Another possible definition of ambiguity: an input that is
perceived in two different ways at the same time. To say that it
is perceived means that there are two perceptual structures,
well-defined, that already exist, and that the input satisfies
the conditions needed for both of them to produce a perceptual
signal.
This definition applies outside language, too: If I get in my
car, drive it 25 feet, and turn it off again, two perceptions are
produced: (1) I have uncovered the part of the road where the car
was, and (2) I have covered the part of the road where the car
now is. Both results are outcomes of the action; most likely,
only one (at most) is an intended outcome. If I were "punning,"
both could be intended outcomes -- using one action to create two
perceptions.
The Test would be needed for a second party to find out which, if
either, was the intended outcome.
The observed speech varies from the socially standardized
references in ways that can at best be described
statistically...
(I think this perspective, familiar to linguists, answers
Bill's qualms about Labov's work.)
The Test would be needed for a second party to find out which, if
either, was the intended outcome.
In the case of nonverbal perception, the ambiguity is in the
observer's perceptions, with no basis for assuming that it is
also in the observed's. This is why you must say "which, if
either" (or "if any"). Did the person moving the car even
perceive the space that the car covered, before or after moving
it? What if she was controlling line of sight, or shade? She
could move the car a few feet or a car length without looking at
the surface before or behind the car.
In the case of language, the ambiguity, the alternatives
of structure (information), is a socially established property of
the utterance. You need some version of the Test to determine
which, but you would never hedge the question with "if any".
(The version of the Test typically is to proceed with
conversation as though the interpretation you have assumed--
often without considering that there might be alternatives-- is
the correct one. If there's a wrong assumption, error might
arise in one or both of you farther on. If you never notice, it
didn't matter [yet]. In other words, one's acting on
expectations in conversation and cooperative interaction
constitute many repetitions of the Test, though not intentionally
so, because of which results are often misinterpreted. And so
on, yakatayakatayak.)