| |
Abstract:
Frazier and Rayner (1990) compared words with multiple
meanings (e.g., the ambiguous word "pitcher") with words with
multiple senses (e.g., the abstract (metonymic) and concrete
sense of "newspaper"; as in "He read the newspaper" and "He
worked for the newspaper"). They found that, when the
disambiguating information was delayed, words with multiple
senses behaved like monosemous control words. However, when
disambiguating information preceded the target word, the
unpreferred reading of a word with multiple senses required extra
processing. This result goes against what is normally assumed in
figurative language research that, in an appropriate context, the
literal and figurative readings of a word or phrase are processed
equally fast (e.g., Gibbs, 1984, 1989, 1994; Gibbs and Gerrig,
1989; Inhoff, Lima, and Carroll, 1984; Cacciari and Glucksberg,
1994; Onishi and Murphy, 1993). Since there were a number of
problems with the interpretation of Frazier and Rayner's (1990)
items, a.o. that not all of them where metonymic, we used eye
tracking to investigate the on-line processing of two different
types of metonymy: PLACE FOR INSTITUTION and PLACE FOR EVENT, as
in (1) and (2) respectively:
(1) The new applicants consulted with the university
(2) A lot of Americans protested during Vietnam
In addition to the metonymic condition, the same places
appeared in contexts supporting their literal sense ("A lot of
tourists visited the university" and "We made a trip around
Vietnam"). In two other conditions, the places were substituted
by other places that did not have a conventionalized metonymic
reading ("consulted with the lighthouse"/"visited the lighthouse"
and "protested during Finland"/"trip around Finland"). The
results of these two experiments clearly showed that
conventionalized metonymies are processed immediately and
extremely fast, which is incompatible with the classic linguistic
model of figurative language processing (Searle, 1979). Moreover,
unlike homonyms, neither the relative frequencies of the senses,
nor whether the literal meaning was the most basic one (according
to dictionary listings), influenced reading times. The two types
of metonymies also seemed to be processed differently: the
anomolous metonymic condition caused much less initial disruption
for the PLACE FOR EVENT constructions. The data will be
interpreted in a parallel model in which the processor 'homes in'
on the intended sense after initially accessing an
underspecified, general meaning.
Frazier, L. and Rayner, K. (1990). Taking on semantic
commitments: Processing multiple meanings vs. multiple senses.
Journal of Memory and Language,
29, 181-200.
Gibbs, R. (1989). Understanding and literal meaning.
Cognitive Science,
13, 243-251.
Onishi, K. H. and Murphy, G. L. (1993). Metaphoric reference:
When metaphors are not understood as easily as literal
expressions.
Memory and Cognition,
21 (6), 763-772.
|
