Phonological constraints and free variation in compounding: A corpus study of English and Estonian noun compounds.
Item
-
Title
-
Phonological constraints and free variation in compounding: A corpus study of English and Estonian noun compounds.
-
Identifier
-
AAI3213181
-
identifier
-
3213181
-
Creator
-
Sepp, Mary.
-
Contributor
-
Adviser: Martin Chodorow
-
Date
-
2006
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Language, Linguistics | Language, Modern
-
Abstract
-
This research was designed to examine the patterns of variation in the phonological and/or orthographic form of Estonian and English noun compounds. Estonian noun compounds generally occur in one of two forms: N1(nominative) + N2, as in kool + meister ("schoolmaster"), or N1(genitive) + N2, as in kooli + opetaja ("schoolteacher"). Some Estonian compounds vary freely in form---e.g., veebsepp/veebisepp ("webmaster"). English noun compounds exhibit orthographic variation, as they may be written in three ways: closed ("bookstore"), hyphenated ("dot-com"), or open ("space station"). Many English compounds also vary freely---e.g., cellphone/cell-phone/cell phone. The principal goal of this study was to use statistical data derived from corpora to determine which variables best account for the choice of variant compound forms.;The 1,094 Estonian compounds used in this research came from a one million word corpus of Estonian literary and news texts. Data on variation of form were obtained from Google searches of the World Wide Web. Results showed a strong preference for genitive forms, and it was posited that this preference is due to general principles of ease of pronunciation and ease of perception.;Phonology is also a factor in the distribution of English compounds. A number of phonological variables were examined in the current study: number of syllables, presence of compound stress, vowel sequences across internal lexical boundaries, and double consonants across internal lexical boundaries. Frequency data for these variables were extracted from a fourteen million word English corpus. Results of multiple regression analyses showed that the number of syllables in the compound is a stronger predictor of orthographic form than the other phonological features that were tested. Phonology was not assumed to be the only influence, however; lexical features were also examined. Results indicated a substantial contribution of the second constituent in predicting whether the compound would be open or closed, and a lesser, though important, contribution of the first constituent. A regression analysis combining phonological and lexical variables accounted for about 68% of the variance in the orthography of 707 high frequency English noun compounds.
-
Type
-
dissertation
-
Source
-
PQT Legacy CUNY.xlsx
-
degree
-
Ph.D.
