Re: The Homunculus Problem

From: HARNAD Stevan (
Date: Tue Jun 04 1996 - 22:47:30 BST

> From: "Perry, Yvonne" <>
> Date: Sat, 25 May 1996 14:23:02 +0100 (BST)
> The homunculus problem is basically "Do we need a homunculus for
> understanding imagery?" A homunculus is the idea of a little man in our
> heads which sees the images (and words) we see and understands these
> words and images for us. The problem Behaviourists saw with this
> contention is that it is not an expanation of how we understand
> imagery, it is just a repetition of the original problem.

So far, fine.

> Theorists such as Remez believe that our imagery detecting functions are so
> complex that they require something such as the homunculus to process them.

Does Remez imply there is a little man in our heads, thinking for us?

> Theorists such as Fodor believe that we rely on our homunculi do do our
> cognition for us.

Again, do they mean we have extra minds in our heads? Then what's going
on in those minds?

> However, not everyone agrees that a homunculus is
> necessary for seeing images.
> Artificial intelligence is often cited as an example of how the
> homunculus is not necessary. The problem with the homunculus is that it
> never comes up with computation. It just happens.

What you should have said is that there is no homunculus problem with
computation. The homunculus is unnecessary; computation is mindless and
mechanical rule-based symbol manipulation.

> When faced with the
> propostion: "John is taller than Mary. Jill is shorter than Mary." if
> we are then asked "Who's taller- John or Jill?" we may conclude that we
> require a homunculus to work out the problem as it requires us
> envisaging the characters. However, if we used dolls to make a PHYSICAL
> comparison we are discharging the homunculus.

What do you mean? How do we use dolls in our heads? And who uses them?

> Behaviourists believe that processing the problem in the head is no
> good. They do not believe in the homunculus. Proposionalists believe
> that it is acceptible to work out the problem in one's head, but also
> that we should get rid of the homunculus.

This reply is rather vague and in some places seems confused; kid-sib
certainly couldn't get an understanding of the homunculuc problem from
this unless he alreday knew it in advance.

See the skywriting and skyreading on this topic to sort it out.

This archive was generated by hypermail 2b30 : Tue Feb 13 2001 - 16:23:45 GMT