Conditionals and Discourse

本项目使用条件句来研究语篇结构在意义构建中的作用。该项目采用形式化和清晰的话语理论（被理解为话语移动的层次系统），研究话语结构如何在句子层面上被纳入意义的构建。重点是一组条件结构，其含义似乎超出了传统的真理条件（话语敏感条件句，DSC）。呼吁与语篇结构的互动提供了一个自然的帐户，而不需要特设的句法或语义移动的DSCs。本文的研究支持了句子意义和语篇结构之间的关系比以往所认为的更加丰富和清晰的假设，从而为人们对语篇在意义构建中的作用的新的思考方式提供了支持。主要的实证研究将集中在两种类型的效果上，这两种效果将被用来说明条件句中的语篇敏感性：穷尽性效果和合意性效果。它们在文献中已经在“条件完备性”和“选择性”的标题下进行了讨论。条件完备性（英语：Conditional perfection）（“exhaustivity effects”）：语义学的解释是，如果p，q，p是q的条件。然而，当听到条件if p，q时，听众往往会学习到p是q的穷举条件的丰富含义。这意味着p不仅是q的一个条件，而且是q（p-q）的一个必要条件：（1）如果你修剪草坪，我会给你5美元。听到（1），听众会得出结论，不仅修剪草坪是获得5美元的一个条件，而且只有修剪草坪才能获得5美元。像（1）这样的例子中的神秘之处在于，观察到的与先行词相关的穷尽效应并不是由条件句的常规语义所预测的。这种效应是一种强有力的现象，被称为“条件完美”。条件句的先行词中某些焦点敏感算子的存在可能会引起带有合意性的解释，而这是结构的传统语义所无法预测的。这种结构被称为“选择性”结构：（2）如果飞机晚点5分钟，（我就会接通电话了）。给出（2）的话语，听者不仅理解到晚点是接通电话的条件，而且还理解到说话人希望接通电话。然而，在（2）中没有一个词项负责这一意义，这仍然是传统语义学无法解释的。

This project uses conditional constructions to investigate the role of discourse structure in the construction of meaning. Adopting a formalized and articulated theory of discourse (understood as a hierarchical system of discourse moves), the project investigates how the structure of discourse is factored into the construction of meaning at the sentence level. The focus is on a set of conditional constructions whose meanings appear to go beyond conventionally conveyed truth-conditions (discourse sensitive conditionals, DSCs). Appealing to the interaction with discourse structure provides a natural account of DSCs without requiring ad-hoc moves in the syntax or semantics. The study of DSCs supports the hypothesis that the relation between sentence meaning and discourse structure is richer and far more articulated than has previously been thought, providing support for new ways of thinking about the role of discourse in the construction of meaning.The main empirical focus will be on two types of effects that will be argued to illustrate discourse-sensitivity in conditionals: exhaustivity effects and desirability effects. They have been discussed in the literature under the headings of "conditional perfection" and "optativity".A. Conditional perfection ("exhaustivity effects")Standardly, semantic accounts predict that in a conditional of the form if p, q, p is a condition for q. However, upon hearing a conditional if p, q, listeners often appear to learn the enriched meaning that p is an exhaustive condition for q. This means that p is not just a condition for q, but also a necessary condition for q (p - q):(1) If you mow the lawn, I will give you $5.Upon hearing (1), listeners standardly conclude that not only is mowing the lawn a condition to obtain $5, but also that only by mowing the lawn will the $5 be obtained. What is mysterious in cases like (1) is that the observed exhaustivity effect associated with the antecedent is not predicted by the conventional semantics of the conditional. The effect is a robust phenomenon that has come to be known as "conditional perfection".B. Optativity ("desirability effects")The presence of certain focus sensitive operators in the antecedents of conditionals can give rise to interpretations flavored with desirability that are not predicted by the conventional semantics of the structure. This type of structure is known as "optative":(2) If only the plane had been delayed for five minutes, (I would have made my connection).Given an utterance of (2), the listener doesn¿t only come to understand that a delay was a condition to make the connection, but also that the speaker desires to have made the connection. There is, however, no lexical item in (2) responsible for this meaning, which remains unexplained by the conventional semantics.