Taking as our point of departure an important distinction between gender as a semantic/conceptual property and gender as a formal morphological property, we argue, assuming a broadly Distributed Morphology/exoskeletal theoretical approach, that formal gender languages, for gender to be visible to the inflectional system, exploit a mechanism that translates semantic/conceptual gender into formal gender by valuing an unvalued formal gender feature. We propose that even though all formal gender features on the DP spine are initially unvalued, only the lowest locus of the unvalued gender feature is valued by the translation mechanism. Subsequent gender agreement is accomplished by probe – goal valuing. Next, we inquire about the exact locus of the lowest formal gender feature. We argue first that gender in the Norwegian DP is a formal feature on an already existing functional head. Then, based on Norwegian – English language mixing, we argue that the lowest locus of the formal gender feature may vary depending on the size of the non-gender chunk mixed from English. In other words, there is no designated functional head where the lowest gender feature is always generated, and the formal unvalued gender feature at any functional head can in principle be the lowest formal gender feature.

中文翻译：

---

性别“翻译”和分布式性别：来自挪威 DP 和语言混合的证据 *

将性别作为语义/概念属性和性别作为形式形态属性的重要区别作为我们的出发点，我们认为，假设广泛分布的形态学/外骨骼理论方法，正式的性别语言对于性别是可见的屈折系统，利用一种机制，通过评估未评估的正式性别特征，将语义/概念性别转化为正式性别。我们提出，即使 DP 脊柱上的所有正式性别特征最初都没有被重视，翻译机制也只会重视未被重视的性别特征的最低位点。随后的性别一致性是通过探测-目标评估来完成的。接下来，我们询问最低正式性别特征的确切位置。我们首先争辩说，挪威 DP 中的性别是已经存在的功能性头脑的正式特征。然后，基于挪威语 - 英语语言混合，我们认为正式性别特征的最低位点可能会因英语混合的非性别块的大小而异。换言之，不存在始终生成最低性别特征的指定功能头，并且任何功能头处的正式的未赋值性别特征原则上可以是最低的正式性别特征。