Re: Person distinctions in languages?

On Tue, Feb 01, 2005 at 08:15:53PM -0800, Arthaey Angosii wrote:
> Emaelivpeith Steven Williams:
> > Are there any languages that break the pattern of
> > first-second-third-(fourth) person? I'm trying to
> > model a system that doesn't use such distinctions, but
> > I can't figure out how to make it coherent or
> > intelligible without perhaps a model to base it on.
>
> My conlang, Asha'ille, uses (long-term) emotional closeness between
> speakers to distinguish persons. Check out my description at:
>
>     http://arthaey.mine.nu:8080/~arthaey/conlang/grammar/persons.html
>
> Teoh's Ebisédian also has an intimate/non-intimate distinction, but
> I'll leave that to him to discuss. :)
[...]

Ebisédian's system is similar to Asha'ille's: there is a set of
singular 1st person pronouns, but no plurals. There are no 2nd/3rd
person pronouns, but instead a set of intimate and distant pronouns,
inflected for gender/number/case. The Ebisédian mode of thought is
that there is the speaker himself/herself, then a set of people whom
the speaker consider to be intimate (close friends, family, spouse,
associate), and then the set of everyone else (acquiantances,
strangers, etc.).

The intimate pronouns always refer to someone intimate with the
speaker, or at least someone the speaker considers to be "on my side".
The distant pronouns are used to refer to everyone else. (The distant
pronouns are also used for everyone in formal contexts, but we will
disregard them for the purposes of this discussion.) Note that this is
*independent* of who the speaker is addressing. Because of this,
Ebisédian has the unusual property that often the exact same sentence
can be spoken to different people, and they will each get a different
understanding of it, yet their interpretations are consistent with
each other.

Take for example, King A speaking with male servant B and his daughter
Princess C. King A may say something like this:

uso'   chi'd0           th0're jobu'            uro  3k3'.
please DIST:MASC:SG:ORG give   INTIM:FEM:SG:RCP this inscription:CVY

Now, if King A were addressing servant B, he would understand this to
mean that _chi'd0_ referred to him, and _jobu'_ referred to princess
C, and that he was to give the inscription to Princess C. If King A
were addressing Princess C, she would understand that _chi'd0_
referred to servant B, and _jobu'_ to herself, and that servant B was
about to give her the inscription.

In fact, the King could be addressing them both at the same time, and
each of them would understand correctly which pronoun was referring to
whom. The King could in fact be addressing the populace, and not
servant B or Princess C directly, and they would understand that he
wishes for servant B to hand the inscription to princess C. (The
singular distant masculine pronoun _chi'd0_ could not refer to the
populace themselves, you see; the King would use a plural epicene
distant pronoun for that.)

Now, I can hear people in the back objecting that this doesn't quite
work in all cases, and that there are many cases where it would be
ambiguous to whom the pronouns refer. And I agree, there are many
cases of ambiguity, and Ebisédian is highly context-dependent.
However, Ebisédian also has what are called "noun association markers"
which can be prefixed to pronouns to distinguish between different
people referred to by the same pronoun. This is sometimes employed to
resolve ambiguities that arise from the pronominal system.

There are 3 noun association markers, corresponding with the 3
prefixes ki-, ci-, and ro-. (In the most literal sense, they mean
"red", "green", and "blue", but so do many other grammatical morphemes
in Ebisédian, so a literal reading is not tenable. In any case, it is
useful to think of them as "coloring" pronouns differently in order to
disambiguate between different referents.) By convention, ki- is used
for the first pronoun referent, roughly meaning "the former", ci- the
second, roughly meaning "this one", and ro- the third, roughly
meaning "the other".

Let's see this at work in a conversation between a party of 4 men, who
are therefore considered to be "intimate" by each other.

Man A (speaking to Man B):
	ni      ghi'  di     iso'i    le's ki-cw'm3          moo'ju?
	REL:LOC what AUX:LOC time:LOC go   #1-INTIM:MASC:CVY city:RCP
	When will you [ki- hereby established to refer to Man B] go to
	the city?

Man B:	oro  ky'ri.
	next day:LOC
	Tomorrow.

Man C (nodding at Man D):
	a'ne     lyy's   iro  ro-cw'm3          zo  ki-cw'm3?
	INTERROG go:PERF EMPH #3-INTIM:MASC:CVY and #1-INTIM:MASC:CVY
	Will he [i.e., Man D, henceforth associated with ro-] go with
	him [i.e. Man B, associated with ki-] ?

(Note that Man C could have addressed this to anyone or everyone in
the party and they would get the correct interpretation.)

Man D:	my'e.
	Not_so
	No.

	eb3'    zo  uro  ro-cw'm3          zokyy' mangu'    loo'ru.
	1sp:CVY and this #3-INTIM:MASC:CVY follow horse:RCP countryside:RCP
	I will be with this-him [referring to man A - ro- cannot refer
	to the 1st person, so it is available to be assigned to man A
	when man D is speaking] following the horse into the countryside.

Man A:	ji'e. i're   ro-cw'm3          zo  eb3'.
	yes   indeed #3-INTIM:MASC:CVY and 1sp:CVY
	Yes. Indeed he [man D] is with me.

Man C:	ghe'     ebi'?
	what:ADV 1sp:LOC
	What about me?

Man A (replying to Man C):	b0'mu cu-cw'm3          uu'ri.
				wait  #2-INTIM:MASC:CVY here:LOC
				You wait here. (Or, "he remains here",
				to the other party members.)

Man B:	oso  cu-cw'm3          zo  eb3'.
	wish #2-INTIM:MASC:CVY and 1sp:CVY
	I wish him (or "you", if spoken to man C) to come with me.

Man A:	0so'.
	ought_to_be
	Let it be so.

So you can see that the noun association markers can be used to
disambiguate up to 4 parties referred to using the same pronoun (the
association tag referring to yourself is "free" for you to assign to
another party, since it is clear that the same tag cannot refer to
yourself when you're speaking). So although this conversation is very
contrived (in a more realistic setting, the men would be addressing
each other by name), it does work better than it could with, e.g., the
English pronominal system under the same circumstances (if you didn't
use names, you'd get very confused who "he" refers to after a few
lines into this conversation).


T

--
First Rule of History: History doesn't repeat itself -- historians merely
repeat each other.

Reply