Abstract：
Theories of spatial term meanings often focus on geometric properties of objects and locations as the key to understanding meaning. For example, in English, \"The cat is on the mat\" might engage geometric properties characterizing the figure (\'cat\', a point) and the ground (\'mat\', a plane) as well as the geometric relationship between the two objects (\'on\', + vertical, 0 distance from ground object). However, substantial literature suggests that geometric properties are far from sufficient to capture the meanings of many spatial expressions, and that instead, force-dynamic properties of objects that afford containment or support relationships may be crucial to the meanings of those expressions. I will argue that both approaches are needed to understand the variety of spatial terms that appear in language and further, that spatial terms fall into two distinct sets, one represented by geometric properties of figure and ground and their spatial relationships, and the other by the force-dynamic properties of objects and their relationships. This division of labor within spatial terms has many consequences, with the two types differing in the nature of the acquisition problem and likely learning mechanisms, the extent and kind of cross-linguistic variation that has been observed, and the application of pragmatic principles to spatial terms. Speculatively, the two types may also be rooted in different cognitive systems and their neural substrates.
摘要：
空间术语含义的理论通常侧重于对象和位置的几何属性，这是理解含义的关键。例如,在英语中,“猫在垫子上”可能会参与表征图形的几何属性(“猫”，一个点)和地面(\'mat\',平面)以及两个物体之间的几何关系(\'on\',+垂直,与地面物体的距离为0)。然而,大量文献表明，几何性质远远不足以捕捉许多空间表达式的含义，相反，提供包含或支持关系的对象的力动态属性可能对这些表达式的含义至关重要。我认为这两种方法都需要理解语言中出现的各种空间术语，空间术语分为两个不同的集合，一个由图形和地面的几何特性及其空间关系表示，另一个是对象的力动态属性及其关系。这种在空间范围内的分工有很多后果，由于这两种类型在习得问题的性质和可能的学习机制上不同，已经观察到的跨语言变异的程度和种类，以及语用原则在空间术语中的应用。推测,这两种类型也可能植根于不同的认知系统及其神经基底。
