From: Brian Murphy <brian.murphycs.tcd.ie>
Subject: English Affect Word-List
Query for this summary posted in LINGUIST Issue:
Regarding Query: http://linguistlist.org/issues/17/17-2582.html#1
First of all, thanks to all that contributed: Maite Taboada, Aubrey Nunes,
Nora Wiedenmann, Elena Bashir and Cynthia Girand.
Aubrey Nunes and Nora Wiedenmann pointed out phonological patterns in
English (e.g. shlep, slink) and German (e.g. Schlampe, Klampfe) that
correlate with negative connotations.
Maite Taboada pointed me to an online discussion group on the computational
treatment of sentiment/affect
(http://tech.groups.yahoo.com/group/SentimentAI/), which hosts several
hand-crafted lists, including her own, based on Martin and White's (2005)
Appraisal framework (http://tech.groups.yahoo.com/group/SentimentAI/).
Cynthia Girand directed me Cynthia Whissel's Dictionary of Affect
(Laurentian University). This is the most comprehensive resource I have
found with more than 8000 English words. Each word has been experimentally
examined, and carries an average value for elicited human judgements of
pleasantness, activation and imagery. The dictionary is freely available
for non-commercial academic purposes.
Incidentally, I didn't make my purposes clear in the initial posting. I am
interested in argument structure realisation, and the fact that some
patterns involve advantage (e.g. beneficial ''for'' in English) or
disadvantage (e.g. the Chinese bei construction) to an event participant. I
am trying to establish the strength of these effects from corpus data.
Trinity College Dublin
Respond to list|Read more issues|LINGUIST home page|Top of issue