Pablo wrote :
Mathias M. Lassailly wrote:
>
> >You just have to coin to affixes :
> >
> >one showing that the predicate uses an Aspective root verb (A)
> >
> >one showing that the predicate uses an Unaspective root noun (U)
> >
> >kjak-A (to bite) >< kjak-U (a bit)
> >
> >di-kjak-A : to bite >< di-kjak-U (to make a bit)
> >
> >It lokks maybe odd, but it's lintiguistically well known and VERY EASY :-)
>
> It is. But it's not necessary! In this case, for example,
> the context should tell you if the nominal form of kjak-
> means "the action of biting" or "the effects of biting".
Yes, you're right when the case is tagged.
However, when you tag only a theme 'case' (it's not a case ! Aaaaarrrggghhhhhh ! ),
then you may have :
zu-jol di-seat
'leg : bite'
Now Herman mayl think :
'as for the leg : it bites'
since to him 'kjak' is the verb 'to bite' (of course)
and Carlos will think
'as for the leg : it's a bite'
since to him 'kjak' is 'a bite'
of course he won't because 'context' is clear, but try with image', 'seat' and
any other result and unergative instrument, you'll see what I mean.
That's why I wrote all that, but ok ! Majority rules :-)
>
> >
> >Modifiers incorporate themselves inherently the 'case tag' :
> >
> >'a bitten (leg)' incorporates already the 'pe-' patientive as 'P' :
> >
> >pe-jol di-kjake > mu-kjake-P (a-jolo)
> >the leg is bitten > the bit-TEN leg
> >
> >'a biter (dog)' incorporates the 'R-' ergative as 'R' :
> >
> >a-frar pe-jol di-kjake > mu-kjake-R a-frar
> >the dog bites the leg > the bit-ING dog
>
> I agree with the idea of marking a modifier as ergative
> or patientive. This should be done by agglutinating the
> case prefixes
>
> muwakjako afraro [...]
> the biting dog (does something)
>
> where mu- "modifier", a- "agent" (separated by a glide -w-).
> Note gender agreement.
>
To me
muwakjako = mu-a-kjako = (made) by a bite
mu-kjak-a-o = biting
mu-kjak-pe-o = bitten
> >
> >The suffix -P/S is a way to make a phrase a modifier.
> >
> >Nouns integrate the case even deeper, in the unaspective form :
> >
> >*the bit-TEN : a-kjake-P
> >the bit-ER : a-kjake-R
> >
>
I wanted to say :
the bitten = kjak-pe-o
the biter = kjak-a-o
I put 'a-' case in front just to show it's a noun.
> apekjako "the bitten (one) [does something]"
> a- "agent" pe- "patient"
>
> that is a- [pe- kjak-] -o, you can change a- for pe- and make
>
> awakjako pepekjako dikjake
> "The biter bites the bitten one"
>
> where awakjako = a- [a- kjak-] -o, i. e. the agent of the biting
> is the agent of the (another) biting.
See above
>
> You can't take this too far tho. IMO we should have simple
> derivations like English -er, -or for agents and leave the
> case system alone.
Ok. I thought it would be easier to remember since it's an ergative language.
>
>
> [snip]
> >So both H & C would have NO PROBLEM using the same words in the same language, H thinking he's
> >deriving noun from verb and C thinking reversely
> >
> >BUT
> >
> >To do that with unergative instruments and results, you need do something special :
> >
> >H : to be the seat of >< the seat : C
> >
> >not : to sit on >/< the seat
> >
> >H : to be the fruit of >< the fruit : C
> >
> >not : to bear fruit >/< the fruit : C
> >
> >In other words, H must accept that the verbal root is deponent, not active when
> >C derives this verb
> >from unergative instruments and results
> >and C must accept that unergative instruments and results derive into deponent
> >verbs, not ergative
> >ones.
> >
> >I'm not inventing anything here.
> >
> >I'm describing the origin of unergative and antipassive forms in ergative
> >languages drifting towards a
> >nominative system.
> >
> >NOW, do you want to know why ?
> >
> >This is because C goes from UNASPECTIVE nouns into ASPECTIVE verbs
> >
> >while H goes from ASPECTIVE verb register into UNASPECTIVE noun register :
> >
> >H : 'to bite' (asp) > 'the bit' (unasp)
> >C : 'a bit' (unasp) > 'to bite' (asp)
> >
> >So you need TAG the U/A REGISTER on the ROOT to say
> >
> >'hey, I'm speaking of the noun embodying the action' (U)
> >
> >or
> >
> >'hey, I'm speaking of the action embodied by that noun' (A)
> >
> >The predicate di- alone CANNOT do that.
>
> I repeat my previous opinions: we don't need U/A tags because
> common sense when creating the word and context when using it
common sense make Japanese derive 'image' into 'to appear' and Europeans into 'to see'.
> should tell you which one is it. If you want to make things clear,
> write glosses like these:
>
> XXXX- "to sit, a seat"
> YYYY- "a fruit, to give fruit"
to me :
XXX : 'to be used as a seat'
YYY- 'to grow as a fruit'
>
> If you want to say "the action of sitting", use pred-XXXX "to sit"
Then you can't have the chain :
mieru > miru > miseru > misaseru
because you need the first verb to be deponent like 'to be apparent', 'to be the seat
of' and 'to grow as'.
> plus the case tag corresponding to it as a nominal root. For example
>
> "Sitting here is not allowed"
> (here) theme-pred-to_sit neg-pred-allow.passive
>
> "allow" would be glossed "to allow, a permission"
>
Why not 'to be allowed' from noun root 'allowance' ?
