Search

Manambu, a Ndu language from the East Sepik Province of Papua New Guinea, has a complex system of marking positive and negative imperatives. Imperative is the main means of marking [...]

Yalaku, a previously undescribed Ndu language from the East Sepik Province of Papua New Guinea, has two core cases (nominative and accusative) and four oblique clausal cases (locative-instrumental, allative, dative, [...]

Case markers are thought of primarily as nominal morphemes, indicating the function of a noun phrase in a clause. In a few languages of the world case markers also appear on verbal forms. Such 'versatile' cases can express (i) temporal, causal and other relationships between clauses, and (ii) aspectual and modal meanings within a clause. Core cases tend to express aspectual and modal meanings, while oblique cases tend to be used as clause-linkers. The recurrent semantic differences between case morphemes as nominal markers, as clause-linking devices, and as exponents of clausal categories are rooted in the inherent polyfunctionality of these 'chameleon' morphemes: the specific meaning of any instance is affected by the morphosyntactic context in which it occurs. The conclusions are corroborated by a case study of Manambu, a Papuan language with extensive use of cases on nouns and on verbs, as exponents of aspectual and modal meanings and as clause-linking devices.

The Sepik River Basin in New Guinea is a locus of substantial linguistic diversity, with several genetically related and unrelated languages in continuous contact. The inhabitants of the area divide into 'River-dwellers' (i.e., those who live on the Sepik River) and 'Jungle-dwellers' (i.e., those who live in the bush). The two groups differ in their ways of subsistence, their knowledge of each other's languages, and the impact of language contact. This article focuses on Manambu, a language of the Ndu family spoken by a warlike group of River-dwellers, and the ways its grammar has been influenced by the languages of the neighboring Jungle-dwellers, the Kwoma and the Yessan-Mayo. Lexical influence from the closely related Iatmul (also spoken by River-dwellers) is restricted to a number of ritual genres (now obsolete). Patterns of interaction between Jungle-dwellers and River-dwellers and the effects of language contact in the Middle Sepik are compared to the situation in the multilingual Vaupes area in northwest Amazonia. Different means of subsistence, life styles, and patterns of interaction are responsible for differences in contact-induced change in the two eases.

Borrowing, or diffusion, of grammatical categories in language contact is not a unitary process. In the linguistic area of the Vaupes in northwest Amazonia, several different mechanisms help create new contact-induced morphology. Languages which are in continuous contact belong to the genetically unrelated East-Tucanoan and Arawak families. There is a strong cultural inhibition against borrowing forms of any sort (grammatical or lexical). Language contact in the multilingual Vaupes linguistic area has resulted in the development of similar--though far from identical--grammatical structures. In Tariana, an Arawak language spoken in the area, reanalysis and reinterpretation of existing categories takes place when diffusion involves restructuring a pre-existing category for which there is a slot in the structure, such as case. A new grammatical category with no pre-existing slots may evolve via grammaticalization of a free morpheme--this is how aspect and aktionsart marking was developed. The development of a five-term tense-evidentiality paradigm involves a combination of strategies: reanalysis with reinterpretation accounts for the obligatory tense marking, and the history of visual, inferred and reported evidentials. The nonvisual evidential evolved via grammaticalization of a lexical verb while the most recent, assumed, evidential involves reanalysis and reinterpretation of an aspect marker and grammatical accommodation.

This article discusses the way 'correct' and 'incorrect' language uses are rationalized by Tariana speakers in the linguistic area of the Vaupes River basin in Brazil. This area is known for its institutionalized multilingualism due to linguistic exogamy operating between the Tariana and speakers of a number of languages belonging to the East Tucano subgroup of Tucano family. There is a strong constraint against language mixing in Tariana. This constraint operates predominantly against loan forms and items that contain Tucano-like sounds. A few morphosyntactic constructions calqued from East Tucano languages are also identified as 'incorrect' Tariana. An additional mechanism, which helps determine what is 'correct' and what is not, is constant reference to the way in which representatives of older generations speak.

Serial verbs in Tariana, a dialect of North Arawak, suggest valid defining properties cross languages. Coexistence of such verbs in one language supports the concept of iconic motivation.

The South American Indian languages Warekena and Bare of the Upper Rio Negro area, Brazil, have a number of structural similarities regarding the marking of different prosodic domains. These domains include the h-prosody, the vowel nazalization and dipthongization, and the vowel harmony triggered by the glottal fricative h. A feature designating a word in one language may designate a phrase in another. Morphemes and syllables depend on each other. Pausal forms marking the boundaries of phrases and utterances are the languages' most unusual characteristic.

An over view of the Vaupes area is given. The principles of choosing a language and use of different languages in varied circumstances are discussed.

This article describes an unusual and complicated system of genders and classifiers in Palikur, a northern Arawak language spoken in Brazil and French Guiana. Palikur has three genders (masculine, neuter, and feminine); gender assignment is based on a combination of semantic features (humanness, animacy, size, and shape). There are two or three gender choices depending on construction type. There are also four distinct types of classifiers: numeral classifiers, verbal classifiers (with two subsets--those occurring on stative verbs, which are frequently used as modifiers in noun phrases, and those occurring on transitive verbs), locative classifiers (used as adpositions), and possessive classifiers (generic nouns used in possessive constructions with certain alienably possessed nouns). Different noun classification devices have different functions and scope; all, except possessive classifiers, overlap in their semantics. Classifiers provide cross-categorization of nouns and help the language to structure concepts. Throughout the article, Palikur gender and classifiers are placed in typological perspective.

Research focusing on eight grammatical systems and their cooccurrence within grammars has led to the creation of a dependency hierarchy. This hierarchy indicates that systems linked with the predicate argument (person, reference classification and number) depend on the systems associated with the predicate (tense, aspect and evidentiality) and on that linked with the clause as a whole (polarity). Systems linked with the predicate of the clause (tense, aspect and evidentiality) depend on that associated with the whole clause (polarity).