On the classification of Taushiro
August 21, 2008
When in Lima I spent quite a bit of time in taxis, where I inevitably fall into discussions about what I am doing in Peru. When I explain that I am a linguist working on Peruvian Amazonian indigenous languages, the driver normally responds with a either puzzled silence or just assumes that I talking about Quechua. Either way, this gives me an opportunity to talk about the linguistic diversity of Peruvian Amazonia — which truly is impressive — the scale of which many a taxi driver has seemed to find quite amazing.
But when I was in Lima a few weeks ago, things went quite differently: I was in a taxi, and for the umpteenth time I was explaining what I was doing in Peru, and the taxi driver responded, without dropping a beat, “Oh, you mean like Taushiro” (except, of course, in Spanish). “Well, er, yes, exactly,” I fumbled, as I tried to assimilate the fact that the driver had just named one of the most obscure Peruvian Amazonian languages there is. I think the driver was somewhat alarmed at my disconcerted response and helpfully explained that there had recently been a television program that focused on the last speaker of Taushiro, a linguistic isolate spoken in Loreto. Now, for the benefit of all of us that missed the original program, Nila Vigil, over at Instituto Lingüístico de Invierno has made it available here. Check it out.
In relation to her post, Nila made a pair of comments here on Greater Blogazonia about the classification of Taushiro (here), which I now want to take up. Specifically, Nila remarks that the apparently incorrect classification of Taushiro as a Zaparoan language continues to be propagated in many linguistic and anthropological works. In fact, if one takes a closer look at how Taushiro has been treated by classifiers of Amazonian languages, as we shall do now, we find an excellent illustration of the kind of confusion and borderline linguistic malpractice that bedeviled Amazonian comparative linguistics until relatively recently.
As Nila notes, we can probably trace the modern-day prevalence of the incorrect classification of Taushiro as Zaparoan to Beuchat and Rivet’s (1908) influential article delimiting the Zaparoan family (available here via the amazing Persee site). For the most part, this article does an excellent job with the Herculean task of sorting through the truly vast number of colonial era ethnonyms and then delimiting the Zaparoan family. Their achievement is all the more impressive for the fact that they only had access to very scanty linguistic data, most of it of low quality. In fact, the only significant error in their classification, as far as I am aware, is their inclusion of Taushiro in the family.
A close reading of the article suggests that Beuchat and Rivet had no data on Taushiro, but instead based their classification on the word of colonial era writers who claimed that ‘Pinche’ (a colonial era name for Taushiro) is related to Andoa, which is quite unambiguously a Zaparoan language. In particular, Beuchat and Rivet seem to be indebted to Hervas (1800), a Spanish translation of Hervas (1784). Lorenzo Hervas’ Wikipedia biography here suggests that his linguistic methodology mainly consisted in working with Jesuits who had been ejected from the Americas after the 1767 order of Carlos III, and who were personally familiar with the languages of particular groups. An examination of Hervas’ classifications suggests that they were based at least as much on geographical proximity as on properly linguistic grounds.
In particular, Hervas (1784 : p. 61) ( the whole work is available here) enumerates the following languages as ‘dialects’ of Andoa: Arafo, Chudaviño, Gae, Murato, Pavo/Pabo, Pinche, Simigae, and Bobonazo. Of these, Andoa, Gae, Simigae and Bobonazo are clearly names for Zaparoan languages (as is evident by looking at Beuchat and Rivet (1908)), Murato is colonial era name for Candoshi, and Pinche is, of course, Taushiro. Hervas thus treats languages from three distinct language families as dialects of Andoa, thereby seeding, I believe, the belief that Taushiro is a Zaparoan language. The only obvious thing that these languages have in common is that all these languages were spoken on the Río Pastaza, or on the nearby Río Tigre. In other words, the languages appear to have been grouped together on the basis of geographical proximity.
The fact that colonial-era writers made an error about the classification of Taushiro is not surprising, but this error has proven to be quite resilient, as Nila remarked. Solís (2003: 200), for example, remarks:
La lengua taushiro … debe ser considerada como una rama divergente dentro de las lenguas que conforman el grupo Záparo.
[The Taushiro language ... should be considered a divergent branch among the languages that form the Zaparoan group.]
And Wise (1999: 312) remarks
Taushiro is possibly a Zaparo language, but its classification has not been confirmed.
