First Steps with Caquinte

I am writing to you from Sepahua, in the Department of Ucayali in Peru, where I have come for a few days of internet-fed work and relaxation after three and a half weeks in Kitepámpani, a Caquinte (Kampan Arawak) community in the headwaters of the Mipaya River in the Department of Cuzco. On Friday or Saturday I will return to Kitepámpani for a little more than three weeks of work, before making my way to Lima for a conference on nominalization at the Pontificia Universidad Católica del Perú (PUCP). I arrived in Pucallpa the evening of June 24, where I met and stayed with Ken and Joy Swift, members of the Summer Institute of Linguistics (SIL) who have worked with Caquintes since 1975 to, among many other things, translated the entirety of the New Testament into the language. Prior to my fieldwork this summer, Ken's 1988 MA thesis -- primarily a survey of nominal and verbal morphology in the language -- was the sole linguistic work on the language. Hung up by the lack of an available seat on a float plane to Sepahua, I flew instead to Atalaya on July 1, and took a boat along the Urubamba River to Sepahua the following day.

North American Float Plane
Pucallpa-Atalaya

Sepahua is a frontier town of some 3,000 residents located on the border of the departments of Ucayali and Cuzco at the confluence of the Sepahua and Urubamba rivers. It has grown substantially in the last 15 years due to increased oil and natural gas exploration and the migration of (primarily) Panoan peoples from the headwaters regions to the east. This has resulted in a central area concentrated around the port, the market, and the Catholic mission, surrounded by a series of barrios dominated by particular ethnic groups that lay further afield. Many of the now older adults of these barrios were already grown when they first entered into sustainted contact with outsiders. This, combined with an influx of Peruvians from the highlands in search of work (e.g., in the lumber industry), results in a diverse environment that in many ways barely hangs together. For example, non-native residents cannot own land, because the entire city occupies a legal limbo in which the terrain is titled as a comunidad nativa (native community) but federally it is both a district and provincial capital.

I spent two nights in Sepahua provisioning for everything from rice to a surge protector, and left on the Shao-Ling for Nuevo Mundo on July 3, some four and a half hours further up the Urubamba. These boats (like the one from Atalaya) are usually in the range of 75 feet longand powered by two outboard motors in the rear. Passengers typically sit in cheap seats removed from cars and screwed to the boat deck, in other makeshift seats, or on the floor. They are also responsible for the transport of commercial goods between here and Camisea, where the local route terminates.

Nuevo Mundo is a Matsigenka community that is now attached to a large base controlled by the Spanish conglomerate Repsol. It is an odd juxtaposition of, on the one hand, outside oil and gas workers, and, on the other, a community that is still Matsigenka-dominant. There I stayed one night at the newly expanded Hostal Max (three floors!), watching Fourth of July celebrations the following day on DirectTV. From Nuevo Mundo I radioed Kitepámpani, and left for the approximately four-and-a-half-hour canoe-ride up the Mipaya River on July 4.

Repsol Base at Nuevo Mundo (Urubamba River)

Having entered the Mipaya, you are now climbing into the various headwaters that feed the Urubamba. From the Mipaya proper you enter the Agueni River, and at the confluence of the Agueni and Yori rivers lies Kitepámpani. These rivers are increasingly rocky and shallow, and you have to get out of the canoe and haul it over some dozen rocky patches in the river during the course of your ascent. The Andean foothills come more and more into view, and when you arrive, the view is stunning. (Unfortunately the internet is so slow that photos will have to wait.)

