Jackson's Mary 

The story of Frank Jackson's Mary is a philosophical thought experiment proposed by Jackson in his article "Epiphenomenal Qualia"(1) and extended by him in his "What Mary Didn't Know"(2). The argument is intended to motivate what is often called the "Knowledge Argument" against physicalism.

The thought experiment was originally proposed by Jackson as follows:

"Mary is a brilliant scientist who is, for whatever reason, forced to investigate the world from a black and white room via a black and white television monitor. She specializes in the neurophysiology of vision and acquires, let us suppose, all the physical information there is to obtain about what goes on when we see ripe tomatoes, or the sky, and use terms like 'red', 'blue', and so on. She discovers, for example, just which wavelength combinations from the sky stimulate the retina, and exactly how this produces via the central nervous system the contraction of the vocal cords and expulsion of air from the lungs that results in the uttering of the sentence 'The sky is blue'. [...] What will happen when Mary is released from her black and white room or is given a color television monitor? Will she learn anything or not?"(1)

In other words, Jackson's Mary is an omniscient scientist who by stipulation knows everything there is to know about the physical science of colour, but has never experienced colour. The question that Jackson poses is - once she leaves her black and white room and experiences the colour red for the first time, does she learn anything new?  And if she does learn something new, then what exactly is it that she learns?

More formally laid out, the following argument is contained in the thought experiment:

(P1)  Physicalism is true, and all mental phenomena are in some way really physical.  In other words, all knowledge is physical knowledge.

(P2)  Mary knows all and every piece of physical knowledge in regards to human colour vision, prior to her release from the black-and-white room.  In other words, Mary knows all the knowledge that is available about colour vision.

(P3)  Upon leaving the room and witnessing the colour red first-hand, she obtains new knowledge -- what experiencing the colour red actually "feels like".

(C)  There was some knowledge about human colour vision Mary did not have prior to her release.  Therefore, not all knowledge is physical knowledge.  And therefore, (P1) (physicalism) is false

But Jackson's description of this thought experiment contains a subtle trap -- in an inviting equivocation of meaning for the concept "physical knowledge".  Because the (P1) premise of the formalized argument is not explicitly stated by Jackson, he leaves available two different ways that the concept of "physical knowledge" can be understood.  On the one hand, if one accepts the implied instruction to hold physicalism true (to consider P1 to be true), then the concept of "physical knowledge" must, ex hypothesi, include the physical basis underlying the first-person "feels-like" experience of seeing red for the first time.  The subtle trap arises because Jackson never offers a suggestion of how Mary could learn that information, and because we cannot conceive of how reading about stuff in a text book could lead Mary to accurately imagine what the experience of red would be like.  Jackson sets the trap by side-stepping this difficulty - he simply stipulates, by fiat, that she has acquired the necessary knowledge. 

On the other hand, if one glosses over the implied instruction to hold P1 true, then the concept of "physical knowledge" can be limited to only the third-person objective (cum scientific) knowledge -- stuff that is usually meant by the term "physical knowledge".  Then, necessarily, Mary does learn something when she acquires the first-person phenomenal experience that goes along with the third-person physical knowledge.  So the apparent paradox presented by Jackson's Mary can be seen to rest on an equivocation.

Most people, including most philosophers, approach this thought experiment with an already preconceived notion of the answer.  Most people unconsciously employ a dualist mind-set when they envisage the kind of knowledge that Mary possesses -- separating the third-person objective "scientific" knowledge from the first-person subjective "feels-like" phenomenal knowledge.  They unconsciously do not consider the kind of first-person subjective "feels like" knowledge acquired by an experience of the colour red as falling within the label of "all and every piece of physical knowledge in regards to human colour vision".  So most people fail to follow Jackson's implied direction to consider physicalism true and will admit, readily or with angst, that Mary does learn something when she first experiences the colour red.  And this answer poses a serious problem for those philosophers who would rather be physicalists.

