dialect in novels

[Dennis R. Preston]

>Arnold is right, of course, especially about the "psycholinguistic
>effects" of "real" garden path sentences. They influence processing
>(and the other "path" is hard to get to). I wouldn't like to count
>an "I have.." or lots of other "garden pathlets" as such items, for
>they surely cause none of this processing difficulty. My point was
>to distinguish them from ambiguous sentences, for "The horse raced
>past the barn fell" is unambiguous since it "cannot" mean "The horse
>raced past the barn."

dInIs



>bethany dumas:
>   >>dInIs:
>
>   >>The horse raced past the barn collapsed.
>
>  >I have always used
>
>  >The horse raced past the barn fell.
>
>  >What's the original?
>
>the original is: The horse raced past the barn fell.  from
>articles and books by tom bever (and various collaborators) in
>the early 70s.  (i'm at home in bed with a nasty cold/flu and
>so can't check my library, but if anyone really really needs to
>know exact details, i could supply them next week.)
>
>the effect is *not* created by just any sentence whose initial part is
>capable of several very different continuations.  such sentences are
>everywhere.  *any* sentence beginning Subj HAVE could be continued
>with the form of HAVE understood as a main verb (possessive HAVE, as
>in "I have a Camry"; causative HAVE, as in "I have my students write
>essays on the verb HAVE"; affective HAVE, as in "I had an asteroid
>crash into my car, omigod"; obligative HAVE, as in "I have to go now";
>possibly others) or as an auxiliary verb (perfect HAVE); up to the
>point where the form of HAVE is uttered, the initial portion is
>many-ways ambiguous, but this seems to cause no problem for listeners,
>who process on, holding these possibilities in mind until later
>material determines an interpretation.  (speakers do this with
>ambiguous words in general.)
>
>garden-path sentences are different.  speakers seem to strongly
>prefer one interpretation, and then get hung up if the other one
>turns out to be the right one.  in the standard examples, the
>preferred interpretation is the one with the least embedded structure
>in it ("raced" interpreted as the verb of the main clause rather
>than as the verb in a postnominal modifier).
>
>there are lots of neat wrinkles in this.  for one thing, the effect
>shows up robustly only when the verb form is plausible either as the
>verb of the main clause or as a postnominal modifier.  sentences
>beginning "a dressmaker fashioned" will be understood with "fashioned"
>as a verb with subject "a dressmaker", but one beginning "a dress
>fashioned" will be understood with "fashioned" as a modifier of "a
>dress".
>
>the point is that in the garden-path phenomenon, ordinary people
>processing speech in real time tend to boggle.  (you can detect
>boggling experimentally via, for example, the time it takes for
>people to decide whether some sentence they are presented with,
>visually or orally, is a sentence of english.)
>
>arnold (zwicky at csli.stanford.edu)

--
Dennis R. Preston
Department of Linguistics and Languages
Michigan State University
East Lansing MI 48824-1027 USA
preston at pilot.msu.edu
Office: (517)353-0740
Fax: (517)432-2736