There’s
something about Mary is the first book
devoted solely to Frank Jackson’s knowledge argument (KA). The editors provide
a lengthy introduction, which explains the significance of the KA and
introduces different lines of reply to it. This is followed by Jackson’s
original argument, several of his follow up articles including his eventual
rejection of the KA and by a wealth of contributions by prominent philosophers.

Like
most thought experiments Jackson’s knowledge argument exploits strong
intuitions. The brief outline of the argument (which can be read in variations
20+ times throughout the book!) is as follows: Mary, a brilliant
neuro-scientist, who specializes in color-vision, knows "everything
physical" about colors. But, because she has been locked since birth in a
black-and-white room and can only access the world via black and white monitors
she has never seen a color herself. When she is finally released from the room
and experiences color for the first time it seems clear that she "learns
something new". Thus we have 2 claims:

1. Mary knows everything physical about color vision (knowledge
   claim, KC)
2. Mary learns something new when she sees color for the first
   time (learning claim, LC)

Both claims seem to be true and
thus seem to support the conclusion that physical knowledge "leaves
something out" and hence that physicalism is false.

The
editors have grouped the replies to KA in the following 5 sections.

Part II Does She Learn Anything?
(Dennett, Robinson)

Part III The Ability Hypothesis
(Lewis, Pettit, Tye)

Part IV The Acquaintance
Hypothesis (Churchland, Biglow and Pargetter, Conee)

Part V Old Facts, New Modes
(Loar, Nida-Ruemelin, Chalmers)

Part VI Did She Know Everything
Physical? (Horgan, Stoljar, Hellie, Van Gulick)

Because most of the authors who
attempt refutation concentrate their efforts on LC while KC seems to be almost
unanimously accepted LC is the main focus of these sections.

Dennett’s
response to KA is unique insofar that he does not focus on LC. Instead, he
argues that we cannot fully appreciate what KC amounts to; essentially Mary
knows a lot but not everything about colors. Subsequently Dennett’s claim is
that if Mary truly knew everything physical, she would not learn
something new upon first seeing a red object. So for Dennett the problem is
quantitative, he does not seem to accept that there is a qualitative difference
between "intellectual" knowledge and experiential knowledge.

Most of the other contributors seem
to hold that such a qualitative difference exists and debate whether or not it
has implications for the truth or falsity of physicalism. In part III to V we
find variations of a common theme that can be illustrated in the following
example: I can read a "complete instruction manual" about how to ride
a bike or I can actually ride a bike. In the latter case I will gain practical
abilities and abilities to imagine and to remember that are not accessible to
me in the former case. It is at times difficult to see the relevance of the
examples constructed by individual authors. Most readers will probably be able
to follow Lewis theorizing about a person who never tasted Vegemite (even
though Lewis does not demonstrate that the physiology of vision and taste is
analogous in the relevant aspects, he merely relies on our intuition that it
is).

Pettit attempts to convince the
reader that what Mary learns when she sees red for the first time is comparable
to what a person who has never seen motion (in his example Eve who was raised
in a stroboscopic room) would learn when she sees motion for the first time.
While the cases may be analogous in same aspects a defender of KA may hold that
what is different between the cases is the difference that makes the
difference. Even if we were to accept that no motion qualia exists this does
not prove that color qualia exist neither.

Much of Tye’s argument seems to rest
on the assumption that the conclusion of KA needs to be read bi-directional:
KA: A person who knows everything physical does not know what it is like
to see color. Reverse KA: A person who knows what it is like to see
color knows everything physical. I do not think
Jackson is committed to such a view, in fact several authors point out that
phenomenal knowledge is often pre-theoretic knowledge (e.g. Chalmers: we know
the phenomenological properties of water before we know that water is H₂O).
Thus, for KA it is not relevant whether or not a person can reliably
distinguish between numerous shades of red and correctly remember the
difference between red₁₇ and red₁₈. Rather, Jackson
claims there is a "before and after" difference for a person who sees
any color for the first time. Possibly this difference is smaller for a person
who experiences red₁₇ for the first time but already was familiar
with red₁₈ than for Mary but it still requires an explanation.

The
examples chosen by Bigelow and Pargetter provide the least obvious connection
to the problematic of KA. These authors introduce thought experiments by David
Lewis (which were conceived in a rather different context) to support their
version of the acquaintance reply. They compare the different possible perspectives
of people who hear about a trial in which the defendant and the judge have the
same last name (Hoffman). They analyze what a person learns when he realizes
that in the sentence "Hoffman is just as bad as Hoffman" the first
instance of Hoffman refers to the defendant and not to the judge (or vice
versa). While this may be an interesting puzzle for the philosopher of language
it is not clear how it can contribute to the refutation of KA. In Lewis example
factual verbal explanations are completely sufficient to illuminate all
perspectives of the puzzling case. By contrast Jackson’s claim in KA is that no
amount of factual (physical) information can give Mary the phenomenal
perspective on color qualia.

In
general the opponents of the KA claim that the acquisition of new abilities,
direct acquaintance, and new modes of accessing old facts do not threaten
physicalism because no new factual knowledge is acquired. Proponents of KA
claim that no matter how you call what Mary gains it is something that cannot
be described within the complete physical story and, therefore, physicalism is
false.

Opponents
and defenders of the KA alike seem to agree that once Mary has seen her first
red object she will be able to remember what it was like and to recognize new
instances of red objects. To me this raises an important question: if
there is a non-physical phenomenal component to color experience it is not
clear whether this component has to be always present (as dualist hold) or
whether it is only present when an agent has a particular phenomenal experience
for the first time and/or whether there is a difference in the phenomenal
component of an actual color experience and a remembered/imagined color
experience (as KA seems to indicate). My own suggestion is that neurophysiologic
research and not philosophical speculation will provide the relevant answers.

One can
always argue about the selection and assembly of articles of an anthology and I
suggest that, overall, the editors made good choices. But, I was rather puzzled
about the placement of Robert van Gulick’s superb article at the very end of
the book. Considering that he provides an excellent classification system for
the individual replies to the knowledge argument (which contains most of the
selected articles for this volume) it would have been very useful to read his
article first. This would allow especially the non-philosophical reader to
situate the individual articles within a logical framework.

From
the individual articles I highly recommend Churchland’s excellent combination
of neurophysiology and philosophical argument (and for the psychological
interested reader this may be the only relevant article). Chalmers’ critical
evaluation of numerous replies to the knowledge argument is very informative
(even though I do not agree with his conclusions) and readers interested in
analytical philosophy will undoubtedly appreciate the contributions by Lewis,
Bigelow and Pargetter, Nida-Ruemelin, and Stoljar.

While one might
wonder whether a 456-page book is needed for a 20-year-old argument, it makes a
number of important philosophical contributions. Many contemporary philosophers
hold that thought experiments are essential for testing our intuitions. They
value Jackson’s creation because Mary "initiated" a wealth of responses,
many of them supported by thought experiments of their own. The ongoing
discussion about Mary has resulted in a re-evaluation of the initial intuition.
Not everyone holds anymore that obviously she will learn something new
when she leaves the black and white room. This may be the most important
contribution of Mary: She forced us to find better arguments to defend our
intuitions, or — as in Jackson’s case — to abandon very strong intuitions.

© 2005 Christina Behme

Christina Behme, MSci (Biology, University Rostock,
Germany) is currently a PhD student in the philosophy department at Dalhousie University,
Halifax, Nova Scotia, Canada. Her research interests are philosophy of mind and
psychology, cognitive science, and philosophy of language.

Categories: Philosophical