But as pointed out by Daniel Dennett(3), people who approach this though experiment with the unconscious adoption of a dualist theory of subjective consciousness are not following the instructions laid down by Jackson.  If one reads the article within which Jackson presents his though experiment, he clearly intends to start off with the assumption (if only for the sake of argument) that physicalism is true.  Otherwise, there is no point to his though experiment.  In other words, without the starting presumption that physicalism is true, the though experiment begs the question.  From the context of the article, the rational (cum charitable) assumption is that Jackson knew this.

If you start off with the assumption that the subjective first-person "feels like" experience is not amongst the physical knowledge available to Mary, then obviously she learns something upon her release from her room.  And obviously, by both pre-supposition and by logical conclusion, physicalism is false.  But if physicalism is true by pre-supposition, then necessarily, the subjective first-person "feels like" experience must fall within "all and every piece of physical knowledge in regards to human colour vision".  And that means that before she leaves the black-and-white room, Mary must know, by stipulation, what the subjective first-person "feels like" experience of red will feel like -- since, ex hypothesi, whatever it "feels-like" is physical. 

The trap awaits the unwary because Jackson does not provide the reader with any visualizable means that Mary might employ to acquire all that first-person subjective "feels-like" physical knowledge she is stipulated to possess.  He does provide some suggested means of her acquiring all the third-party scientific physical knowledge.  But he does not provide any clues to how she might acquire the first-person subjective-perspective knowledge that he stipulates must be physical.  He simply stipulates that she has acquired it.  And that leaves the reader primed to fall into the trap of assuming this first-person subjective-perspective "feels-like" knowledge cannot be among the "all and every piece of physical knowledge in regards to human colour vision".  What is lacking, in other words, is imagination.

Despite pretending to hold (P1 -- Physicalism) true, there have been a number of attempts to provide explanations of what Mary might learn when she leaves her black and white room.  Nemirow(4) and Lewis(5) present the Ability Hypothesis.  David Lewis in the postscript to "Mad Pain and Martian Pain" comments:

.... knowing what it is like isn't the possession of information at all. It isn't the elimination of any hitherto open possibilities. Rather, knowing what it is like is the possession of abilities: abilities to recognize, abilities to imagine, abilities to predict one's behavior by imaginative experiments (p. 131).

Lawrence Nemirow presents almost the same view:

Knowing what an experience is like is the same as knowing how to imagine having the experience (p. 495).

The Ability Hypothesis argues that Mary only obtains an ability (knowledge-how) to do something, not the knowledge (-that) of something new.  Lewis calls it an ability to "remember, imagine and recognize."  This hypothesis assumes that however an ability is physically realized in the brain (and if physicalism is true, then it has to be physically realized in the brain -- baring the possibility that "muscle memory" is not a brain phenomenon), it would not be captured in "all the physical knowledge" that Mary, by stipulation, has already been granted.

Earl Conee(6) presents another alternative - the Acquaintance Hypothesis.  This sets up a third category of knowledge.  Acquaintance knowledge is separate from knowing-that (factual knowledge) and knowing-how (ability knowledge).  Conee argues that what Mary learns post-release is just acquaintance knowledge.  Knowing something by acquaintance "requires the person to be familiar with the known entity in the most direct way that it is possible for a person to be aware of that thing".  Since experiencing a property is the most direct way to become acquainted with the property, Mary gains acquaintance with color qualia after release.  She does not acquire any new item of propositional knowledge by getting acquainted with qualia.

Another similar approach to dealing with the apparent paradox of Mary is the Phenomenal Concept Strategy.  According to Loar(7), Carruthers(8), and Tye(9), phenomenal concepts are recognitional concepts of experience. A recognitional concept, unlike a theoretical concept, contains only the subjective perceptual experiences of acquaintance with its instances.  And according to Perry(10) and O'Dea(11), phenomenal concepts are a form of indexical concepts -- in the same manner as "I", "this" and "now" are indexical.  Phenomenal concepts thus share somewhat the same approach as does the Acquantance Hypothesis of Conee.  But the idea of the Phenomenal Concept Strategy is that what Mary learns when she leaves her black and white room is a new set of concepts around which to organize all the physical information she already has about colour qualia.  Same physical facts, different organizational hierarchy with which to think about them.

