an email to:
    conlang-unsubscr...@yahoogroups.com

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/
 
------------------------------------------------------------------------

Reply via email to