The acquisition of voicing an vowel alternations in German

这个项目的目的是研究德语单复数对的语音习得和元音变化。语态的交替是德语的一个规律过程，从某种意义上说，所有复数形式的词干-词尾浊音障碍都对应于单数形式的词干-词尾浊音障碍。在文献中，这个过程被称为最终奉献（见Beckmann， 1997，以及其中的参考文献）。元音交替，也被称为Umlaut，是一种过程，在这个过程中，单数的后词干元音对应于复数的前词干元音。这是一个不规则的过程，因为并非所有复数中的前词干元音都对应于单数中的后词干元音。音系学理论对这些过程的习得做出了不同的预测，这是基于它们如何解释变化。区分规则交替和不规则交替的理论预测，一旦获得规则交替，所有相关的新项目都会受到该交替的影响。规则交替是盲目应用规则或约束的结果，而不规则交替是逐项学习的结果。另一方面，不规律的变化永远不会外推到新的项目上。根据与其他物品的相似性来解释变化的理论，预测了规则物品和不规则物品的逐渐习得。每当一个新项与另一个交替出现的项相似时（在某种程度上将在下面明确说明），新项也将交替出现。基于一项关于荷兰语语音交替习得的研究（Kerkhoff, 2007），我设计了一组实验来检验这些假设。该项目的主题是当前音韵学理论中关于音韵学是分类还是随机问题的争论的中心，而像DFG的优先框架这样的场所，其中许多数据被纳入音韵学理论的范围，似乎是进行这项研究的绝佳框架。

The goal of this project is to study the acquisition of voicing and vowel alternations in singular-plural pairs in German. Alternation of voicing is a regular process of German, in the sense that there are all stem-final voiced obstruents in the plural correspond to a stem-final voiceless obstruent in the singular. In the literature this process is known as final devoicing (see Beckmann, 1997, and references therein). Vowel alternation, also known as Umlaut, is a process in which a back stem vowel in the singular corresponds to a front stem vowel in the plural. It is an irregular process, since not all front stem vowels in the plural correspond to a back stem vowel in the singular. Phonological theories make different predictions concerning the acquisition of these processes on the basis of how they explain alternations. Theories that make a difference between regular alternations, which are a consequence of blind application of rules or constraints, and irregular alternations, which are learned on an item-by-item basis, predict that once a regular alternation is acquired all relevant new items will be affected by the alternation. Irregular alternations, on the other hand, are never extrapolated to new items. Theories that explain alternations based on similarities with other items, predict a gradual acquisition of both regular and irregular items. Whenever a new item resembles–in a way that will be made explicit below–another item that alternates, the new item will also alternate. Based on a study on the acquisition of voicing alternations in Dutch (Kerkhoff, 2007), I have designed a set of experiments to test these hypotheses. The topic of this project is at the center of a current debate in phonological theory about the question if phonology is categorical or stochastic, and a venue such as the priority framework of the DFG, in which many data are brought into purview of phonological theories, seems an excellent framework in which to conduct this research.