In Shifting Animacy, Peter de Swart and Helen de Hoop argue that animacy in general and animacy shifts in particular can be better understood if we take into account the distinction between grammatical and conceptual animacy. They argue that possible mismatches between conceptual and grammatical animacy licence animacy shifts. Their approach seems promising to me,1 but I would nevertheless like to take this opportunity to discuss further some of the points made in the paper. If I understand their claims correctly, de Swart and de Hoop propose that linguistic animacy is discrete and binary and conceptual animacy is gradient and non-binary. Linguistic animacy is modelled with binary features and conceptual animacy is modeled with a hierarchy (the animacy hierarchy). I view their proposal as an invitation to consider further the question whether featural analyses are necessarily discrete and analyses that appeal to hierarchies necessarily gradient.

中文翻译：

---

梯度，特征和层次结构

彼得·德·斯瓦特（Peter de Swart）和海伦·德·霍普（Helen de Hoop）在《变态的动画》中指出，如果我们考虑语法和概念上的动画性之间的区别，可以更好地理解一般的动画性，尤其是动画性的变化。他们认为，概念性和语法性许可许可的动画性可能会发生不匹配。他们的方法对我来说似乎很有前途，1但是我还是想借此机会进一步讨论本文提出的一些观点。如果我正确地理解了他们的主张，de Swart和de Hoop提出语言的灵巧性是离散的和二元的，而概念性的灵巧性是梯度的和非二元的。语言的动画能力是用二进制特征建模的，而概念动画效果是用层次结构（动画能力的层次结构）建模的。