Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/71552
Title: A corpus-based textual analysis of irony and sarcasm in scripted discourse
Authors: Laszlo, Anna Xenia
Advisors: Huang, Chu-ren (CBS)
Keywords: Discourse analysis
Wit and humor
Issue Date: 2017
Publisher: The Hong Kong Polytechnic University
Abstract: The elusive nature of irony and sarcasm, being the pragmatic concepts they are, is seized from the point of view of lexis and grammar. The thesis takes a corpus linguistic and textual analytic orientation. The idea is that, contrary to most previous research bringing out pragmatic aspects as essential in detecting irony and sarcasm, specific lexico-grammatical patterns may be just as good indicators of these phenomena. To test this hypothesis, a soap opera corpus is called in that is representative of spoken data; however, in a written format without intonational marking. The premise is that irony and sarcasm can be identified based on lexico-grammatical patterns. Whereas, irony boasts numerous theories and approaches both theoretical and experimental, sarcasm in its own right is a far more neglected area. The methodology involves the annotation of a tweet dataset for evaluative expressions (as defined by Thompson & Hunston, 2000) of which the most frequent ones are subjected to search in the soap opera corpus. This preprocessing phase involves fifteen keywords, of which five are verbs and ten are adjectives. Once a sample of five hundred examples is obtained for each keyword, they are filtered for ironic/sarcastic examples based on the notions of contrast (Colson & O'Brien, 2000b), reversal of evaluation (Partington, 2007), and conflict (Camp, 2012), which are adequate in cases where tone or facial gestures are not available. In the subsequent step, potentially irony and sarcasm-prone patterns are identified and more examples are generated based on these patterns. These are manually analyzed in terms of lexical and grammatical aspects often invoking corpus linguistic notions, such as semantic preference and semantic prosody. The results support the few studies that ventured to claim that irony and sarcasm research could benefit from considering the lexical properties involved. Certain patterns in themselves can point toward irony and sarcasm in the text as they have been conventionalized to a sufficient extent, while others need the aid of the context. While all keywords are equally prone to both irony and sarcasm, certain fixed patterns show preference to either one of them. Irony and sarcasm are distinguished along the lines of evaluation reversal advocated by Partington (2007). Whereas irony fits the proposed good/bad dichotomy, sarcasm cannot be described in such a diametric and clear-cut way. Sarcasm, namely, involves two meanings which might not be diametric opposites. Further on, the ill-intention and hostility attributed to sarcasm, which is missing in irony, prompted the introduction of a new term of art, sarcastic prosody to encapsulate all those cues found in the sarcastic remark and in its co-text that convey this ill-intention and hostility either through lexico-grammatical cues or a "virtual intonation" accompanying the pattern. Prosody here is meant as a derivative of its "original" sense in semantic prosody (Sinclair, 2003). The data analysis suggests that each word (and pattern) is best examined on its own merits; and that irony and sarcasm do show interesting differences in their manifestations depending on the word (and pattern) under consideration. The main contributions of the thesis are the employment of lexico-grammatical patterns in the analysis as well as the treatment of irony and sarcasm as discrete concepts.
Description: 196 pages
PolyU Library Call No.: [THS] LG51 .H577P CBS 2017 Laszlo
URI: http://hdl.handle.net/10397/71552
Rights: All rights reserved.
Appears in Collections:Thesis

Files in This Item:
File Description SizeFormat 
991021988856803411_link.htmFor PolyU Users167 BHTMLView/Open
991021988856803411_pira.pdfFor All Users (Non-printable)2.46 MBAdobe PDFView/Open
Show full item record

Page view(s)

15
Citations as of Feb 19, 2018

Download(s)

2
Citations as of Feb 19, 2018

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.