There are 8 messages in this issue.
Topics in this digest:
1a. Re: THEORY: Lost of final-syllable rhotic.
From: Alex Fink
2a. Re: Proto-Jardic noun morphology
From: Herman Miller
2b. Re: Proto-Jardic noun morphology
From: Herman Miller
2c. Re: Proto-Jardic noun morphology
From: BPJ
2d. Re: Proto-Jardic noun morphology
From: Roman Rausch
2e. Re: Proto-Jardic noun morphology
From: Herman Miller
2f. Re: Proto-Jardic noun morphology
From: Jörg Rhiemeier
2g. Re: Proto-Jardic noun morphology
From: Roger Mills
Messages
________________________________________________________________________
1a. Re: THEORY: Lost of final-syllable rhotic.
Posted by: "Alex Fink" [email protected]
Date: Fri Feb 15, 2013 4:05 pm ((PST))
On Fri, 15 Feb 2013 10:43:08 -0200, Njenfalgar <[email protected]> wrote:
>I think there is a universal tendency to drop word-final and syllable-final
>stuff. :-)
On Fri, 15 Feb 2013 16:52:57 +0000, David McCann <[email protected]>
wrote:
>Obviously there's a general tendency to simplify syllables to CV, which
>different languages resist to different degrees.
There's certainly a pretty strong tendency to these things. But,
crosslinguistically, I'd want to ascribe the bulk of it to ultimate effects of
perceptual factors. Stops before vowels, e.g., have a brilliantly salient and
distinctive release burst that's easy to latch onto in a speech stream,
something which stops after vowels have no analogue of. So it's only natural
for a language's speakers to generalise this and pay more attention to CV
transitions in general than VC ones, and as a consequence of this comparative
neglect, consonants which only have a VC transition -- like final ones -- would
be subject to more erosion, since it's easier to miss the cues for what they
are.
One reason I think this is an argument of Blevins, regarding the Australian
sprachbund, whose languages are barse ackward phonologically in a lot of ways:
for instance, *initial* consonant weakening and loss is extremely common;
languages like Arrernte have been reasonably analysed as having basic syllable
structure VC. But there are other features of the area which suggest that VC
transitions are more important than CV. For instance, retroflexion and
laminality contrasts on consonants (and graveness???), which Australia has in
spades, are among the few consonantal contrasts which have more distinctive
effects on a previous vowel than a following. And Australia has lots of
non-assimilated nasal plus stop clusters, which again aren't so awkward if
you're reading the place of the nasal off its effects on the previous V. It's
reasonable to think that the VC syllable stuff might also be an effect.
Alex
Messages in this topic (9)
________________________________________________________________________
________________________________________________________________________
2a. Re: Proto-Jardic noun morphology
Posted by: "Herman Miller" [email protected]
Date: Fri Feb 15, 2013 5:45 pm ((PST))
On 2/15/2013 4:12 PM, Jörg Rhiemeier wrote:
> Long vowels are often more "tense", i.e. closer to the edges of
> the vowel space, and therefore perhaps more resistant against
> assimilatory processes. At least, this does not seem implausible
> to me. Maybe /o/ was fronted and /o:/ only centralized, which
> later shifted back to /o/.
Or I could be going about it wrong, maybe Proto-Jardic had *eu, *ou, *au
in these words, which developed into /ö/, /ó/, /o/.
>> Well, it's a sound that really doesn't seem to fit with the rest of the
>> Proto-Jardic sounds. The Jarda sound is more like trying to say /j/ with
>> your tongue tip curled back than it's anything like an American English
>> /r/. When it dissimilates after /r/, it changes to /j/. So I don't know
>> that I'd call it a rhotic even though I write it as<ṛ>.
>
> What you describe seems to be a palatalized retroflex approximant.
> Certainly an interesting phoneme. Perhaps from a Proto-Jardic
> cluster */rj/ or something like that?
