"Pragmalinguistics" or
the occupation with
pragmatic aspects of
language can be
important where
computational linguists or
artificial intelligence researchers
are concerned with
natural language interfaces
to computers, with
modelling dialogue behavior, or
the like. What
speakers intend with
their utterances, how
hearers react to
what they hear,
and what they take
the words to
mean will all
play a role
of increasing importance
when natural language
systems have matured enough
to cope readily
with syntax and
semantics. Asking a
sensible question to
a user or
giving him a reasonable
response often enough
depends not only on the "pure" meaning
of some previous
utterances but also
on attitudes, expectations,
and intentions that
the user may
have. These are
partly conveyed in
the user's utterances
and have to
be taken into account,
if a system
is to do
more than just
give factual answers to
factual requests.
Thus someone
who wants to
construct a natural language system
might look at
Mey's book Pragmalinguistics with
some expectations as
to what he
should consider or what
guidelines he should
follow, or maybe just
to find out
what the current
state of the
art in pragmatics is.
However, he will
find little of
that in this book.
Pragmalinguistics is a
collection of articles dealing with
many different thingssome
of the articles
could instead of
being called pragmatic
or pragmalinguistic be
labelled sociolinguistic. Most
authors that have contributed
to this volume
are located in Scandinavia, and
thus maybe it
gives a good
impression of the
concerns of North
European linguists. The article
by Bang and
Door gives a
critique of the linguistic
theories of Lyons,
Habermas, Bernstein, and Ehlich
and Rehbein from
a Marxist point of
view. The article by
Qvortrup criticizes Transformational Grammar--and in
particular the Language
Acquisition Devicemagain
from a Marxist
point of view
and with a breath-taking
ignorance of what
TG is all
about. Lindberg's article on
units of speech
is trivial and Utaker's
on semantics is
outdated. Blakar writes on
language as a
means of social power.
His paper is
anecdotal; he draws
conclusions without stating from
what premises; and
he is on the
whole not very
explicit. Gregersen postulates
in his article on the relationships
between social class
and language usage that
an economic analysis
of "objective class positions" has to precede
sociolinguistic studies proper, but
fails to show
how the results
of such an analysis
will influence sociolinguistics. Haeberlin writes
on class-specific vocabulary
as a communication problem.
His ideas have
been published before
and in more
detail. But he
at least makes substantial
and concrete claims,
and he has a
reasonable framework for
his research, even
though he admits that
the results he
has obtained in
his statistical studies are
only preliminary in
nature. He found,
for instance, that members
of the middle
class have a higher
ability to gather
the meanings of
new words in conversations than
members of the
lower class do. Jacobsen
writes on language
and emotions much
from the point of
view of a
psychotherapist. The emphasis of
his article is
more on explaining
emotions than on explaining
the relationship between
language and emotions. Olsen's paper
is on psychopathology, interaction and pragmatic
linguistics. Sondergaard's topic
is the neurolinguistic concept
of the ontogenesis
and disinte- gration of
smooth articulation.
Andersen is concerned
with the syntax
of texts and the
syntax of actions.
He has been
influenced by work done
at SRI International, and
his analysis of actions
resembles the SRI
action graphs. It
may be worthwhile to
look at the
differences in detail.
Bjerg wites on public
speech acts, and
Gloy states some postulates for
a theory of
linguistic manipulation.
Schank's article--the only
one in the
book that carries the
term computational in its title--gives
a summary of
Conceptual Dependency Theory
because he
feels that
is the (only?)
prerequisite for computational pragmatics. In his
closing paper on
critical language theory
Mey points out a
number of phenomena
having to do
with the pragmatics of
natural language that
should be dealt with
by an integrated
linguistic theory.
Pragmalinguistics is a
book with an
unfortunate history, which delayed
its publication for
a long time maybe
for too long.
It is not
very useful for
some- one who expects
concrete results applicable
to the construction of
a natural language
system. But it may
be of
interest anyway, as
it gives a
different (often Marxist) perspective
on linguistic phenomena
that some may not
have considered to
be linguistic phenomena
at all.
Комментариев нет:
Отправить комментарий