Assumptions about Communication

[Sherman Wilcox]

Somehow, I knew when I hit that "Send" button Tom would come back at me!

Tom, I actually don't think we're saying very different things. I certainly
never meant that no animal communication is intentional. If horses and
chickens exhibit intentional communication, I have no problem with that. If
intentionality is deeply rooted in our brain's evolutionary history, that's
fine too.

The one place I think you might have overstated the case is where you say
"there is no bloody way in the work [world?] meaningful 'communication' can
take place without the intention to communicate." You'll have to explain the
qualifier 'meaningful', of course, but it seems to me that some pretty
meaningful communication *can* take place without the intention to
communicate. At least that's the possibility I'd like to leave open for now.

When Noel says, "Surely information without intention is not information" it
makes my head hurt (in a good way). I'm not trying to be difficult, but I
think we need to be careful with what we mean by information. I've been
trying to teach myself to think of information not in terms of
*instructionist* models but rather as *selectionist* (the distinction is
made by G. Edelman and H. Plotkin). And, as for my skepticism about the
impossibility of communication taking place without the intention to
communicate, I think I'm skeptical that information without intention is not
informative. (And I just now see that David Tuggy might also question this
claim.)

So, all I really meant in my original post was that I think it's worthwhile
to investigate intentionality and communication separately (I haven't read
the DuBois article, but I bet I'll like it when I do), to distinguish
intentionality from intentional communicate. For example, I'm not sure I
want to say that from the start infants intend to communicate. I think
intention to communicate develops, quite possibly because caregivers treat
infants' vocal behavior *as if* it were intentionally communicative. It's
not communication because the child intended it to be, it's communication
because the caregivers sanctioned it to be.

In other words, I think it's worthwhile to consider the possibility that
intentional communication develops ontogentically. And if so, that it
develops phylogenetically. So it becomes then a matter of looking for
precursors. I take it this is what Matt Cartmill (physical anthropologist)
is talking about when he says:

"To understand the origin of anything, we must have an overarching body of
theory that governs both the thing itself and its precursors. Without such a
body of theory, we have no way of linking the precursor to its successor,
and we are left with an ineffable mystery, like the one that Chomsky and
Lenneberg have always insisted must lie at the origin of syntax."

Apply this to the emergence of intentional communication and I think it
leads us to consider a concept behavioral ecologists rely on when they
discuss the evolution of communication, that of 'intention movements': "many
signals have evolved from incidental movements or responses of actors which
happened to be informative to reactors. Selection favoured reactors who were
able to anticipate the future behavior of actors by responding to slight
movements which predicted an important action to follow" [the quote is from
"Behavioural Ecology : An Evolutionary Approach" by J. R. Krebs & N. B.
Davies].

I want to entertain the hypothesis that this ability "to anticipate the
future behavior of actors" is a precursor to intentional communication.
While these movements communicate, and are intentional (although I bet some
of them aren't), they are not intentionally communicative (which is why the
authors above use the phrase "happened to be informative").

What I like about this view, though I don't see that the animal
communication people ever recognize this, is that it places the emphasis not
on production, the intention to communicate, but on comprehension, the
ability to garner information (I might even say the ability to generate
information on the part of the perceiver/reactor) from signals produced,
intentionally or not, by others. Not unlike the child example above, the
reactor "treats" the actor's behavior as if it is communicative.

A pleasant side effect of focusing on comprehension rather than production
is that it shifts our perspective from a monologic to a dialogic one. If we
dwell too much on production, we too easily forget that information is in
the eye of the beholder (it's that instructionist/selectionist difference
again). But by focusing on the comprehender, the reactor who is selected
because she was able to "put meaning into" the movements of actors, or the
caregiver who treats the child's behavior as if it were intentionally
communicative, we necessarily have to consider the dyad.

And I'll be darned if I don't think all of this leads to the work of
Rizzolatti that's getting so much press these days, on mirror neurons and
the link between action, the capacity to recognize action, vision, and
gesture.

Thanks, folks. This is fun!

-- Sherman