The acquisition of tense and aspect in a morphology-sensitive framework: Data from Italian and Austrian-German children

This article criticizes the wide-spread view, sometimes referred to as the "aspect first hypothesis" (initiated by Antinucci and Miller 1976 and supported by Bloom et al. 1980; Bickerton 1981; Weist et al. 1984; Shirai and Andersen 1995, among others), according to which a universal acquisition path is postulated in the tense-aspect domain, based on the leading role of actionality (or Aktionsart) and aspect. According to this view, children build their competence starting from the pervasive correlations atelic therefore imperfective therefore present vs. telic therefore perfective therefore past, before gradually learning to disentangle (i.e., freely combining) the various actional, aspectual, and temporal components. The alternative view advocated here (typologically-oriented and morphologically-sensitive) claims, instead, that children start out with no predefined strategy and extract the relevant information out of the individual language's morphological structure. The data stem from four longitudinal corpora relating to three Italian children and one Austrian German child, showing that: (i) the strong correlation between actionality, aspect, and tense can only be supported if activity and stative verbs are lumped together within the category of atelic predicates. Once activities are separately examined, their behavior stands out as absolutely incompatible with the traditional view. (ii) In the relevant languages, there can be earlier understanding of the temporality-oriented morphology as contrasted with the aspect-related categories. (iii) The analysis does not support the so-called prototype hypothesis (Shirai and Andersen 1995), since the examined children were strongly affected by their linguistic input from the very beginning. (iv) The children presented (to a greater or lesser extent) a notable verb spurt that very briefly preceded the first uses of the Past tenses. In conclusion, the actual acquisition path followed by the analyzed children undermines the hypothesis of a universal acquisition pattern, supporting instead the view that acquisition depends on the specific morphological shape of the target language.