Re: new parts of speech/cases

Garrett Jones <alkaline@...> wrote:
 >>>
 > i guess you called "patient" and "focus" after Rick Morneau?
yeah i did. :) btw do you know any other "real" linguists that use the same
system he does?
<<<
no, i don't. but RM doesn't say that his 3-case system is how actors work in
the real world's linguistics. he clearly states that his "invention" is a
way to devise a workable verbal system analysable by people speaking very
different natlangs. my feeling is that his system is handy but completely
misses the point with important kinds of predicates like "to fill a tank
with water", "to plant a garden with flowers", etc, for which it is hardly
easy to tell the focus from the patient. these verbs are the ones which
imply a spatial, temporal or notional preexisting or postexisting relation
between the actors of the verb. example: "the tank is full of water", "the
garden is planted with flowers". the roles of "patient" and "focus" don't
work well for these verbs: you need introduce other attributive roles such
as the situative SIT, the possessive POSS, the locative LOC etc.

however, i've also noticed that indonesian has a specific "passive" voice
whose subject can only be a focus:

Kuncing melihat tikus.
Cat sees mouse.
The cat can see the mouse.

Tikus kelihatan kuncing.
Mouse vision cat.
The mouse can be seen by the cat ("is the vision of the cat")

In the above example, "mouse" is a focus.

Kuncing memakan tikus.
cat eat mouse.
The cat eats the mouse.

Tikus dimakan kuncing.
mouse eaten cat.
The mouse is eaten by the cat.

In the above example, the "mouse" is a patient.
but you can also say "Tikus dilihat kuncing", thus doing as if the focus
"tikus" were a patient.

But Rick Morneau says that in "the cat eats the mouse", the "mouse" is a
focus because "mouse" is the "pattern" of eating. He does so because he is
embarassed because the cat is a patient as well according to his definition
(the "patient" is the guy "experiencing the state focused at"), so he gets
two patients--an "active" one (the cat) and a "passive" one (the mouse) and
he tries to devise a solution by "demoting" the "passive patient" into a
focus.

By doing so, Rick Morneau tries to neutralize what french linguists call
"the direction of the relation between an entity and its behaviour": at the
sentence level, the entity is the base actor (let's call it improperly
"subject") and the behaviour is the predicate (let's call it wrongly
"verb"). There are "attributive" relations between the subject and the verb
called "endocentric" because the verb "focuses inwards on the subject's
state" and "active" relations called "exocentric" because they focuses on an
object's state. "The cat eats the mouse" is endocentric because the
direction is really from the mouse to the cat, or said in other words, the
state focused at is the cat's state, not the mouse's state. "The cat eats at
the mouse" is exocentric because it focuses on the mouse's painful change of
state. Rick Morneau clearly states that some verbs focuse on the state of
their subject and others more on the state of their object, but he doesn't
want to tag the two kinds of verb differently, so he found an easy solution:
(i) since "to eat" may be either transitive or intransitive and (ii) the
subject of the intransitive verb is tagged as a patient (iii), then the
subject of such verb should keep being tagged as a patient when it's
transitivized and (iv) the other patient of the verb (the mouse) is tagged
as a focus.
I find the solution pretty cool, but at the same time there's little chance
that a natlang uses the same trick.

>>>
my focus-object doesn't include addressee or result, those are both under
the patient-object.
<<<
RM tags the result as a focus. he states that there is a difference between
(i) verbs whose focus is the pattern of the performance like "song" in "i
sing a song", and (ii) verbs whose focus is a result created but the action
like "flour" in "i grind up flour", but he says he can tag both result and
pattern with the same tag because they are mutually exclusive, i.e. there's
no verb of which two core actors are a result and a pattern: you don't say
"i sing a song up into something" or "i grind up flour something".
RM doesn't tag the addressee very clearly.

>>>
I am going to re-tag the role of the paper in the
sentence you gave as "instrument/medium", which is a third case. A minyeva
sentence can contain multiple instances of any case.
i = (case) patient-object
e = (case) instrument/medium
bo = (causijunction) distant recipient/benefactor
the sentence would look like so:
le fedu  i da kleta  e mlukiqa bo i te  vlenu.
I  write - a  letter - paper   -- - the man.
<<<
using Rick Morneau's system, i would say:
le fedu u da kleta i mlukiqa bo i te vlenu.
this because the letter is a result or a pattern of action and therefore is
tagged as a focus.
the unfortunate paper is one actor undergoing a change of state from a blank
page into a written document.
but "medium" sounds cool too. there are passive mediums (paper), active
instruments (pen) and materials, facilities, recipients, vehicles, etc. ;-)
i'm kidding you. only the core cases of the verb need be crammed under a
tag.

>>>
I suppose since causijunctions always have the patient-object case after
them, the 'i' wouldn't necessarily have to be said. Here are the different
possibilities of objects the 'fedu' verb can take
(first the cases)
u = (case) focus-object
a = (case) location
[snip examples]
<<<
your causijunctors' objects may be either focus or patient:
commitative: with, along (focus i guess)
adversive: against, from ("protect from", "fight against"): focus? patient?

>>>
-ge = (suffix) verb promoter (state -> action)
zena le u te  dine
read I  - the book
le zenage i te  vlenu
I  read - the man
I read to the man
<<<
ok, so basically you promote the patient "I" into an agent directing the
action of reading towards a new patient "man" while the book keeps being a
focus. in other words, you make an endocentric verb "to read a book" into an
exocentric verb "to read aloud to someone". i do exactly the same in my
conlang by suffixing "to" to the verb in order to focus to another actor
than the focus: "i read-to the man the book".

>>>
marking the book as the focus makes it the mental focus, but also allows it
to be a 'patient pivot' as far as the causation string goes, so that a
change of state can be specified for it.
le zenage mlusi u te dine i te vlenu.
i read, giving the man a book injury.
le zenage u te dine mlusi i te vlenu
I read the book, injuring the man.
yeah, my system is on crack. Think it's workable? Could you come up with any
other hard examples for my system to handle? The challenge would be
appreciated :)
<<<
is "le" automatically considered the agent of "mlusi" just because "vlenu"
is tagged as a patient?
what in the case that the two core actors of the verb are a patient and a
focus with no agent like in "to smell", "to know", etc?
example:
le zenage u te dine coni i te vlenu
I read the book, known by the man.

i think tagging "the book" as the focus actor of "to read" and "to injure"
simultaneously is not workable. you've got that in english sentences like:
"i called him an idiot" where "him" is the patient of "to call" as well as
of "to be an idiot", but since "idiot" may itself be analysed as the focus
of "to call", you could say that "him" is the patient of "to call" in
relation to the focus "idiot".

mathias

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp.

Reply