Re: Agency and the lexicon, and Modexúr Olo

----- Original Message -----
From: "Nathaniel G. Lew" <natlew@...>
To: <CONLANG@...>
Sent: Saturday, October 19, 2002 11:28 AM
Subject: Re: Agency and the lexicon, and Modexúr Olo


> >
> >The Bantu connection is not deliberate but welcome, as it puts me in
> familiar waters.  What do y'all think of this system?  does anyone know of
> any langs, con- or nat-, which have a similar system?
> >
>
> (Re: Your subj=AGENT / subj=EXPERIENCER distinction that has semantic
> results for the meaning of the predicate)
>
> (Sound of one blowing one's own horn.)
>
> Bendeh does something similar with markings on the verb.  Every predicate
> is either essentially transitive or intransitive in the lexicon, but verb
> markings differ according to transitivity (j- prefixes for transitive and
> p- prefixes for intransitive).  Transitive predicates inflected with
> intransitive p- prefixes become neuter, and intransitive predicates
> inflected with transitive j- prefixes become causative.  This process is
> completely productive.
>
In fact I think your system is a much closer parallel to the Bantu/Shona
system than mine.  The really interesting thing about your system is the
lack of default valency. How do you order your lexicon to account for the
meanings of each root?  I've taken to multiple entries: It makes it easier
to search for the proper word, but takes up a lot of space.  For your second
example I'm using a very odd structure that amounts to "Anyone who was able
saw the house".  I doubt this will remain as the standard way to say such
things, but I haven't worked out a proper system for reducing valency. For
an increase in valency like your fourth example, I use the verb
'du'='cause', which is also subject to the agent/experiencer distinction
(perhaps wrongly).


>    Fym jelzil ecus.
>    mother TRANS-past-see OBJ-house
>    Mother saw (could see) the house.
Shona: Amai        va-ka-ona    imba
            c2.Mother c2-Prf-see  c9.house (transitive verb used without
verbal extension)
            Mother has seen the house (c2, c9 indicate class/gender markers)

Modexúr: Bu-ka    mo        i      bur
                  see-Prf  mother Det  house
                  Mother saw/could see the house

>
>    Cus pelzil.
>    house INTRANS-past-see
>    The house was visible.
Shona:    imba        ya-ka-on-eka
                c9.house c9-Prf-see-NEUT (transitive verb used as an
intransitive stative verb, valency decreased by 1)
                The house was visible

Modexúr:    bu  su     [i      inse-be]   bur
                    see Past  [Det able-Rel]  house
                    Whoever was able saw (could see) the house

I strongly dislike the last construction, but it is analogous to the way the
passive is (sort of almost) formed.  Perhaps the answer is to embrace the
pseudo-passive:

    bu  inse su    [o-be]         bur
    see able Past [exist-Rel] bur.
    Those who exist were able to see the house/whoever was there could see
the house.

That seems to me a much more natural solution.  I must say at this point I
envy your intransitivity marker.  mind if I pinch it?

>
> (zil, "see," is inherently transitive.)
>
>    Zak pelzur.
>    story INTRANS-past-good
>    The story was good.
Shona:    ngano    i-no-naka
               c9.story c9.-Prg-good (intransitive stative verb with no
extension)

Modexúr:    iyo    i      wabról
                    good Det story
                    The story is good
>
>    Fym jelzur ezak.
>    mother TRANS-past-good story
>    Mother improved the story.
Shona:    Amai          va-ka-nak-isa            ngano
               c2.mother  c2-Prf-good-CAUS  c9.story (causative extension
used to "transitivize" a stative verb)
               Mother goodens (makes better) the story.

Moxedúr:    Mo       du      iyo    i      wabról
                    mother cause good Det  story
                    Mother causes (the story is good)

Alternately

                    Mo       du      [ris           [i     iyo     [i
wabról-(za)]]]
                    Mother cause [be.more [Det good [Det story-(Gen)]]]
                    Mother makes the goodness of the story be more.


>
> (zur, "good," is inherently intransitive.)
>
> Obviously, in Bendeh, the results are less interesting in your project,
> because the semantic alternation is mechanical or "logical".  What
> particular appeals to me about your idea is that the pairs of meanings are
> linked metaphorically/psychologically.
>
> I think that this is a result of your careful attempt to find pairs of
> meanings where the SUBJ="agent" and the SUBJ="experiencer" are the same.
> At the same time, perhaps unwittingly, but crucially, you keep the
> predicate either transitive or intransitive in both uses.  This forces you
> to come up with interestingly-related pairs of meanings.
This distinction historically derives from ergative constructions.  The word
orders were S-Vt-O and Vi-S, like so:

    i nan kanto
    I eat greens
    I eat greens

    brokas I
    strong  I
    I'm strong

Eventually sentences like these became fused in cause/effect relationships
(among other similar structures), which led to all intransitive verbs
becoming transitive verbs of causation:

    [i nan]   kanto   brokas  i
    [I eat]    greens  strong  I
    I eat greens and it makes me strong> greens strengthen me

At the same time, Intransitive verbs were occuring in relative clauses,
which suffixed them with -go (from the verb "to be", which in-turn became
re-interpreted as intransitives

     i brokas-go
     I strong-be
     I who am strong

    brokas-go i
    strong-be  I
    I am [the one who is] strong

Which leads to modern Modexúr:

    bróyasko e
    strong      I
    I am strong

    i kando bruya [xa be nan i]
    Det1 greens strengthen [Det2 Rel eat Det1]
    Greens make one strong (strengthen those who eat them)

So, as you can see, similar transitive/intransitive pairs and triplets (and,
conceivably, quadruplets) were created in this way, among them (A=agent,
E=experiencer, S=subject, Vi=intransitive verb, Vt=transitive verb)