And Pozzi-Escot (1998) simply classifies Taushiro as Zaparoan without any further comment. What I find remarkable about these classificatory claims is that they are not accompanied by any evidence to support them, nor do they cite anyone else who provides such evidence. In fact, I suspect that what has happened is that the idea of Taushiro as Zaparoan has simply gained legitimacy by having been repeated for so long, by so many people. How anyone who has actually looked at lexical and grammatical data on Taushiro could consider this language to be Zaparoan is beyond me. Here, do it yourself: thanks to the excellent SIL Peru site, you can download a Taushiro grammatical sketch (here) and a Taushiro vocabulary (here). Please write me if you find any plausible evidence of a Taushiro-Zaparoan connection. All the solidly Zaparoan languages exhibit numerous grammatical similarities and connections, and Taushiro doesn’t even come close. Of course, it may ultimately be the case that Taushiro is a Zaparoan language — it is certainly possible that there is a very distant relationship here — my point is simply that no one who has judged this hypothesis favorably has presented any evidence in its favor. In fact, according to Fabre (here (pdf)), Alicea (1976), who carried out fieldwork on the language, found no evidence to support a Zaparoan connection.
Well, so much for the Zaparoan classification of Taushiro — but the story does not end there! A common fate for Amazonian isolates is that they end up being included in a number of different language families by different classifiers, and Taushiro is no exception. Tovar (1961: 151) groups Pinche with Omurano, another unclassified language, and Loukotka (1968:156) groups Pinche with Candoshi, which is generally believed to be an isolate. Interestingly, if one looks at Loukotka (1968), the author is quite upfront that he has no sources or data on Pinche (Taushiro) at all! How Loukotka manages to classify Taushiro with Candoshi on the basis of no data at all is a mystery — but it’s certainly a nice trick. It’s also pretty clear that Tovar’s (1961) is also an immaculate classification. Note, by the way, that Hervás original classification groups Taushiro and ‘Murato’ (a colonial era name for Candoshi) together, so the ultimate responsibility for this classification may also be laid at Hervás’ feet.
Interestingly, Kaufman (1994) tentatively proposes that Taushiro belongs to a grouping that includes Candoshi and Omurano, in effect combining Tovar’s and Loukotka’s proposals. To his credit, Kaufman identifies one phonological correspondence between the three languages, lending at least a measure of credibility to the proposal. In my opinion, it will probably never be possible to arrive at any solid conclusion regarding the position of Omurano in this grouping, as the only data on this extinct language, of which I am aware, are modest wordlists in Tessman (1999 ) and Villarejo (1959), both of which are of dubious quality. However, the question of the genetic relationship between Candoshi and Taushiro seems fairly straightforward to settle, as Candoshi is a vital language, and obtaining substantial lexical and grammatical data should pose no insurmountable difficulties. In fact, trying to evaluate the Candoshi-Taushiro hypothesis with the available lexical materials (here) might even make a nice term paper or undergraduate thesis project.
To summarize, then, classifiers have grouped Taushiro with the Zaparoan family, with Candoshi, and with Omurano, the latter two of which are otherwise considered isolates in their own right by most present-day Amazonianists. The evidence for a Zaparoan connection is non-existent, and the evidence for either a Candoshi or an Omurano connection is thin, but in the case of Candoshi, may ultimately prove viable.
Alicea Ortiz, Neftalí. 1976. Apuntes sobre la cultura taushiro. DEL 56. ILV.
Campbell, Lyle. 1997. American Indian Languages. OUP.
Hervas, Lorenzo. 1800. Catálogo de las lenguas de las naciones conocidas, y numeración división y clase de éstas según la diversitad de sus idiomas y dialectos. Madrid: Librería de Ranz.
Hervas, Lorenzo. 1784. Catalogo Delle Lingue Conosciute E Notizia Della Loro Affinita’, E Diversita’. Cesena: Gregorio Biasini.
Kaufman, Terrence. 1994. The native languages of South America. In C. Mosley and R.E. Asher (Eds.), Atlas of the world’s languages. New York: Routledge: pp. 46-76.
Loukotka, Cestmir. 1968. Classification of South American Indian Languages. Los Angeles : Latin American Center, University of California
Pozzi-Escot. 1998. El Multilingüismo en el Perú. Cusco: Centro de Estudios Regionales Andinos “Bartolome de Las Casas.
Solís, Gustavo. 2003. Lenguas en la Amazonía Peruana. Lima:. Visual Service S.R.L.
Tovar, Antonio. 1961. Catálogo de las lenguas de América del Sur. Buenos Aires: Editorial Sudamericana.
Tessman, Gunther. 1999 . Los indígenas del Perú Nororiental. Quito: Abya Yala.
Villarejo, Avencio. 1959. La selva y el hombre. Lima: Ausonia.