But again, these two alternatives both assume that the physical facts about whatever physical substratum underlies acquaintance knowledge or conceptual concepts are not included within the "all the physical knowledge" that Mary has already been granted.

Daniel Dennett argues in response to these alternatives(12) that Mary would not, in fact, learn something new on her release.  Dennett asserts that if she already truly knew "everything about colour" or "all physical knowledge" as Jackson stipulates in his thought experiment, that knowledge would necessarily include a deep understanding of why and how human neurology causes us to sense the "qualia" of color.  Moreover, that knowledge would also a deep understanding of the knowledge base underlying both abilities and acquaintance -- which themselves must (ex hypothesi) be realized in physical phenomena.  Dennett argues that Mary is stipulated to already have available to her all the necessary physical knowledge that underlies all the possible abilities, acquaintances, and phenomenal concepts that these suggested alternatives propose she might learn on her release.  There is nothing left for her to learn.

Even Howard Robinson(13), in his critique of Dennett's solution to the puzzle of Mary, falls into the trap set by Jackson.  His objection is -

"But she will not know by looking at the banana what physical colour impression that is, what input of light waves it is giving her unless she already has some way of relating how it experientially feels and her current physical states. It might be argued that she knows what the physical stimulus is as part of knowing 'all the physical information', but here the ambiguity enters.  . . .  She may have 'written down in exquisite detail, exactly what effect a yellow object would have' on her nervous system, but she couldn't tell by looking at an object whether it was having that effect, so wouldn't know that that look was the yellow or the blue one. Dennett's implication that Mary could, through physical knowledge, acquire the ability for direct recognition of colours is mistaken."

Dennett's argument is that if physicalism is presumed to be true, then Mary's "ability for direct recognition of colours" (whether by knowledge-that, knowledge-how, knowledge-by-acquaintance, or by phenomenal concepts) is necessarily part of the physical knowledge she is stipulated to possess.  Of course Mary can tell by looking at a red object whether it was having the proper "reddish" effect on her.  By stipulation of the thought experiment, that knowledge and ability is part of the physical knowledge she is credited with having.  Of course Mary, before she leaves her black and white room, already knows what it "feels-like" to see red for the first time.  By stipulation, what it "feels-like" is a physical process in the brain, and Mary is stipulated to already know all about that sort of thing.

The only other alternative approach to Mary's circumstances, is to challenge the stipulation that Jackson makes in his thought experiment.  One of the possible readings of the alternatives described above is as a denial of Jackson's stipulation that Mary knows "all of the physical knowledge".  Each of them, in their various ways, segregates out of "all physical knowledge" a particular kind of physically implemented knowledge.  Any of these alternative hypotheses can be seen as an argument that it is simply not conceivable for Mary to have acquired "all of the physical knowledge" that Jackson stipulates she has acquired.  And this denial approach meshes well with our intuitions on the matter.  Learning an ability -- like riding a bike, or playing concert piano -- is simply not at all learnable by acquiring knowledge-that by reading books.  Gaining an intimate acquaintance with the "qualia" of experience, no matter how strongly based on physical phenomena, can also not be learnable by acquiring knowledge-that by scientific investigation.  Setting up some phenomenal concepts just does not seem possible without the "feels-like" experience to ground their indexical nature.  But all that the denial approach to the problem of Mary does, is surface the ambiguity in the concept of "learn".  Jackson's thought experiment merely stipulates that Mary has already acquired all of the necessary physical knowledge.  He does not mention if or how she has learned all that.  And that is the entry to his trap.

Ultimately, the answer one gives to the question of what exactly Mary will learn on her release will depend upon how one views the Mind-Body relationship.  If one is pre-disposed to consider that physicalism is true, then one will conclude that Mary learns nothing that threatens physicalism -- thus "showing" that physicalism is true.  But if one is pre-disposed to consider that physicalism is false, then one will conclude that Mary does learn something that threatens physicalism -- thus "showing" that physicalism is false.  All based on whether one is inclined to include within the scope of "all the physical knowledge" the first-person subjective "feels-like" experiences.  In other words, all based on whether one is pre-disposed to think that the subjective first-person conscious experience is nothing more than the physical reactions of the physical components within the physical brain.