Kitepámpani was first settled in 1975 when Caquintes were eager to have missionary-linguists from the SIL be able to land a float plane in their territory. Prior to this, and for some time afterwards, Caquintes, like other Kampan Arawak peoples, lived mainly in single-family settlements scattered across large swaths of forest. Historical Caquinte territory, however, actually lies over the mountain range that separates the Urubamba and Tambo drainages. Here they lived in the headwaters of the Pogueni River -- on which the current Caquinte community of Tsoroja is still located -- until approximately 1959, when a particularly deadly attack by neighboring (and related) Asháninkas, forced them to scatter. (Asháninkas, the militarily dominant indigenous group in this region, had even prior to this period consistently conducted raids on Caquinte settlements, kidnapping Caquinte women to sell in places such as Atalaya and Pucallpa.) I am still working out the details of subsequent migrations, but at least one group of Caquintes crossed the mountains, ultimately arriving in the Picha River -- another tributary of the Urubamba upriver of the Mipaya -- and settled in the Matsigenka community of Puerto Huallana. In the early 1970s this group moved to a settlement across from Nueva Luz, the largest Matsigenka settlement between Sepahua and Nuevo Mundo, and some four years after that they moved to a settlement on the lower Huiriticaya the closest left-bank tributary of the Urubamba downriver of Nuevo Mundo. Separately, and somewhat later in the mid-1970s, another group of Caquintes made its way from the upper Pogueni, via a series of small headwaters tributaries in the Urubamba drainage (the Shampavireni and Tireni), ultimately forming part of the group that founded Kitepámpani in 1975. Although Kitepámpani was titled in 1983, all but two of the families that came to settle it eventually returned to the Pogueni River, and the current residents of Kitepámpani represent yet another migration from the Pogueni after this date. At the time of this second migration, some of the Caquintes that had settled on the lower Urubamba (see above) also came to live in Kitepámpani.

Confluence of the Ageni & Yori Rivers
Mipaya River Headwaters

These recent migrations and subsequent marriage patterns have resulted in, among other things, a linguistic situation in which the two most recent generations of Caquinte-speakers speak a language that is, at least lexically, influenced in spontaneous speech either by Matsigenka (for those that came to the Urubamba drainage) or by Asháninka (for those who remained in the Pogueni). Many current residents of Kitepámpani, for example, are the children of mixed Matsigenka-Caquinte marriages, or have two Caquinte parents but a Matsigenka spouse. While all adults are easily able to differentiate between what is Matsigenka and what is Caquinte, acquiring a "pure" Caquinte by listening to my surroundings is nearly impossible, which has severely complicated one of three main prongs of my work this summer.

Kitepámpani, Echarate, La Convención, Cusco, Perú

That work has been divided between: 1) spending time with residents in an effort to learn to speak Caquinte (although social integration has been more difficult than I had hoped due to the fact that it is mainly women in the community that I need to be working with for linguistic reasons); 2) building a lexical database (currently at over 1,500 headword entries); and 3) developing a text corpus from which I can begin more detailed linguistic analysis. Despite major social changes in the last five years due to Repsol's exploration in the region (which has resulted in, among other things, cement homes and running water -- see later blog post), many residents remain Caquinte-dominant. Some men -- including the schoolteacher, who was born there and is the only local Caquinte to have attained a university degree -- speak Spanish fluently (but some speak primarily Matsigenka and not Caquinte); some women understand Spanish quite well, but only one speaks it fluently. This has resulted in a workflow in which I collect texts from one of three women over three days and then work with the schoolteacher, Miguel Sergio Salazar, to translate them into Spanish on the fourth day. While I can rely on my lexical knowledge of Matsigenka somewhat (see here for work that I have carried out on a large Matsigenka text corpus written by Haroldo Vargas Pereira and collected by Christine Beier and Lev Michael) Caquinte and Matsigenka are most divergent lexically (i.e., and not in their grammar, in which they are remarkably similar), e.g. (Matsigenka:Caquinte): panko:tsovironaki 'house'; tsamairentsi:nigankitehi 'swidden'; shima:pamakaviri 'fish'; tsinane:mankigarentsi 'woman'; sekatsi:aintochapaki 'manioc'. You can imagine how such differences were a significant obstacle to my communication at the outset.

Yori River

Another significant obstacle to my communication has been the nature of the grammar of Kampan Arawak languages itself. That is (for the linguistically inclined), these languages are highly polysynthetic, head-marking languages with typically only one postposition that encodes either location at or motion towards. A non-core NP with any other thematic role needs to be licensed by one of a set of approximately ten applicative suffixes on the verb. I have had many encounters in which I find myself quite proud to have constructed a well-formed verb -- with all necessary categories like person, aspect, reality status, and a host of modal categories -- only to realize I left off the requisite applicative, and so by the time I get to the applied object it is completely uninterpretable and I have to start all over!

I plan to continue the same kind of work when I return in August, and at the end of that month I will write a series of blog posts on Caquinte history and their current interactions with oil and natural gas companies and the significant social changes that those interactions have brought about in the last five to ten years.