Well, there's the interesting point that Jarda doesn't have any "śr" or
"źr" clusters, even though C + /r/ clusters are common, but "śṛ" and
"źṛ" clusters do occur. Maybe the Proto-Jarda for these was *srj-,
*zrj-. On the other hand both "r" sounds occur after initial /s/ and /z/
(sr-, sṛ-, zr-, zṛ-). (Maybe a vowel was lost in sṛ- and zṛ- words.
Certainly Jarda seems to have lost many short vowels.)
Palatalized rhotics can produce all sorts of strange sounds like the
Czech fricative trills, so why not a sound like the Jarda /ṛ/?
Messages in this topic (15)
________________________________________________________________________
2b. Re: Proto-Jardic noun morphology
Posted by: "Herman Miller" [email protected]
Date: Fri Feb 15, 2013 6:56 pm ((PST))
On 2/15/2013 5:42 PM, BPJ wrote:
> I second the suggestions that the various case endings arose from an
> amalgam of gender markers, case markers and postpositions.
> My conlang Sohlob is another example of a head-initial language with a
> head-final ancestor. S case endings arose from PS postpositions which were
> themselves often case inflected nominals. PS had case but most PS case
> endings were lost by phonetic attrition. A main exception is the ergative
> ending which arose from an instrumental ending plus a discourse or topic
> marker and the accusative (S is split-ergative-accusative) which arose from
> the same marker without an instr ending. On top of this S has developed
> some prepositions from verbs.
Besides compounds, I see that classifiers (typically used with numbers)
usually precede nouns.
wam-ṛom-ṛal-ğom laz-a
64-8-2-(classifier) year-GEN
74 years
dim-śav-ļed-na kirja-n
new-many-(classifier)-DAT school-DAT
newly-many (an unprecedented number of) schools
Maybe Proto-Jardic had more words (like adjectives) that preceded nouns,
and other instances of head-final syntax. Adjectives and prepositions in
Jarda are actually forms of verbs. As Jarda came to use verbs more, and
the few remaining postpositions ended up as case suffixes, it started
looking more like a head-initial language.
It seems like there has to be an intermediate mixed stage. I don't see a
language going from Japanese-style SOV to Hawaiian-style VSO overnight.
So an SVO word order with postpositions might be best for Proto-Jardic.
(Finnish and Chinese have postpositions, so those might be good examples.)
Messages in this topic (15)
________________________________________________________________________
2c. Re: Proto-Jardic noun morphology
Posted by: "BPJ" [email protected]
Date: Sat Feb 16, 2013 4:11 am ((PST))
I'd rather think that you'd go from SOV to VSO by some sort of fronting
process. It kind of feels that you would need that to go from SOV to SVO as
well and then once again to go to VSO. I might have read something about
the subject, so I'll take a look in my stash. Remember that languages with
morphological marking of participants tend to be rather more free with
constituent order on that level, allowing O or in particular V to be
fronted for emphasis. Shame to say I haven't much contemplated *why* Sohlob
went from SOV to VSO but only assumed that it had to do with the change
from hierarchical to split ergative. Also since PS verbs were marked for
both participants it was a complete clause in itself which probably
facilitated fronting. Googling "basic word order shift" and "basic word
order change seems to bring up quite a lot. I'm disinclined to open pdfs on
my phone though.
Den lördagen den 16:e februari 2013 skrev Herman Miller:
> On 2/15/2013 5:42 PM, BPJ wrote:
>
>
> I second the suggestions that the various case endings arose from an
>> amalgam of gender markers, case markers and postpositions.
>> My conlang Sohlob is another example of a head-initial language with a
>> head-final ancestor. S case endings arose from PS postpositions which were
>> themselves often case inflected nominals. PS had case but most PS case
>> endings were lost by phonetic attrition. A main exception is the ergative
>> ending which arose from an instrumental ending plus a discourse or topic
>> marker and the accusative (S is split-ergative-accusative) which arose
>> from
>> the same marker without an instr ending. On top of this S has developed
>> some prepositions from verbs.
