Se describe la sintaxis pronominal según la jerarquía de las personas que existe en las lenguas de la gran familia Mawetí-Guaraní.1 El enfoque nuevo resulta de la interpretación de las formas mismas que se usan en la sintaxis pronominal: Cuando se trata de 1,2,3 > 3 observamos la sintaxis normal, es decir que el verbo transitivo recibe el prefijo personal de los verbos activos y el objeto pronominal aparece después del verbo si es un objeto animado; el objeto pronominal inanimado se omite. Sin embargo, cuando se trata de 3 > 1,2 o 2 > 1, la sintaxis cambia radicalmente, omitiéndose la marca personal sujeto del verbo transitivo. Lo que en inglés es el pronombre de objeto (por ejemplo, en "she saw me" o "you called me"), aparece en Mawetí-Guaraní delante del verbo, en forma de pronombre personal (o de marca personal/posesiva de los nombres). Yo interpreto esta mudanza como un cambio de verbo activo con objeto a una construcción existencial: en lugar de decir "ella me vio", la mayoría de las lenguas mawetí-guaraníes dice "hubo vista mía por parte de ella". En mbyá, por ejemplo: a'e xer-exa 3.pron 1sg.n rel-ver/vista Este ejemplo es especialmente significativo porque contiene la construcción típicamente nominal con el elemento rel de la flexión relacional tan característica de estas lenguas. Debido a la distinción difusa entre radicales verbales y radicales nominales en mawetí-guaraní, el cambio ("switching") de una clase de palabras a la otra es fácil.

In this contribution I analyze, in the languages of the South-American Tupi-Guarani family, the syntax of what are object pronouns in English or Spanish. The description of word classes in Tupi-Guarani languages with its fuzzy distinction between nouns and verbs is the basis of the word class switching we observe in Tupi-Guarani pronominal syntax. The existing hierarchy 1 > 2 > 3 is shown by active syntax in the case of 1,2,3 > 3 and 1 > 2. Inanimate 3 p objects are included in the active verb forms and are not expressed by a special object pronoun. However, if the personal objects are higher than the subject (3 > 1,2 or 2 > 1), we observe a switching from active syntax with verbal person marking for subject function to nominal existential constructions. There is no more any active verbal form marked by a verbal person marker, but a nominal predicative construction characterized by nominal determination devices. What is an object pronoun in English (“she saw me”, “you called me”) appears as a Tupi-Guarani personal pronoun which determines the noun-verb (“possessive” construction, such as “my seeing” = “seeing me”). Instead of saying “she saw me” speakers of most Tupi-Guarani languages say “there was her seeing me”. Many basic “verbs” behave as nouns whenusing relative inflection. For instance in Mbyá: a'e xe r-exa 3.pron 1sg.n rel- seeing,usually translated by ‘he/she saw me’. The translation, however, being pragmatically correct, does not allow understanding the structural meaning of the Mbyá expression.

In this contribution nearly 200 lexical items of Aché, an understudied Tupi-Guarani language of Paraguay, are identified as of Tupi-Guarani (TG) origin and compared with their cognates in 20 other TG languages. The comparison not only shows that the lexicon of Aché is basically TG - no other linguistic elements have been identified until now -, but also allows for establishing aspects of the historical phonology of Aché. Comparing the phonological system of Aché with that of other TG languages makes clear that the closest contact languages are Mbyá and Xetá. Lexical and phonological comparison, however, call for the hypothesis that Aché has in its history undergone more than one intense phase of linguistic contact, probably with the Yuki-Siriono cluster of Bolivia, with Mbyá, but also with classical and modern Paraguayan Guarani.

En este artículo se identifican unos 200 lexemas del aché (familia Tupí-Guaraní) con sus cognados en 20 de las demás lenguas de la familia. Este rastreo lexicológico no sólo muestra que la base del léxico aché es fundamentalmente tupí-guaraní (TG) - hasta ahora no se encontraron elementos de otra lengua no TG - sino que permite también elaborar aspectos de la fonología histórica del aché. Comparando la fonología del aché con la de las demás lenguas TG, se ve que las lenguas más cercanasson el mbyá y el xetá. Resulta de la comparación lexicológica y fonológica que el aché probablemente ha tenido en la historia de su formación contactos con varias lenguas TG como son el grupo siriono-yuki de Bolivia, el mbyá y/o xetá, el guaraní clásico y, en la actualidad, el guaraní paraguayo moderno.

Hoje conhecemos dez dicionários de língua geral amazônica, seis deles não publicados. Em comparação, a língua geral paulista continua pouco documentada. Estudam-se seis dicionários de língua geral amazônica setecentistas, em especial três manuscritos ainda não publicados. Da comparação desses dicionários, resulta a identificação de Anselm Eckart como autor de um e de Anton Meisterburg como possível autor de outro manuscrito. Analisa-se a interdependência de Eckart, Meisterburg e da "Prosodia de Lingua". Descrevem-se mudanças ocorridas na LGA e discutem-se caraterísticas da LGP. Tudo isto serve para rever os problemas téoricos do conceito de "língua geral" proposto por Aryon Rodrigues e problemas da história da sua formação.

"Língua Geral" ('General Language') was a language of indigenous origin (Tupinambá) spoken by mestizos on the coast of Brazil. Rodrigues' definition of the term and his distinction of Southern and Amazonian "Língua Geral" are discussed in the light of recent linguistic publications and of existing and newly discovered dictionary manuscripts of Amazonian "Língua Geral". The result of the analysis of three of those manuscripts is that Anselm Eckart is the author of one of them and Anton Meisterburg probably the author of another one. Both have a third one (Prosodia de Língua) as one of their main sources. The study is completed by the description of diachronic change in Amazonian "Lingua Geral", as well as by the discussion of some characteristics of the scarcely documented Southern "Língua Geral".