>
> >NOW you don't give any example for subclauses, and that's unfortunately the main
> >reason why SOV is SO
> >difficult for Europeans to speak :
> >
> >'the dog who-m I painted red bites me'
> >
> >Then you have :
> >
> >a-si di-wiv-ul a-fraro pe-si di-kjake
>
> I would say
>
> a asi pefraro diwivul apiko pesi dikjake
>
> The first "a" tells you there's another a-prefixed pronoun later
> (in a-pik-o, where pik- is a "relative pronoun").
>
That's great. I like it, and I would like others would understand that.
>
> >I told PABLO about 'resumptive pronouns' used in spoken Japanese, but actually
> >he managed something
> >different :
> >
> >We mean : the head the subclause is like the theme of a main clausE :
> >
> >ti = who/which = the latter one
> >
> >ki = who/which = the former one
> >
> >'ti' and 'ki' both mean 'the dog' (fraro) in the two following examples :
> >
> >a-fraro pe-ti a-si di-wivul pe-si di-kjake
> >
> >'the dog who-m I painted red bites me'
> >
> >fraro pe-ti a-si di-wivul a-ki pe-si di-kjake
> >
> >'the dog who-m I painted red, he bites me'
> >
>
> I like this one. Shouldn't it be _afraro peti ..._
> so that it agrees with _aki_?
Ok. I mean, YOU found that one (and you know it :-). But I feel nobody else but me
understood. I tried explain, but I think you would be better at it.
>
> a-fraro [a-si pe-ti di-wivul] a-ki pe-si di-kjake
> | ^
> | |
> \___________________________/
well, then the same problem as in Japanese arises because you could understand
a-fraro [a-si pe-ti di-wivul] pe-si di-kjake
as
a-fraro a-si [pe-ti di-wivul] pe-si di-kjake
I'm afraid you need start the subclause with the pronoun to show : 'hey, here is a
new phrase starting !'
>
> If this is OK, I'd drop my separable case tags
> proposal.
>
> But here's another idea: why not use two relativizing
> morphemes? Say gi- to prefix the phrase first, then
> gij- as a relative pronoun standing for it then.
>
> a-si pe-gi-fraro diwivul a-gij-o pe-si di-kjake
> "The dog-that I painted red, he bites me"
>
Very interesting. The problem is then that you must think of the subclause headed
with fraro before or at the same time as you mention it as patient of the main
subclause.
I think it's almost impossible at length.
I prefer your first idea. But why not ask the others ?
>
> (zu- = theme)
>
> [[a-si pe-gi-fraro diwivul] a-gij-o pe-si di-gi-kjake] a-si zu-gij-e di-kite
>
> [[The dog-that I painted red] it that-it-bites me] I see it
>
> Think of it as a pointer. The gi- prefix marks the "address" where you're
> going to find the phrase you're referring to. When you need it, you reference
> it by using gij-.
>
Same doubt as above.
>
> About the modifiers:
>
> mufraro zukjako diqaunul
> "The dog's bite was hard"
> (More like "referring to the dog, the bite was hard")
I read it :
'referring to the dog, as for the bite ON her it was hard' :-)
>
> If you see this as ambiguous, we could have as you proposed:
>
> muwafraro zudikjako diqaunul
> "The made-by-the-dog action-of-biting was hard"
I understand that. But I think you mix here adjectives and subclauses.
For adjectives I would be ok for :
ys mu-wa-fraro kjako diqaunul
'the dog's bite was hard'
For connective I would go gor
ys-kjako ti ya-fraro gi di-qaunul
the dog's biting was hard
>
> with the dog first marked as "agent", and the bite marked
> as the predicate (i. e. the action performed by the agent).
>
> Another (classical) example:
>
> "the fear of the enemy"
> mod-agent-enemy pred-fear "fear caused by the enemy"
> mod-undergoer-enemy pred-fear "fear felt by the enemy"
>
Yes.
>
>
> >I suggest two structures :
> >
> >one subclause structure with pronouns
> >
> >one modifier structure : adjectives mu- and adverbs su-
>
> I agree on the subclause structure (the way I proposed it :)
> I'm not sure if morphologically distinguishing between adjectives
> and adverbs is necessary. In the worst case, we could make
> nominal modifiers agree in gender, and verbal modifiers agree
> in tense or aspect. And there must be some kind of word order!
I'm not sure I could learn all that agreement stuff.
Easy languages made me lazy.
You'll have a bad pupil here.
>
>
> To everybody: I think it's time that we decide on some basic
> issues, namely gender, evidence, tense, and aspect. I'll restate
> my proposal with a slight modification:
>
>
> Gender (for "nouns" and maybe agreeing modifiers)
>
> person (sentient being)
> animate (animal or vegetal)
> thing (physically perceivable)
> concept (an action or abstract object)
>
That's for nouns, isn't it ?
>
> Evidence (for all PoS's)
>
> actually perceived (you saw it/listened to it/etc.)
> indirectly perceived (i. e. by its effects)
> hearsay (they told you about it)
> hypothetical (you don't imply it's real or it happened)
That's for verbs isn't it ?
>
> These would be suffixed. Gender would be compulsory,
> not evidence (and evidence would go first,
> root + [evid_tag] + gender_tag)
>
>
> --Pablo Flores
>
Mathias
-----
See the original message at http://www.egroups.com/list/conlang/?start=17187
--
Free e-mail group hosting at http://www.eGroups.com/