>>
>
> Besides compounds, I see that classifiers (typically used with numbers)
> usually precede nouns.
>
> wam-ṛom-ṛal-ğom laz-a
> 64-8-2-(classifier) year-GEN
> 74 years
>
> dim-śav-ļed-na kirja-n
> new-many-(classifier)-DAT school-DAT
> newly-many (an unprecedented number of) schools
>
> Maybe Proto-Jardic had more words (like adjectives) that preceded nouns,
> and other instances of head-final syntax. Adjectives and prepositions in
> Jarda are actually forms of verbs. As Jarda came to use verbs more, and the
> few remaining postpositions ended up as case suffixes, it started looking
> more like a head-initial language.
>
> It seems like there has to be an intermediate mixed stage. I don't see a
> language going from Japanese-style SOV to Hawaiian-style VSO overnight. So
> an SVO word order with postpositions might be best for Proto-Jardic.
> (Finnish and Chinese have postpositions, so those might be good examples.)
>
Messages in this topic (15)
________________________________________________________________________
2d. Re: Proto-Jardic noun morphology
Posted by: "Roman Rausch" [email protected]
Date: Sat Feb 16, 2013 7:24 am ((PST))
>It seems like there has to be an intermediate mixed stage. I don't see a
>language going from Japanese-style SOV to Hawaiian-style VSO overnight.
In particular situations, one can also find VO or VS in Japanese as a sort of
clefting techinque, or when adding an afterthought. Something like:
_Meiwaku da, ano ko wa
nuisance COP, that child TOP
'That child is a nuisance'
_Oboete inai, sono namae wo_
remember.GER PROG.NEG, that name ACC
'I don't remember that name'
My proto-language is also SOV and I intend to derive a VSO language out of it.
The first idea I had was to make this kind of construction increasingly more
common, until it flips over to VSO, but I don't know how realistic that is.
Messages in this topic (15)
________________________________________________________________________
2e. Re: Proto-Jardic noun morphology
Posted by: "Herman Miller" [email protected]
Date: Sat Feb 16, 2013 7:46 am ((PST))
On 2/16/2013 7:11 AM, BPJ wrote:
> I'd rather think that you'd go from SOV to VSO by some sort of fronting
> process. It kind of feels that you would need that to go from SOV to SVO as
> well and then once again to go to VSO. I might have read something about
> the subject, so I'll take a look in my stash. Remember that languages with
> morphological marking of participants tend to be rather more free with
> constituent order on that level, allowing O or in particular V to be
> fronted for emphasis. Shame to say I haven't much contemplated *why* Sohlob
> went from SOV to VSO but only assumed that it had to do with the change
> from hierarchical to split ergative. Also since PS verbs were marked for
> both participants it was a complete clause in itself which probably
> facilitated fronting. Googling "basic word order shift" and "basic word
> order change seems to bring up quite a lot. I'm disinclined to open pdfs on
> my phone though.
Jarda can have SVO or OVS word orders by fronting the subject or the
object, so maybe Proto-Jardic had basic SVO order and allowed V to be
fronted. (It's a little harder to see how a verb-final language could
switch to SVO, but I don't need to go back that far for now.)
http://www.prismnet.com/~hmiller/lang/Jarda/nouns.html
ğjub-en ķitṛa-ja ģundi (VSO)
catch-PERF bat-ERG moth-ABS
"the bat caught a moth"
ğjuben ģundi ķitṛaja (VOS)
ķitṛaja ğjuben ģundi (SVO)
ģundi ğjuben ķitraja (OVS)
(I see also that "ķitṛaja" on the web page contradicts the "ķitṛala"
that I've got in my Jarda documentation. I don't know which one is more
recent, but "ķitṛaja" sounds more like the correct form. Maybe it could
be a dialectal variation.)
