In this article we describe and develop an optimality-theoretic (OT) analysis of foot-level (secondary) and word-level (primary) stress in Nanti, a Kampa language of Peru. The distribution of stress in Nanti is sensitive to rhythmic factors, syllable quantity, vowel quality, and to whether a syllable is open or closed. The interaction of these independent variables produces a complex, multigrade stress scale married to an iterative stress system whose default preference is alternating, iambic rhythm. While each of the interacting factors in this system is familiar to phonologists, Nanti is special because the particular combination of influences and factors in Nanti contributes to a complexity of interactions that has not been documented in any other language to date.

This paper presents an internal classification of Tupí-Guaraní based on a Bayesian phylogenetic analysis of lexical data from 30 Tupí-Guaraní languages and 2 non-Tupí-Guaraní Tupian languages, Awetí and Mawé. A Bayesian phylogenetic analysis using a generalized binary cognate gain and loss model was carried out on a character table based on the binary coding of cognate sets, which were formed with attention to semantic shift. The classification shows greater internal structure than previous ones, but is congruent with them in several ways. RESUMEN:Este artículo presenta una clasificación interna de la familia Tupí-Guaraní en base a datos léxicos de 30 lenguas tupí-guaraníes y 2 lenguas no tupí-guaraníes, awetí y mawé. Un análisis filogenético que utiliza el modelo Bayesiano de ganancia y pérdida de cognado binario generalizado se llevó a cabo en una tabla de carácteres basado en la codificación binaria de los cognados afines, que se formaron con atención al cambio semántico. La clasificación muestra una estructura interna mayor que las anteriores, pero que resulta congruente con ellas de varias maneras. PALABRAS CLAVE: Lenguas amazónicas; Tupí-Guaraní; Filogenética; Clasificación de lenguas.* Affiliations for the authors of this paper are: Chousou-Polydouri-Laboratoire Dynamique Du Langage, cnrs, Donnelly, Michael, OʼHagan-university of caLifornia, berkeLey; Meira-Museu Paraense emíLio goeLDi; Bartolomei, Wauters-inDePenDent schoLar.

Key Words Amazonia, language change, linguistic area, discourse area, ethnography of speaking ■ Abstract In indigenous lowland South America there are several discourse forms and processes that are shared by groups of people of distinct genetic linguistic affiliations; this leads us to posit this large region, which we label greater Amazonia, as a discourse area, a concept that parallels the notion of linguistic area. The discourse forms and processes we examine are ceremonial dialogue, dialogical performance, templatic ratifying, echo speech, ceremonial greeting, ritual wailing, evidentiality, speech reporting practices, parallelism, special languages, and shamanistic language use. We hypothesize that in lowland South America, discourse is the matrix for linguistic diffusion, i.e., that linguistic areas emerge within discourse areas. What we propose then is a discoursecentered approach to language change and history, parallel to a discourse-centered approach to language structure and use. Our survey includes a plea for a careful archiving of recorded and written materials dealing with lowland South American discourse.

TitleThe Nanti reality status system: Implications for the typological validity of the realis/irrealis contrast The Nanti reality status system: Implications for the typological validity of the realis/irrealis contrast Lev Michael University of California, BerkeleyAbstract This paper describes the reality status system of Nanti (Arawak) and argues that it constitutes an instance of a canonical reality status system. The relevance of such a system is examined in the light of literature that casts doubt on the typological validity of reality status as cross-linguistic grammatical category. It is shown that reality status in Nanti is an obligatory inflectional category, and that the distribution of realis and irrealis marking across Nanti construction types hews closely to expectations based on a notional understanding of the 'realis' and 'irrealis' categories grounded in a contrast between 'realized' and 'unrealized' situations. It is also shown that the Nanti reality status system does not exhibit evidence of being based, either synchronically or diachronically, on semantically narrower notions that could account for the distribution of reality status marking in the language, without recourse to the more generalized notions of realized and unrealized events. It is suggested that the Nanti reality status system might serve as a suitable canonical system around which a canonical typology of reality status might be built.

Cabral (1995, 2007, 2011) and Cabral and Rodrigues (2003) established that Kokama and Omagua, closely-related indigenous languages spoken in Peruvian and Brazilian Amazonia, emerged as the result of intense language contact between speakers of a Tupí-Guaraní language and speakers of non-Tupí-Guaraní languages. Cabral (1995Cabral ( , 2007 further argued that the language contact which led to the development of Kokama and Omagua transpired in the late 17th and early 18th centuries, in the Jesuit mission settlements located in the provincia de Maynas (corresponding roughly to modern northern Peruvian Amazonia). In this paper I argue that Omagua and Kokama were not the product of colonial-era language contact, but were rather the outcome of language contact in the Pre-Columbian period. I show that a close examination of 17th and 18th century missionary chronicles, Jesuit texts written in Omagua and Kokama, and modern data on these languages, make it clear that Omagua and Kokama already existed in a form similar to their modern forms by the time European missionaries arrived in Maynas in the 17th century. Moreover, I show that several key claims regarding ethnic mixing and Jesuit language policy that Cabral adduces in favor of a colonial-era origin for Kokama are not supported by the available historical materials. Ruling out a colonial-era origin for Omagua and Kokama, I conclude that Proto-OmaguaKokama, the parent language from which Omagua and Kokama derive, was a PreColumbian contact language.