I thank Robyn for striving to bring clarity to the "metaphor and effort"issue. Here is my stab at it from an experimental point of view:
If one accepts that a metaphoric expression comes with extra effects (even
weak effects) with respect to comparable non-metaphoric expressions (let's
call the latter "taut-talk") it follows that these additional effects
should come at a cost. This cost is cashed out in effort. That is why I
think it makes sense to suppose that metaphors should be effortful to
process. This idea was inspired by RT (obviously), especially Sperber and
Wilson's paper on loose talk (1986). As it turns out, experimental data
support such an analysis. What is this evidence:
The most compelling evidence in favor of this idea comes from tasks in
which people can't help but generate a metaphor. In the classic work
coming out of Glucksberg's lab, one finds that participants -- when asked
to make truth-judgements about literally false statements like "Some roads
are snakes", "Some jobs are jails" or "Some marriages are iceboxes"-- take
longer to respond "false" than they do to control statements like "Some
roads are jails." This longer reading time is readily interpretable as
being due to the extra effort applied to detecting and deciphering the
metaphoric interpretation. At the very least, this means that metaphor
demands resources enough to distract someone doing a truth-judgment task
(indeed, the relatively longer reading times were the basis for the claim
that metaphors were generated automatically). In itself I think that this
indicates that a statement with metaphoric appeal takes longer to process
than a comparable statement without such appeal. (For evidence on the
relative advantages of metaphor for comprehension, i.e. how metaphor
provides additional effects in experimental situations, see Noveck et al.;
but that would be a different argument.)
Gildea and Glucksberg's concerns lied elsewhere, i.e. with showing that
metaphoric interpretations are readily accessible (in much the same way
people's ability to do categorisation is), but there is a distinction to be
made between automaticity and relative effortlessness. That is, a metaphor
can be both readily accessible and extra-effort demanding (in fact, once
generated it must be using up some resources...I'd argue more than a
non-figurative equivalent). This distinction has often been overlooked.
There is often the perception that "readily accessible" is equivalent to
"no-extra-effort" due to, I think, the literature's collective
determination to show that metaphor is not special. So, people like Gerrig
(1989) link Gildea and Glucksberg's data (which, again, I believe shows
that metaphors require extra-effort but are readily accessible) with those
that indicate that metaphors are as easy to comprehend as literal
expressions, leaving the impression that metaphors demand as little effort
to comprehend as comparable literal expressions. Even Gerrig's (Gerrig &
Healy's) data indicate that metaphors are more effort-consuming when 1)
compared to literal equivalents and 2) when they precede a context rather
than when they follow it (compare "The night sky was filled with drops of
molten silver." and "Drops of molten silver filled the night sky.")
So, what I'll call stand-alone metaphors (experiments that present
metaphors with no context) arguably do take some effort. Once in context,
the story changes some. If there is enough supporting background
information, it appears that a metaphoric expression can appear as
effortless as a literal one. This explains why some studies show no
differences between literal and metaphoric sentences when the same sentence
can have a literal or figurative reading. Compare these two passages (as
cited by Gerrig, 1989) where the reading times for the last sentence in
each is identical:
Literal: Joan didn't want to put her silk blanket in her automatic dryer.
Although it was January, she risked putting it on the clothesline. The
winter wind gently tossed the lacy blanket.
Figurative: Joan looked out into her yard with great excitement. Over
night, a layer of snow had covered the ground. The winter wind gently
tossed the lacy blanket.
Note, however, that we are no longer comparing metaphoric with
non-figurative equivalents. The contexts are dramatically different,
making the effects different as well; we are no longer talking about
"extra" effects in the way a literal/figurative comparison does.
Nevertheless, the data suggest that ease-of-information-integration is
equivalent in the two cases.
Metaphors do take some effort, but preceding context can arguably ease the
burden. With enough contextual information, an utterance's or sentence's
metaphoric potential can be readily capitalized. However, when a metaphor
arrives on the scene unexpected, it eats up time and this implies a cost.
This is why I think a referential expression in our experiments (and
others') appear effort-demanding -- they are unanticipated...in much the
same way a non-contextualized, predicative metaphor is. When texts are
generally taut-talk (or when there is silence) before the onset of the
metaphor, it will take some effort in order to arrive at the intended
interpretation.
Gerrig, R. (1989). Empirical constraints on computational theories of
metaphor. Cognitive Science 13, 235-241.
Gerrig and Healy (1983). Dual Processes in Metaphor Understanding:
Comprehension and APpreciation. JEP:Learning, Memory and Cognition, Vol.
9, No. 4, 667-675.
Gildea, P. & Glucksberg, S. (1983) On understanding metaphor: The role of
context. J of Verbal Learning and Verbal Behavior, 22, 577-590.
Glucksberg, S., Gildea, P. & Bookin, H.B. (1982). On understanding
non-literal speech: Can people ignore metaphors. JVLVB, 21, 85-98.
Noveck, I. Bianco, M., and Castry, A. (2001). The costs and benefits of
metaphor. Metaphor
and Symbol 16, 109-121.
Sperber, D. & Wilson, D. (1986). Loose Talk. Proceedings of the
Aristotlean Society, 86 (1985-1986), 153- 171. [Reprinted (1991) in S.
Davis (ed.) Pragmatics: A reader, 540-549. NY: Oxford University Press].
This archive was generated by hypermail 2b29 : Fri Jun 29 2001 - 16:53:11 GMT