‘Case’ and ‘grammatical relations’ are central to syntactic theory, but rigorous treatments of these concepts in surface-oriented grammars such as Dynamic Syntax are pending. In this respect, Japanese is worthy of mention; in this language, the nominative case particle ga, which typically marks a subject, may mark an object in certain syntactic contexts, and more than one instance of ga may be present within a single clause. These patterns cannot be captured if we simply assume that ga marks a subject. In the present article, we aim to advance formal aspects of the framework, especially the mechanism of ‘structural underspecification,’ by proposing that the parse of a case particle maximally excludes potential landing sites of an unfixed node at the time of parsing the case particle, delaying the resolution of the unfixed node until a subsequent stage of structure building. This maximal exclusion approach to structural underspecification accounts for a range of case marking patterns and their connections with grammatical relations.

中文翻译：

---

动态语法中结构欠规范的“最大排除”方法

“案例”和“语法关系”是句法理论的核心，但在面向表面的语法（例如动态语法）中对这些概念的严格处理尚待确定。在这方面，日本人是值得一提的；在这种语言中，通常标记主语的主格助词 ga 可能在某些句法上下文中标记一个宾语，并且单个子句中可能存在多个 ga 实例。如果我们简单地假设 ga 标记了一个主题，则无法捕捉到这些模式。在本文中，我们旨在推进框架的形式方面，特别是“结构欠规范”的机制，通过提出案例粒子的解析在解析案例粒子时最大限度地排除未固定节点的潜在着陆点, 将未固定节点的解析延迟到结构构建的后续阶段。这种对结构性欠规范的最大排除方法解释了一系列案例标记模式及其与语法关系的联系。