Linguistische Berichte Heft 214

The Nominal Mapping Parameter gave rise to an acquisition model which makes clear and falsifiable statements about the acquisition of articles in German. It claims that children acquiring a Germanic language omit articles for a longer period of time than children acquiring a Romance language, as they have to decide for each noun separately whether it is count or mass. Thus, article drop in the acquisition of Germanic languages is considered to be a lexical learning problem. This study presents a critical discussion of this model on the basis of data from ten children acquiring German. Although these children pass through a protracted period of article omissions, omissions are unrelated to the count/mass distinction.

The paper compares two important approaches to tense. Both Reichenbach (1947) and Klein (1992 and subsequent work) assume a point of speech at which an utterance is made and an event time at which the eventuality described takes place. Reichenbach proposes as a third point the reference time that is generally understood as the point in time relative to which the event time is located. According to Klein this third point is the topic time and has to be defined as the time for which the particular utterance makes an assertion. The paper argues that both the topic time and the reference time are problematic.

This paper provides an analysis of the typologically unusual patterns of case assignment in Tlapanec. The proposal challenges some idiosyncratic aspects of former descriptive analyses (i.e. zero-marked ergative and a novel case 'pegative') and provides a theoretically and typologically more adequate analysis instead. The theoretical approach relies on the framework of Distributed Morphology (DM). According to this theory the output of syntax can be impoverished before the lexical insertion of concrete vocabulary items takes place. This framework is modified in the account proposed here, insofar as the step of impoverishment of morphosyntactic features is designed as an optimality theoretic competition. I will show that all of the constraints that are crucial for this competition emerge from independently established prominence scales by a modified version of harmonic alignment. Thereby the process of impoverishment is given a typologically/conceptually more plausible base.