Achard, Michel2019-05-162019-05-162017-122017-10-31December 2Lanneau, Bazile Rene. "Schema-guided comprehension of noun-noun compounds: An experimental and corpus-based approach." (2017) Diss., Rice University. <a href="https://hdl.handle.net/1911/105565">https://hdl.handle.net/1911/105565</a>.https://hdl.handle.net/1911/105565This dissertation provides evidence that the comprehension of noun-noun compounds is guided by probabilistic knowledge of the structures of things, events, and situations. These structures are what we refer to as “schemas”. As a theoretical construct, schemas are important because they supply information that is not explicitly given. Understanding language involves inference, and this is particularly true for noun-noun compounds. The constituents of lexicalized compounds are semantically linked in a variety of ways. A door knob is a knob that is located on a door, a wedding dress is a dress designated for use in a wedding, a chicken leg is the leg of a chicken, and so on. The list of relations semantically linking the constituents of noun-noun compounds goes far beyond these Locative, Purposive, and Part-Whole examples. To complicate matters, such relations can be analyzed at varying levels of specificity, creating a trade-off between semantic accuracy and theoretical elegance. Language use therefore offers no deterministic rule for how novel noun-noun compounds should be interpreted. Cloud house, for example, could be understood as 'a house in the clouds', 'a house from which clouds can be seen', or 'a house in the shape of a cloud'. Understanding how such interpretations are constructed would provide crucial clues to how language comprehension works in general. To that end, a random sample of transparent noun-noun compounds from The Corpus of Contemporary American English is analyzed. Over two hundred examples of noun-noun compounds are provided whose meanings appear to be obvious only in light of schematic world knowledge. Additionally, a series of experiments are discussed which provide evidence that interpretations of novel noun-noun compounds can be biased by abstract conceptual schemas. Through reading tasks, subjects are prompted to build schemas by mapping animals onto roles typically filled by humans in schemas of human activity. Subjects are thus primed to interpret novel compounds of the form Animal-Noun anthropomorphically, e.g., crab shirt as 'a shirt worn by a crab'. In conjunction with a broad theoretical discussion, these qualitative and experimental data provide a strong case that schemas play a central role in noun-noun compound comprehension.application/pdfengCopyright is held by the author, unless otherwise indicated. Permission to reuse, publish, or reproduce the work beyond the bounds of fair use or other exemptions to copyright law must be obtained from the copyright holder.noun-noun compoundschemacognitive linguisticspsycholinguisticsconceptual blendingdialogicconstruction grammarframe semanticssemanticslanguage comprehensionconcept combinationcorpus analysisThesis2019-05-16