Re: Behaviourism

From: HARNAD Stevan (
Date: Tue Apr 30 1996 - 18:07:12 BST

> From: Whitehouse Chantal <>
> Date: Tue, 30 Apr 1996 16:47:07 GMT
> The question "what was right about behaviourism". I can't
> really see much that was right about their way of thinking
> at all, apart from the fact that they believed introspection
> to be untestable.

Well, that would already be one thing: That introspective observations
are not really "observables," because the only ones that anyone can
observe are their own. Observations have to be public. Anyone and
everyone should be able to observe them, not just one person.

So behaviourists asked: If introspection does not give you observables,
what ARE the observables of psychology? Their reply was that it was
behaviour, that it, what the organism DOES. We can all see that. And
they were right about that.

[They also conceded that the brain and its activity was also
observable, but they suggested that that was another subject, not
psychology: were they right about that? That's a more controversial
question, and you'll need to come to a judgment, but remember to
support your judgment. If you think observations of the brain SHOULD be
part of psychology, say why, and give examples of what we learn about
psychology from them (brain imaging, for example). If you think they
should NOT be part of psychology, say why (the computationalists say
psychology is about finding the right algorithm that is being
implemented by the brain, but the hardware is irrelevant).]

So there's two, maybe three things. Behaviourists also insisted that we
shouldn't theorise about what's going on in the mind either: Not only is
introspection not a source of DATA (observations); it is not a source of
THEORY either. So we should throw out all theories about what's going on
in the mind. Like the anti-imagery people, they thought that all
"mentalist" theory, everything that spoke about "mental states" (such
as feeling pain, or seeing green, or seeing an image) was homuncular: It
depended on a little man in the head, but no such little man exists.

They were probably partly right about that too: It's still true that a
mentalistic theory is not much more helpful than mental data

The fun starts when you get to what the behaviorists were WRONG about --
but you didn't ask me about that, so I suppose you know...

> The other question I was having trouble expanding on was
> "what are the observables in psychology". Is it enough just
> to discuss the body's visible behaviour (and while doing
> this try to define what behaviour actually is) and brain
> activity or is there something else that I'm missing? What
> about peoples responses to questions or their attitudes or
> is that just another form of behaviour?

Good questions! First, what are people's responses to questions, if not
behaviour? Isn't responding (speaking) something you do? Of course it's
more complicated than that, because when you speak, you're not just
making sounds: What you are saying MEANS something. But what is meaning?
Is it observable? We'll get to this when we get to the symbol grounding
problem. For now, it looks as if the only observables are what people
(and their brains) do.

Attitudes? Have you ever seen (observed) an attitude? You've seen people
BEHAVE as if they had, say, an arrogant attitude. But what you observed
was their behaviour. The arrogant attitude was something you INFERRED
from their behaviour, so it wasn't an observable, it wasn't a datum, it
was a hypothesis. And it could have been wrong: If you see someone
shrug his shoulders (and everyone else sees him do it too, and you have
it on video, to be extra-sure), then that's an observation. It can only
be wrong if your eyes or your video camera were wrong. But the fact that
you saw an ARROGANT shrug is not an observation, it's an interpretation.
It could be dead wrong.

Now attitudes are more than just hypotheses, we know, because we have
attitudes ourselves: Sometimes I can FEEL that I'm being arrogant; in
my own case, it's not a theory, it's an observation. But because of the
other-minds problem, that observation is available only to me, it's not
observable by anyone else. It's not publicly observable. So it's not

So behaviour -- of the whole body, or the brain part of it -- is
observable. But if behaviour and brain structure/function are the only
observables, it doesn't mean they're the only thing there is. Meaning
exists; attitudes exist; the mind exists. It's just part of the
mind/body problem that it's not clear what you can do about them, if
you would like to explain them. Because explanations are explanations
of observables, otherwise how could you ever test whether or not they
were correct?

This archive was generated by hypermail 2b30 : Tue Feb 13 2001 - 16:23:39 GMT