It may be objected that Mary surely does acquire some new proposi-
tional knowledge on her release from the room (Loar, 1990). For example,
she may learn something she would express by saying, ‘Thiscolour [point-
ing at something red] is warmer thanthisone [pointing at something
yellow]’. The knowledge she thereby expresses is surely the knowledgethat
one colour is warmer than the other. But this is knowledge which she
cannot have had before, since it involves recognitional concepts of colour
(‘thiscolour’) which she did not possess when in her black-and-white
room. But now this is just a dispute about how onetypesfacts. Are facts
diVerent if theconceptsused to describe them are diVerent? Or are facts
only diVerent if the worldly objects and properties involved in them diVer?
The objection to the Lewis argument assumes the former. But then that
just returns us to the confusion between sense and reference discussed
above. As Loar points out, that there are someconceptswhich you can
only possess in virtue of having had certain experiences (namely, recog-
nitional concepts of experience) does not show thatwhat is recognised
(namely the experiences themselves) in any way transcends physical or
functional description.
2.3 Cognitive closure?McGinn (1991) has argued that the solution to the problem of phenom-
enal consciousness (the problem, that is, of understanding how phenom-
enal consciousness can be, or be explained in terms of, physical events in
the brain) iscognitively closed to us. He argues,Wrst, that it is a corollary
of the Chomskian claim that we have a variety of innate special-purpose
learning-mechanisms, specialised for particular domains such as natural
language or folk psychology, that there may besomedomains which are
cognitively closed to us. These would be domains which might actually
contain facts suYcient to answer the questions which we can frame about
them, but where the innate structure of our minds means that we shall
forever be incapable of discerning those answers. So these will be domains
which, while notintrinsically(metaphysically) mysterious, must always
remain mysteriousto us. He then presents reasons for thinking that the
realisation of phenomenal consciousness in physical brain-events is one
such domain.
Now, we are certainly inclined to quarrel with theWrst premise of this
argument. From the fact that our minds contain specialised learning-
mechanisms which make the acquisition of knowledge of certain domains
particularlyeasyfor us, it does not follow that there are any domains which
are cognitivelyclosedto us. It only follows that there are domains where
learning will beless easy. Provided that our special-purpose learning-
mechanisms can also be deployed, somewhat less eVectively, outside of
Mysterianism 237