Doctoral Dissertation Research: Reinterpreting Condition B: An Investigation of Pronominal Reference in Romanian

自然语言使人类能够指称世界上的物体，但语言实现这一目的的方式往往是使用看似信息量不足的语言设备：像“他”或“她”这样的代词。不像“哈利波特”这样的专有名词，代词没有固定的指称，因此只能在特定的语境中理解。出于这个原因，它们提供了一个直接的窗口，让我们了解人类大脑是如何在一个环境中追踪有关人和事物的信息的。对听者如何理解代词的现有研究已经对他们如何追踪话语中不同实体的信息产生了实质性的见解。研究表明，不同的语言在这个过程中有很多共同点。与此同时，有证据表明，语言在代词如何指代话语实体方面确实存在差异。这就引出了一个很自然的问题：我们使用语言指代的能力有多少是根植于大脑的普遍特征，也许是独立于语言的，又有多少是来自于我们碰巧说的语言的具体特征？这些问题的答案告诉我们，语言变异和变化背后的社会和文化因素。这个项目通过研究罗马尼亚语来调查这个问题，罗马尼亚语是一种罗曼语，全世界大约有3000万人使用。罗马尼亚语似乎是对代词指称的普遍约束的一个反例。特别是，在英语句子“Harry talks about him”中，说英语的人无法自然地将“him”解释为“Harry”。在罗马尼亚语中，这是不正确的，使罗马尼亚语成为一个明显的反例，否则强有力的跨语言概括。目前的提议测试了罗马尼亚人为什么能以这种明显不同寻常的方式解释代词的假设。研究人员将调查上下文的不同特征，如代词潜在的指称歧义及其句法位置，如何影响罗马尼亚人对代词的解释，以及罗马尼亚人在生产中选择什么类型的代词来指代上下文中不同的先行词。研究者采用心理语言学的方法来解决他们的研究问题，使用一种版本的图片描述范式来研究代词的产生，使用一种版本的优先寻找任务来研究理解。这是第一个关于罗马尼亚语的研究，他们的研究结果将引起理论语言学家和心理语言学家的兴趣，他们对语言如何限制我们在语境中引用的能力感兴趣。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Natural language allows humans to refer to objects in the world, but often the way that language does this is by using seemingly under-informative linguistic devices: pronouns like 'he' or 'she.' Unlike proper names like 'Harry Potter,' pronouns do not have fixed reference, and so they can only be understood relative to a given context. For this reason, they offer a direct window into how the human mind tracks information about people and things in a context. Existing research into how listeners understand pronouns has already yielded substantial insight into how they track information about different entities in a discourse. Research suggests that there is much about this process that is shared across different languages. At the same time, there is evidence that languages do vary in how pronouns refer back to discourse entities. This raises the natural question: how much of our ability to refer using language is rooted in universal features of the mind, perhaps independent of language, and how much of it comes from the specifics of the language that one happens to speak? Answers to these questions inform us of the social and cultural factors that underlie language variation and change.This project investigates this question by looking at Romanian, a Romance language spoken by approximately 30 million people worldwide. Romanian appears to present a counter-example to a supposedly universal constraint on pronominal reference. In particular, in the English sentence 'Harry talked about him,' English speakers cannot naturally interpret 'him' to refer back to 'Harry.' In Romanian, this is not true, making Romanian an apparent counter-example to an otherwise robust cross-linguistic generalization. The present proposal tests hypotheses for why Romanian speakers can interpret pronouns in this apparently exceptional fashion. The researchers will investigate how different features of the context, such as the potential referential ambiguity of a pronoun and its syntactic position, impact the interpretations Romanian speakers assign to pronouns, as well as what types of pronouns Romanian speakers choose in production to refer to different antecedents within a context. The investigators take a psycholinguistic approach to their research questions, using a version of the picture description paradigm to investigate the production of pronouns, and a version of the preferential looking task to investigate comprehension. The proposed studies are the first of their kind on Romanian, and their results will be of interest to both theoretical linguists and psycholinguists interested in how language constrains our ability to refer in context.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.