Messages in this topic (15)
________________________________________________________________________
2f. Re: Proto-Jardic noun morphology
Posted by: "Jörg Rhiemeier" [email protected]
Date: Sat Feb 16, 2013 8:46 am ((PST))
Hallo conlangers!
On Saturday 16 February 2013 13:11:15 BPJ wrote:
> I'd rather think that you'd go from SOV to VSO by some sort of fronting
> process. It kind of feels that you would need that to go from SOV to SVO as
> well and then once again to go to VSO. I might have read something about
> the subject, so I'll take a look in my stash. Remember that languages with
> morphological marking of participants tend to be rather more free with
> constituent order on that level, allowing O or in particular V to be
> fronted for emphasis. Shame to say I haven't much contemplated *why* Sohlob
> went from SOV to VSO but only assumed that it had to do with the change
> from hierarchical to split ergative. Also since PS verbs were marked for
> both participants it was a complete clause in itself which probably
> facilitated fronting.
The shift from SOV to VSO in Albic seems to have happened via an
appositional construction. Like Proto-Sohlob, Proto-Hesperic
marked verbs for both participants, as Old Albic still does.
A sentence like
(1) Imethasa o ndaro em sarem.
AOR-kiss-3SG:P-3SG:A the:M.AGT man.AGT the:F-OBJ woman-OBJ
would thus have to be translated as 'He kissed her, the man, the
woman'. Such appositional constructions would then have been
generalized.
The continental Hesperic languages, of which none has preserved
object marking on verbs except for some relic formations, either
stayed SOV, developed a V2 rule, or became SVO (the latter perhaps
through generalization of the V2 rule). Also, their morphosyntactic
alignments changed, either to nominative-accusative (with some IE-
like quirks such as a syncretism of nominative and accusative in the
neuter gender) or some kind of split ergativity. (PH was, like Old
Albic, active-stative.)
--
... brought to you by the Weeping Elf
http://www.joerg-rhiemeier.de/Conlang/index.html
"Bêsel asa Éam, a Éam atha cvanthal a cvanth atha Éamal." - SiM 1:1
Messages in this topic (15)
________________________________________________________________________
2g. Re: Proto-Jardic noun morphology
Posted by: "Roger Mills" [email protected]
Date: Sat Feb 16, 2013 9:22 am ((PST))
--- On Sat, 2/16/13, Herman Miller <[email protected]> wrote:
Jarda can have SVO or OVS word orders by fronting the subject or the object, so
maybe Proto-Jardic had basic SVO order and allowed V to be fronted. (It's a
little harder to see how a verb-final language could switch to SVO, but I don't
need to go back that far for now.)
ğjub-en ķitṛa-ja ģundi (VSO)
catch-PERF bat-ERG moth-ABS
"the bat caught a moth"
ğjuben ģundi ķitṛaja (VOS)
ķitṛaja ğjuben ģundi (SVO)
ģundi ğjuben ķitraja (OVS)
-=========================================
Kash can also have all three orders in a sentence like this, though SVO is the
"normal" or default order. Is SVO, the O is the new info; VOS suggests that S
is the new info. and there wouold be an intonation break between VO and S, OVS
(or also OSV) is how they "passivize" it. It could further be made into a cleft
sentence (it was a moth that the bat caught) by adding "na ya" -- O na ya,
[inton. break] VS/SV.
Let me say generally-- I've been saving a lot of this thread, because it
contains interesting ideas for developments from Proto-Kash, and probably also
Proto-Gwr, which I'm still cogitating.
(I see also that "ķitṛaja" on the web page contradicts the "ķitṛala" that I've
got in my Jarda documentation. I don't know which one is more recent, but
"ķitṛaja" sounds more like the correct form. Maybe it could be a dialectal
variation.)
Messages in this topic (15)
------------------------------------------------------------------------
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/
------------------------------------------------------------------------
