Argument structure and genitive modification in the Old English noun phrase

当古英语英雄贝奥武夫死去时，他的灵魂离开了他的身体，寻找sofæstra dom，即“正义的审判”。文学批评争论中心名词（或possessum），dom“判断”，和从属属格（或占有者），sofæstra“正义的人”之间的关系;是正义的人做的判断，贝奥武夫是该死的，还是他的灵魂寻求给予正义的判断？占有者可以是被占有者（进行判断）或被占有者（被判断）所表达的行为的施动者。该项目旨在从形式语言学的角度解决sofæstra dom例子中所见的形容词属格修饰类型。它将集中在由属格修饰的所有格名词的例子，这些名词至少有一个主题角色，如施事或受事，附加到它，或者，在更正式的术语中，有论元结构（a-结构）。在现代英语中，这包括“约翰对玛丽的爱”和“约翰的暗杀”。这个项目并不关心像cyninges mearh“国王的马”这样的拥有短语，因为“马”不是一个带有a结构的名词。本文探讨了与论元结构名词的题元角色相关的领属语在语法关系中的作用。对于一个动词短语，如“正义的法官贝奥武夫”，确定主语和宾语是直截了当的，但同样容易的识别不适用于名词短语。一种可能性是，具有论元结构的名词（下文称为A-名词）与相关动词具有相同的A-结构;因此，像dom这样的名词以与动词deman“to judge”相同的方式，对主语和宾语的语法功能进行了次分类。A-名词可以被认为具有名词性的特定语法功能，如POSS，如用于“国王的马”中的“国王”等核心所有者。本文的研究目的之一是确定如何最好地从句法上处理A-名词的物主参与者。我建议主要使用约克-多伦多-赫尔辛基古英语散文分析语料库（YCOE，见http：//www-users.york.ac.uk/YCOE/YcoeHome.htm），这是一个150万字的句法注释语料库。建议采用混合方法收集数据;将使用注释以电子方式搜索语料库中所有文本的1 000个单词样本，以查找占有者/占有者结构。可以从这些搜索结果中提取相关的A名词，然后在整个YCOE中搜索它们。此外，这些A-名词也可以在YCOE中没有包含的诗歌文本中使用古英语语料库词典进行搜索，这些语料库不被解析。使用这种混合的数据收集方法，将创建一个数据库的实例，所有选定的A-名词与形容词所有者。数据库中的每一个例子将标记的语言特征，包括主题的作用或可能的作用的形容词拥有者，任何deverbal形态上的中心名词的存在，两个NP的生命力，以及配价和古英语中相关动词的a-结构模式。然后将利用这一数据库对所确定的各种特征的相对重要性进行一些基本的统计分析。该数据库可在线提供，以便于复制和开放获取。该项目将对英语中早期的A-名词的句法地位进行实证研究。本文的研究将对英语名词短语和论元结构的研究，对古英语的句法研究，以及对英语中名词短语和论元结构的研究，做出理论、实证和历时的贡献。这项研究涉及到自然语言中是否存在变异，以及语言的重复性是否是现实。

As the Old English hero Beowulf dies, his soul departs his body and seeks the sofæstra dom, the "judgement of the righteous". Literary criticism debates the relationship between the head noun (or possessum), dom "judgement", and the dependent genitive (or possessor), sofæstra "righteous people"; are the righteous those doing the judging, and Beowulf is damned, or does his soul seek out the judgement afforded to the righteous? The possessor can be the agent of the action expressed by the possessum (doing the judging) or the patient (being judged). The project proposed seeks to address adnominal genitive modification of the type seen in the example sofæstra dom from a formal linguistic perspective. It will focus on examples of possessors marked by genitive case modifying possessum nouns which have at least one thematic role, such as agent or patient, attached to it, or, in more formal terms, which have argument structure (a-structure). In Presentday English, this includes "John's love for Mary", and "John's assassination". The project is not concerned with phrases of possession like cyninges mearh "king's horse", since "horse" is not a noun with a-structure. Under investigation is how the possessors associated with thematic roles of nouns with argument structure function in terms of grammatical relations. For a verbal phrase, like "the righteous judge Beowulf", identifying subject and object is straightforward, but the same easy identification does not hold in the noun phrase. One possibility is that nouns with argument structure (hereafter A-nouns) have the same a-structure as related verbs; a noun like dom therefore subcategorizes for subject and object grammatical functions in the same way that the verb deman "to judge" does. A-nouns may be thought of as having nominal- specific grammatical functions like POSS, as is used for core possessors like "king" in "king's horse". It is one goal to determine how best these possessor participants of A-nouns can best be treated syntactically.My study will use corpus methodology, as is usual in current historical linguistics. I propose to chiefly use the York-Toronto-Helsinki Parsed Corpus of Old English Prose (YCOE, see http://www-users.york.ac.uk/~lang22/YCOE/YcoeHome.htm), which is a 1.5 million word syntactically-annotated corpus. A mixed methodology is proposed for the data collection; samples of 1,000 words of all texts in the corpus will be searched electronically for possessor/possessum structures using the annotation. Relevant A-nouns can be extracted from these search results and will then themselves be searched for across the entire YCOE. In addition, these A-nouns can also be searched for in the poetic texts not included in the YCOE by using the Dictionary of Old English corpus, which is not parsed. Using this mix of data collection methods, a database will be created of instances of all selected A-nouns with adnominal possessors. Each example in the database will be tagged for linguistic features including the thematic role or probable roles of the adnominal possessor, the presence of any deverbal morphology on the head noun, the animacy of the two NPs, and the valency and a-structure patterns of related verbs in Old English. This database will then be used to produce some basic statistical analysis on the relative significance of the various features identified. This database can be made available online, in the interests of reproducibility and open access. The project will make an empirically validated claim about the syntactic status of A-nouns at an early period in English. The proposed research will make a theoretical, empirical and diachronic contribution to the study of the noun phrase and argument structure, possession in English, and to study of the syntax of Old English. The research speaks to whether there is such a thing as variation in natural language, and to whether linguistic vagueness is a reality.