Any conclusion one reaches in response to the essay's title question is a reflection of the starting premise about physicalism that one starts with.  Does Jackson's Mary in fact learn something, and just exactly what (if anything) does she learn?  Any conclusion you reach is based on the fallacy of begging the question.  It would be more profitable to go back and address the underlying premise (of physicalism) directly.


Notes & References

(1)  Jackson, Frank;  "Epiphenomenal Qualia" in The Philosophical Quarterly, Vol 32, No 127 (Apr. 1982), pp 127-136, URL=<http://www.jstor.org/stable/2960077>

(2)  Jackson, Frank;  "What Mary Didn't Know" in The Journal of Philosophy, Vol 83, No 5 (May 1986), pp 291-295, URL=<http://www.jstor.org/stable/2026143>

(3)  Dennett, Daniel C.;  Consciousness Explained, Little Brown and Company, Boston, Massachusetts, 1991, ISBN 0-316-18065-3. Pgs 399-400.

(4)  Nemirow, Lawrence. "Physicalism and the Cognitive Role of Acquaintance" in Mind and Cognition: A Reader, W. Lycan (ed.). Wiley-Blackwell, Oxford, England, 1991. ISBN 978-0-631-16763-1.

(5)  Lewis, David;  Postscript to "Mad Pain and Martian Pain", in his Philosophical Papers (Volume 1), Oxford University Press, New York, New York, 1983. ISBN 978-0-195-03204-8.

(6)  Conee, Earl;  "Phenomenal Knowledge" in Australasian Journal of Philosophy Vol 72, No 2 (1994). Pp.136-150.

(7)  Loar, Brian;  "Phenomenal States" in Philosophical Perspectives, Vol 4 (1990), Pp 81-108.

(8)  Carruthers, Peter;  Phenomenal Consciousness: A Naturalistic Theory, Cambridge University Press, Cambridge, England, 2000. ISBN 0-521-78173-6. Kindle Edition.

(9)  Tye, Michael;  Color, Consciousness and Content, A Bradford Book, MIT Press, Cambridge, Mass. 2000. ISBN 0-262-20129-1. Kindle Edition.

(10)  Perry, John:  Knowledge, Possibility, and Consciousness, A Bradford Book, MIT Press, Cambordge, Mass. 2001. ISBN 978-0-262-66135-5.

(11)  O'Dea, J;  "The Indexical Nature of Sensory Concepts" in Philosophical Papers, Vol 31, No 2 (2002), pp. 169-181.

(12)  Dennett, Daniel;  "What RoboMary Knows" in Phenomenal Concepts and Phenomenal Knowledge, Alter, Torin (eds), Oxford University Press, Oxford, England, 2006. ISBN 0-19-517165-9.

(13)  Robinson, Howard;  "Dennett on the Knowledge Argument" in Analysis, Vol 53, No 3 (Jul 1993), pp 174-177, URL=<http://www.jstor.org/stable/3328467>


Graham, George & Horgan, Terence;  "Mary, Mary, Quite Contrary" in Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition, Vol 99, No 1 (May 2000) pp 59-87, URL=<http://www.jstor.org/stable/4321045>

Howell, Robert J.;  "The Knowledge Argument and Objectivity" in Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition, Vol 135, No 2 (Sep 2007), pp 145-177, URL=<http://www.jstor.org/stable/40208745>

Kallestrup, Jesper;  "Epistemological Physicalism and the Knowledge Argument" in American Philosophical Quarterly, Vol 43, No 1 (Jan 2006), pp 1-23, URL=<http://www.jstor.org/stable/20010220>

Wikipedia contributors;  "Knowledge argument" in Wikipedia, The Free Encyclopedia. URL=<http://en.wikipedia.org/w/index.php?title=Knowledge_argument&oldid=516972513>.