Capanahua, an Amazonian language from the Pano family spoken in Loreto, Peru , has intriguing syntactic phenomena in the area of clause linking. In particular, Capanahua has several paradigms of morphemes that are used when two clauses are linked. These paradigms mark notions such as argument coreference (so-called switch-reference), sequence of tense, transitivity concord, and others.
On additional fact makes the study of Capanahua a priority: the language is on the verge of extinction, with less than 300 speakers left. Description and documentation of its linguistic features thus becomes an essential task. In addition to the publication of theoretical results, the current project will emphasize the preservation and dissemination of the linguistic materials gathered during fieldwork.
2. Capanahua clause linking
Capanahua has at least 7 morpheme slots for verbal suffixes (cf. Loos 1998). The first slot includes thematic morphemes such as reflexives, reciprocals, causatives, benefactives, intransitivizers, etc. Slot 2 has one type of negation; morphemes that indicate movement appear in slot 3; slot 4 contains aspectual morphemes, slot 5, person, number and tense. Slot 6 incorporates several kinds of morphemes: validation markers, switch-reference (SR) morphemes, subordinators and interrogative markers. Slot 7 is reserved for negation.
Switch-reference markers, which appear in slot 6, encode two notions: coreference relations between arguments in different clauses and sequence of events (cf. (1) from Loos 1999). For example, the SR marker -a§bo in (1) indicates not only that the subject of the know-clause is the same as the object of the tell-clause, but also that the event of knowing precedes the event of telling. Other SR markers indicate different coreference relations as well as different sequence-of-event relations.
(1) hati?ibi hato yo?i-wv
honan-a?bo
all
them tell-imper know-sr
ATell all of
them so they will know@
A second set of slot 6 morphemes involves several peripheral morphemes which are sensitive to the transitive/intransitive nature of the verb (cf. Loos (1998, 1999: 236)). For example, Loos (1999: 236) states that A[s]ubordinate clauses in which the action is semantically prior in time to their main clauses are marked according to the transitivity of the main clause.@ This connection between argument structure and sequence-of-tense is intriguing, and it extends to certain event-oriented adverbs like rapidly, abruply, completely and subject-oriented locative phrases. All of these items have variants that depend on the transitivity of the verb, as shown in (2), also from Loos (1999:239).
(2) a. mapv-ka?in-i
go.up-rapidly(intr)-conj(intr)
AHe goes
up (the bank) rapidly@
b. bi?-ba?in-kin
grap-rapidly(tr)-conj(tr)
AHe grabs
it abruptly@
The third set of slot 6 morphemes refers to evidentiality or validation. These morphemes provide information with respect either the source of the information given by the speaker (first hand, second hand, hearsay, etc.) or the speaker=s attitude towards that source of information (Aeverything indicates something is true, but I don=t believe it@).
3. Current documentation on Capanahua
The literature on Capanahua, and Pano in general, stems mostly from materials gathered by Summer Institute of Linguistics. There is some work on Capanahua phonology (cf. Loos 1969, 1986), and some brief sketches of the grammar (cf. Loos 1998, 1999), as well as some other linguistic materials (Loos B. and Loos, E. 1973, Loos B. 1980, Loos E. 1963, 1973a, b, 1976, 1986). Additionally, it is generally assumed that languages in the Pano family (which includes, among others languages such as Shipibo, Mayoruna and Yaminawa., with larger populations) are linguistically very close, to the extent that discoveries from one can extend to others. Descriptions of the SR paradigm and transtivity concord for Pano languages include Loos (1998, 1999) for Capanahua; Loriot et al (1993), Valenzuela 1999, 2000 for Shipibo, Prost and Prost (1962) for Chacobo, among a few others.
4. Goals of the project
Our project has the following goals:
As mentioned earlier, this project builds on existing descriptions and on fieldwork that will be undertaken during the summer of 2001, by José Camacho, Alberto Elías and Enrique Espinoza. During this preliminary phase of the project, we will set up a field base in the town of Requena, about one day by boat from the Capanahua area. The Capanahua are located in the region of Loreto, Peru, along two rivers: the Rio Blanco and the Tapiche river (see http://amazonas.rds.org.co/MAPAS/Reg/Lor/BaseLD6.htm, areas marked 1, 2 and 3). The total number of members of this ethnic group is not known, but estimates range from 267 (http://amazonas.rds.org.co/atlas/cap2/Capanahu.htm) to 400 (Loos 1999: 229) . It is also not known how many of them speak the language. However, given the total population, it is safe to assume that Capanahuas is on the verge of extinction.
Following advice from Eugene Loos, the most knowledgeable researcher on Capanahua, we will visit the area of Limon Cocha on the Tapiche River, Loreto (Peru), the most accessible location where it is likely to find speakers. A preliminary sociolinguistic survey will assess the current status of Capanahua, number of speakers in the area, and we will distribute a basic questionnaire on the language that focuses on word order, negation, tense morphology, relativization, subordination, evidentiality, SR, noun incorporation, possessives, among other topics. Based on the existing literature, we have determined that these are important areas in the grammar of Capanahua. In addition, we will distribute a questionnaire on coordination, an area of grammar which accordig to Loos (1968), has some interesting property. Given the connection noted between SR and coordination, we expect some preliminary data to bring important insights about precise directions of the following field work phases.
6. Time line for the second phase of the project)
We plan to disseminate the results of our work at three distinct levels. First, as mentioned earlier, stories will be collected in a small booklet to be presented to the communities and the informants whose knowledge will be essential for this project. Second, the codified data and the stories will be made available online (information will be posted in this web page). See also Camacho and Elías (2001)