[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xmca-l] Speech Acts as Grammatical Metaphors

Of course, it goes without saying that some threads are more congenial
to some readers and others to others. I certainly find a great deal in
xmca that is not particularly apposite my tastes and proclivities, but
when I do I just assume that it is addressed to someone else, and I am
invariably proven right.

For many years, I really couldn't understand why the two linguists I
most admire on this planet, Henry Widdowson and Michael Halliday,
don't get along very well. Or rather, they get along rather as
Vygotsky and Piaget did: Widdowson never stops addressing Halliday and
Halliday never replies (the relationship between Halliday and Chomsky
is rather similar: according to Halliday they once shared a swimming
pool at MIT, and got on famously as long as they restricted their
conversation to politics: Halliday has written a good deal about
Chomsky, almost all of it critical, and as far as I know Chomsky has
not once mentioned Halliday in print).

When I last remarked on this to Professor Widdowson, he replied that
for Halliday pragmatics is part of semantics, whereas for him,
semantics is part of pragmatics. This makes sense to me now. I know
that Widdowson, like many linguists of his generation, was much
influenced by speech act theory and the whole approach of notions and
functions which followed from it. Even today, for Widdowson, text is a
kind of footprint left by discourse, but for Halliday when you take
discourse out of text, there is nothing left but paper and ink.

Perlocutionary force--e.g., what my teachers are REALLY doing when
they ask "Can anyone tell me what this is?"--is a central concern of
pragmatics. But in semantics, we can simply treat it as an instance of
nterpersonal metaphor. We use a yes/no question to stand for a
wh-question in much the same way we use a question to stand for a
command ("Would you mind not chewing on Fluffy's tail?") and for much
the same reason: direct speech acts are often very face threatening,
since when you give a command you are casting your hearer in a servile
role, just as when you ask a question you are casting yourself in the
role of an ignoramus (and this, rather than "BS", is probably one of
the most important barriers to broadening and enlivening discussion on

Because perlocutionary force is separable from locutionary force by
metaphor it is possible to separate out these speech acts and give
them some special status and even to derive a whole theory of
pragmatics from them. But Halliday doesn't do that: he simply treats
them as instances of grammatical metaphor, no different in principle
from the lexical variety where we ask one word to stand in momentarily
for another (e.g. "What walks on four legs in the morning, two legs at
noon, and three legs at twilight?). In this case, however, we are
asking one wording (e.g. a question) to stand in for another (a
command), or in some cases to stand in for some legal or commercial

It seems to me that one way to verify this approach is to imagine what
remains of the speech act when we take away the underlying act (e.g.
when we perform it on stage as part of a play). The answer is,
nothing. On the other hand, we can easily imagine a world where speech
acts are performed by gestures rather than by speaking (e.g. when a
dying patient who has lost the ability to speak gets married). The
speech is removed, and the act remains. It seems to me, therefore that
in a speech act the speech is simply a metaphor for the act.

If speech did not have another function, the function of networking
minds and transforming one mind through another and even transforming
itself through the minds of others, then I suppose it would be
adequate to describe speech as a form of object-oriented activity, and
even to describe semantics as a kind of brain trace left by
pragmatics. But speech does, and so it isn't.

David Kellogg
Hankuk University of Foreign Studies