Katharina Haude

Hierarchical Alignment in Movima

In Movima (unclassified, lowland Bolivia), the arguments of a transitive clause are encoded according to the position of their referents in an indexability hierarchy. The argument whose referent is lower in this hierarchy is encoded in the same way as the sole argument of an intransitive clause. This argument, furthermore, is syntactically privileged: it can be relativized and topicalized, while for the argument with the higher‐ranking referent to undergo one of these processes, a detransitivizing voice operation is used. Semantic role assignment is carried out by direct and inverse marking on the predicate. Movima represents a hitherto undescribed case of hierarchical alignment, in which an indexability hierarchy has direct impact on syntax and the less salient noun phrase has the privileged syntactic status.

Subordination in Native South American Languages

In terms of its linguistic and cultural make-up, the continent of South America provides linguists and anthropologists with a complex puzzle of language diversity. The continent teems with small language families and isolates, and even languages spoken in adjacent areas can be typologically vastly different from each other. This volume intends to provide a taste of the linguistic diversity found in South America within the area of clause subordination. The potential variety in the strategies that languages can use to encode subordinate events is enormous, yet there are clearly dominant patterns to be discerned: switch reference marking, clause chaining, nominalization, and verb serialization. The book also contributes to the continuing debate on the nature of syntactic complexity, as evidenced in subordination.

Argument Encoding in Movima: The Local Domain

Argument encoding in Movima transitive clauses is based on a referential hierarchy (1 > 2 > 3 topic > 3 nontopic). Verbal direct and inverse marking indicates the roles (actor/undergoer) of the event participants. The argument with the higher-ranking referent is obligatorily represented by a constituent directly attached to the predicate, while the argument with the lower-ranking referent is represented by a constituent less closely connected to the predicate, aligning with the single argument of the intransitive clause. First and second person can only be encoded in the first way and therefore do not show any alignment effect with the argument of the intransitive clause. However, pragmatic factors can override the person hierarchy: when, for pragmatic reasons, first and second person are expressed by a free pronoun in clause-initial topic position, they can be treated like lower-ranking persons.

Ergativity, valency and voice

This volume is a collection of articles concerned with the typology of valency and valence change in a large and diversified sample of languages that display ergative alignment in their grammar. The sample of languages represented in these descriptive contributions covers most of the geographical areas and linguistic families in which ergativity has been known to exist jointly with well-developed morphological voice, and some languages belonging to families in which ergativity or voice were not previously recognized or adequately described up to now.

Inverse and symmetrical voice: On languages with two transitive constructions

Abstract In voice and alignment typology, a categorical distinction is generally made between inverse systems on the one hand and symmetrical voice systems on the other. A major reason for distinguishing between these two types is the assumption that inverse systems are governed by a hierarchy involving grammatical, semantic, and ontological criteria, while symmetrical voice systems are based on discourse-pragmatic factors. However, the two types also have several important properties in common, in particular the fact that they have more than one nonderived transitive construction. Based on data from three native languages of South America, we show that the line between the two types is not always easy to draw, and that features of the inverse type can coexist with those of the symmetrical-voice type in the same language.

The Expression of Three-Participant Events in Movima

In Movima (isolate, lowland Bolivia), transitive clauses are organized not according to semantic roles, but according to a referential hierarchy (1 > 2 > 3, 3 topical > 3 nontopical). The high-ranking argument expression is phonologically attached to the predicate (see the pronouns in (1) and (2)), the lower-ranking one is phonologically independent (see the NPs in (1) and (2)). Semantic roles are indicated by direct (1) and inverse (2) marking on the predicate.