|Submitter Email:||click here to access email|
In LINGUIST List 10-776, I posted a query concerning a person marking
system which (to me) seemed very odd. In Cubeo (a so-called "middle"
Tucanoan language of Colombia), in one of the two past tenses the
first and second person marking switches between declaratives and
interrogatives. Specifically, first person suffixes in declaratives
are used as second person suffixes in interrogatives, and vice versa.
I asked if anyone had run into this in other languages.
To my surprise, such a person marking system is attested (although it is of
course far from common). Thanks for their responses to Doris Payne
(dlpayne@OREGON.UOREGON.EDU), Claire Hiscock (firstname.lastname@example.org), Adrian Clynes
(email@example.com), Jackson T.-S. Sun (firstname.lastname@example.org), Richard
(= Dick) Hudson (email@example.com), Connie Dickinson
(firstname.lastname@example.org), Ernest McCarus (email@example.com), James L.
Fidelholtz (firstname.lastname@example.org), Stefan Georg (Georg@home.ivm.de), Norvin
Richards (email@example.com), Arthur Holmer (firstname.lastname@example.org),
Michael Cysouw (email@example.com), Wolfgang Schulze
(W.Schulze@mail.lrz-muenchen.de), Timothy Jowan Curnow
(Timothy.Curnow@anu.edu.au), John Peterson
A variety of terms have been used to refer to this person marking
system, including "conjunct-disjunct" (Hale), "self-person
vs.other-person", and "mirativity." (The latter term covers a broader
range of things as well, roughly having to do with old vs. new
Note: the 'Hale' in the above paragraph was mentioned by several respondents,
some of whom referred to Ken Hale (MIT) and others of whom referred to Austin
Hale (SIL). Based on a web search, I believe it is the latter, although I have
not actually seen any of the papers in question.
Some specific comments (slightly edited):
[This marking system has been found in] Lhasa Tibetan, Kathmandu Newari,
some Loloish languages, and probably some further Tibeto-Burman
languages of the Himalayan region. It is also found in a few Mongolic
languages bordering on Tibetan (Monguor and Baoan), where it
was recently discovered by Keith Slater in his 1998 dissertation (Minhe
Mangghuer: a Mixed Language of the Inner Asian Frontier) and, ahem, by
myself in a forthcoming paper on Tibeto-Mongolic language contacts.
[It] has been dubbed the "conjunct/discunct" system by Hale
(for Newari) and its chief explorer, Scott DeLancey. Of his
numerous papers on the matter, one may mention "The Historical Status of
the Conjunct/Disjunct Pattern in Tibeto-Burman", Acta Linguistica
Hafniensia 25/1992, 39-62, and "Mirativity: New vs. Assimilated Knowledge
as a Semantic and Grammatical Category", Linguistic Typology 1,1/1997.
[The suffix marking 1st person in declarative and 2nd person in
might mean] 'source of information/authority', which is the speaker in
declaratives and the addressee in interrogatives.
I recall having seen reports of similar data in the Tibeto-Burman
language Akha spoken in northern Thailand... As I recall,
sentence-final particles have one form if the subject is 1st person in
declaratives and 2nd person in interrogatives, and another form if the
subject is 2nd person in declaratives and 1st person in interrogatives.
These particles also serve to mark evidentiality.
...there was an interesting article by Scott Delancey in
Linguistic Typology, 1:33-52 on Mirativity, which deals with issues
like this...as I recall it, there is a general distinction between 1st
person declaratives and 2nd person questions on the one hand, and
everything else on the other, which many people trace back to
agentivity, expected information, etc. For example, if I do something,
then I am aware of it (generally), whereas when you do something, you
(but not me) will be aware of your having done it (i.e., conciously
having done it). Thus, when I ask you, I use the same form which I use
for myself when making a statement. This has just become
grammaticalized in some languages.
[There is] a good parallel from a Darwga diaclect (Central East Cauacsian)
called Meheb, cf.
nu quli-w le-w-ra
I:ABS house-LOC be:PRES-CL:I-1:Sg
"I am at home"
nu quli-w le-w-u
I:ABS house-LOC be:PRES-CL:I-Q
"Am I at home?"
h/u quli-w le-w
you:SG:ABS house-LOC be:PRES-CL:I
"you (sg.) are at home."
h/u quli-w le-w-ra(-u)
you:SG:ABS house-LOC be:PRES-CL:I-2Sg(-Q)
"Are you at home?"
Here, the morpheme -ra indicates 1Sg in declaratives, but 2Sg in
Ika (Chibchan, Colombia), [described] by an SIL grammarian,
Paul Frank. I do not have his works right here, so I only
can give an account from my own (scanty) notes.... There are
auxiliaries used for Past marking, and one of them has the
following paradigm (this is only part of the complete paradigm):
immediate past past
1 uwin ukuin
2 ukuin uzin
3 vwin uzin
(The orthography is not correct, due to ASCII limitation! The data come
from Frank['s dissertation] 1985, page 89.) [There was more data in the
msg, but I've edited out of this summary for reasons of space--MM]
The situation reminds me of something I read about British English back
in the 50's: a question uses the form of "shall" or "will" expected in the
Shall you attend the meeting?
Yes, I shall.
--Ernest N. McCarus
In addition to the languages mentioned above, Adrian Clynes reports a similar
system in Balinese (Austronesian), described in his 1995 PhD thesis. Claire
Hiscock has been told that for Brazilian Portuguese "in informal sitiations, the
suffixes marking first and second person were the same in exchanges like 'Are
you coming?' 'Yes, I'm coming.' Connie Dickinson is writing a paper for CLS on
this kind of 1/2 person declarative/ interrogative split in Tsafiki (= Colorado,
a Barbacoan language of Ecuador). Timothy Jowan Curnow describes such a person
marking system in an unpublished paper (originally a chapter of his
dissertation) for Awa Pit (= Cuaiquer), another Barbacoan language of Ecuador
and Colombia. Connie Dickinson reports that Cha'paalachi (= Cayapa of Ecuador,
the third living language of the Barbacoan family) and Guambiano (sometimes
called Barbacoan, but of doubtful affinity, IMHO) have similar systems. I have
inquiries out to some of the field linguists I know who work in these languages
for data, but haven't heard back yet. The Barbacoan languages are not known to
be related to Cubeo, but there might have been an areal influence. (Cubeo is
atypical of the other languages in the Tucanoan family in several respects.
Also, the area is well-known for intermarriage between people of different
Several diachronic sources for such a marking system were suggested, among them
re-analysis of mirativity marking or of evidentiality systems, and folk
etymology (resulting more or less from the same sort of mistake we linguists
sometimes make when trying to elicit 1st and 2nd person forms). In the case of
Cubeo, there is a robust system of evidentials, so I suspect that is the source.
(And the evidentials behave oddly in interrogatives of the recent past tense,
perhaps evidence--sorry about the pun--for this hypothesis.)
Summer Institute of Linguistics
|Original Query:||Read original query|
Sums main page