ápar -  correct sth, right sth.  (Vt, S=A/E)
áhwayo -  polite, behaved  (Vi, S=A)
áhwayo -  proper, correct  (Vi, S=E)

bú  -  loosen  (Vt, S=A/E)
búwayo  -  loose  (Vi, S=A/E)

búde  - cut  (Vt, S=A/E)
bújo  -  sharp  (Vi, S=E)

brís  -  create, make  (Vt, S=A/E)
brísko  -  be born  (Vi, S=E)

íne  -  throw up  (Vt, S=A/E)
ínyo -    high  (Vi, S=E)

gáf  -  sicken, wilt  (Vt, S=A/E)
gáfayo  -  ill  (Vi, S=E)

ímos  -  change sth.  (Vt, S=A/E)
músko -  change oneself  (Vi, S=A)
músko  - change, become different  (Vi, S=E)

káyo  -  dry (fish), smoke (meat)  (Vt, S=A/E)
káyo  -  dry oneself off  (Vi, S=A)
káyo  -  dry  (Vi, S=E)

kíkal  -  frighten, scare  (Vt, S=A/E)
kíklayo  -  careful, cautious   (Vi, S=A)
kíklayo  -  afraid  (Vi, S=E)

kúme  -  discipline so.  (Vt, S=A)
kúmesko  -  polite  (Vi, S=A/E)

kúbe  -  benefit sth.  (Vt, S=A/E)
kúvyo  -  plenty, bountiful  (Vi, S=A/E)

lú  -  worry so.  (Vt, S=A/E)
lúyo  -  lukewarm, clammy  (Vi, S=A/E)

líjas -  improve, fix  (Vt, S=A/E)
líjasko  -  beautiful  (Vi, S=A/E)

mínde  -  put down  (Vt, S=A/E)
mínjo  -  sit, be seated  (Vi, S=A/E)

núr  -  bend  (Vt, S=A/E)
núgro  -  bend oneself  (Vi, S=A)
núgro  -  flexible, bend  (Vi, S=E)

skíl  -  crawl  (Vt, S=A)
skílgo  -  crouch down  (Vi, S=A)
skílgo  -  crouching  (Vi, S=E)

tá  -  increase  (Vt, S=A/E)
táyo  -  big  (Vi, S=A/E)

tímb  -  wash (clothes)  (Vt, S=A)
tímbayo  -  wet  (Vi, S=A/E)

vá  -  start sth.  (Vt, S=A/E)
vásko  -  begin (vi.)  (Vi, S=E)

zán  -  cool sth. down  (Vt, S=A/E)
zángo  -  impartial  (Vi, S=A)
zángo  -  cold  (Vi, S=E)



>
> Note, therefore, that in your last examples with nugro, "bend," you
> implicitly shift the OBJ NP in the "agent" version ("I bend the stick")
> into the EXP NP position in the "experiener" version ("the stick is
> flexible"), and shift the meaning from transitive to intransitive.  This
> is precisely parallel to the Bendeh system, but is a small departure from
> the earlier examples.  I am curious - what would a
> transitive "experiencer" version of nugro mean?
I didn't provide a sentence "I bend the stick"  My sentences were:

>e nugro-ka  sodize     e  lubo
>I  bend-Prf  and.then  I   run
>I bend/stretch before I run
>
>mur-me  nugro  imar-o
>wind-in   bend   tree-Pl
>Trees bend in the wind
>
>nugro i      gis   nemar-o olo  wam'  kayo  si    i
>bend  Det thin  stick-Pl   this  if        dry    not Det
>These thin sticks are (probably) flexible if they are not dry.
Although  a better translation for the last sentence is probably "If these
sticks aren't dry, they (probably) bend."

'I bend the stick' would be either:

e  nur-me     i     nemar
I  bend-Prg  Det stick
I'm bending the stick (deliberately, e.g. I pick it up and bend it with my
own two hands)

nur-me    e  i     nemar
bend-Prg I  Det stick
I bend the stick (unwittingly, as if I were to step on it in the course of
doing something else)

It almost seems to me that Modexúr has a lexified version of the Bendeh
system, and that Proto-Modexúr has a lot in common with Bendeh on the matter
of valency change.  I'm anxious to see whether or not I will be able to work
a sort of intransitivity marker into Modexúr

>
> By the way, can your distinction be applied to all predicates, or only a
> select set?  Might you might reserve the true agent/experiencer
> distinction for predicates of feeling, emotion, human experience, etc.,
> and use the same syntactic pattern for the transitive/neuter distinction
> on more physical predicates?  That is sort of what your nugro example
> implies.
Yes, the distinction is fully productive.  So-called "lower animals", in
fact, are always subject experiencers, e.g.

polafax  inze rizmer-o    majáfuwa-vo
retaliate can  starfish-Pl clam-Pl
Starfish can retaliate against clams

Theorhetically any verb can have a subject-agent or subject-experiencer,
though many verbs are hard to imagine as such.  How can one walk
unintentionally (perhaps in one's sleep).  Others can be used as such only
in the literary sense:

rison    i        raga            ti-z'         [xa     e-z'     ruyar]
answer Det1 expression you-Gen [Det2 I-Gen ask]
Your expression (unintentionally) answers my question.

  Funny you should mention verbs of emotion.  I've been finding it extremely
difficult to come up with agentive meanings for these verbs, e.g. how does
one really be jealous on purpose?
    As far as Bendeh is concerned, do you allow verbs to develop two
separate meanings based on the transitivity distinction, or do you make them
pretty much remain as one verb, effected by two affixes?

-Joe