MoL-2007-16: Revealing Concealment: A (Neuro-)Logical Investigation of Concealed Questions

MoL-2007-16: Harris, Jesse Aron (2007) Revealing Concealment: A (Neuro-)Logical Investigation of Concealed Questions. [Report]

Text (Full Text)

Download (1MB) | Preview
[img] Text (Abstract)

Download (3kB)


Natural language is notoriously multi-faceted and thoroughly nefarious. Even the simplest linguistic elements may turn out to support different interpretations in different contexts. And the more one examines natural language, the more disparate and complex the relationship between a word's meaning and the context in which it occurs seems to become. This thesis is about contexts in which nominal elements are not interpreted as nominal elements typically are. To be specific, this thesis is about so-called concealed questions, nominal phrases (e.g., the price of milk) which may be paraphrased as indirect questions (e.g., what the price of milk is). This investigation stands at the crossroads of several lines of research, which are themselves perhaps only tenuously connected. The basic outline of the thesis can be divided into three major types of inquiry: (i) descriptive, (ii) theoretical, and (iii) experimental. It is hoped that the sorts of research discussed here complement one another, giving a richer and deeper understanding of a highly complex and ill-understood phenomenon. I briefly present the basic landscape of the thesis, highlighting the important results. Overview of Thesis: Chapter 2 is primarily descriptive. The core, established properties of concealed questions (CQs) are critically reviewed, and evidence for two new, but revealing, properties is provided. Then, three major accounts are presented: Irene Heim's early, influential account; a string of recent articles by Maribel Romero arguing that CQs are individual concepts, and the recent dissertation of Lance Nathan, exploiting a typological correspondence between CQs and propositions. Chapter 3 is primarily theoretical. A novel semantic account of concealed questions is developed, one that employs the pragmatic notion of conceptual covers. Here, I argue that CQs are best expressed as relations between individual concepts and identity questions, such that the concept provides a constituent answer to the question itself. In particular, I posit a type-shifting rule Q which lifts the nominal to a more complex type <<s,e>,<s,<s,t>>>. Two different ways of viewing this relation are presented, and the cognitive naturalness of Q is briefly discussed and defended. The formal framework is itself developed within a fully compositional Dynamic Montague Grammar with Existential Disclosure. The unique binding properties of Dynamic Semantics is exploited to treat a long standing puzzle regarding an ambiguous interpretation of CQs (Heim's ambiguity). Together, the chapter gives an in depth framework for addressing some of the context sensitivity that CQs display. Throughout, the chapter discusses how CQs relate to propositions, free relatives, and identity questions. Chapter 4 presents the first psycholinguistic experiment on concealed questions. Two separate methodologies, eyetracking and magnetoencephalography (MEG), found that that interpreting CQs taxes the language processor. An interpretation of the results which is broadly compatible with a type-shifting analysis like the one presented in the preceding chapter is subsequently argued for in detail.

Item Type: Report
Report Nr: MoL-2007-16
Series Name: Master of Logic Thesis (MoL) Series
Year: 2007
Date Deposited: 12 Oct 2016 14:38
Last Modified: 12 Oct 2016 14:38

Actions (login required)

View Item View Item