I just read about Mary the color scientist and qualia on wikipedia because someone had mentioned it in the consciousness thread.
My summary for this thread:
The “Mary” argument seems trivially weak "All the information" is ambiguous
“…all the physical information there is to obtain about what goes on when we see…”
Unless having “all” the information includes experiencing something, then she doesn’t have “all” of the information.
If the assumption is that having “all” the information is equivalent to experiencing something, then a gigantic assumption has been made about how we acquire information that (IMHO) is completely inaccurate. When we learn “about” something, we build an abstract model with placeholders for concepts, etc. When we experience a sensation, we have different neurons firing, firing together, synapses being created and strengthened, etc. than when we learn about something in an abstract sense. Trying to argue that these two different forms of information are the same seems like a stretch.
To me it seems obvious that either:
Mary didn’t really acquire “all” the information
If the thought experiment is to pretend that acquiring all the information means that through magic she stimulated her cones to be sure to have that specific information contained within her brain, then she experienced something similar to color already
Am I missing something that would make this thought experiment a compelling argument of some sort?
The idea is that prior to seeing anything red, she knows all the physical information there is to know about redness*. This term “physical” is supposed to limit her knowledge to just the kind of information a physicist would talk about–particle positions and momenta, fields, quantum states, etc–as well as anything that can be logically deduced from such information (together with whatever stipulative definitions make for the best macro-organization of the physical world). The argument says that since she could know all of that kind of information without having any idea what its like to see something red, it follows that the experience of seeing red is not a physical existant. If it exists, it’s not physical.
Yes, but “knowledge about” is not the same as knowledge gained from first hand experience. It is different information.
It doesn’t really “follow” at all that seeing red is not a physical “existant”, because it’s not a logical argument. Mary’s knowledge is mostly independent of the knowledge gained from direct experience, they barely overlap. I’m really struggling to see how any adult could have entertained this logic to the point of it having an entry in wikipedia.
Of course, people who use Mary’s Room to argue for the truth of non-physical facts would agree with you.
Well, here’s the argument put formally:
Physical facts are either facts the form “fundamental physical constituent X has fundamental physical property Y” or else facts which can be derived logically (given stipulative definitions of non-fundamental entities) from facts of that form.
The fact “This is what it’s like to see red” (entertained or uttered by a human being upon seeing something red) is not a fact of the form “fundamental physical constituent X has fundamental physical property Y.”
The fact “This is what it’s like to see red” can not be derived logically from facts of that form, no matter what stipulative definitions of non-fundamental entities one may care to use.
Therefore, the fact “This is what it’s like to see red” is not a physical fact.
Physicalism is the view that all the facts are physical facts.
Therefore, physicalism is false.
The Mary’s Room scenario serves as an “intuition pump” (I think) to illustrate the truth of sentence 3 above. Give Mary all the time in the world, they say, and she’ll never have derived the fact “This is what it’s like to see red” from the fundamental physical facts. How could she? You could give her a purely physical description of say, a dog, and since she knows all the right definitions, she could derive “that would be a dog.” But give her a physical description of someone seeing red, and she could derive “that would be a person seeing red” but never (they say) “that would be what it’s like to see red.”
“What it’s like to see red” is something you can’t get from complete detailed knowledge of the physical facts, so it’s not physical. If there is such a thing as “what it’s like to see red,” then physicalism is false (says the argument).
The argument is straightforwardly valid, so the question is, which premise do you disagree with (1, 2, or 3) and why? (You might disagree with 5 but that would be a terminological dispute, probably not ultimately substantive.)
(Disagreement with 1 may look like a terminological dispute as well, but there may be substance to be had there as it might be important to ask what is supposed to be the significance of talk about such things as “physical” facts or physical entities more generally.)
BTW I think what you’re going for is the “acquaintance hypothesis,”
This requires an argument that “this is what it’s like to see red” is not a new proposition, or if it is, that somehow it doesn’t constitute a new fact–that instead it’s just a new way to present an old fact.
Is your proposal to say that sentence 1 in my summarized argument is false, and that the truth is instead that
Physical facts are facts (F) satisfying one or more of the following three conditions:
a) F has the form “fundamental physical constituent X has fundamental physical property Y”
b) F can be derived logically (given stipulative definitions of non-fundamental entities) from facts of the aforementioned form
c) F has the form “W is what it’s like to experience Z for V”?
Assuming I am reading that correctly, and I believe “c” is the key difference from before, then I think that is correct (or at least closer to correct).
Although, I wonder if we can really even talk in terms of “facts”. Sure there are “facts” that we can agree exist, and that our mind seems able to store and retrieve and roll around and play with as a unit, but it also seems that at the same time, our brain contains an enormous amount of information that is simply not stored as “facts” but rather as state/structure. Stuff that can’t easily be identified as a “fact”.
You say the thought experiment with Mary succeeds but I don’t understand why someone would think it does succeed.
The neurons that our eyes connect to, and the way that information enters and resides in the brain is substantially different than taking in information “about” this same thing (seeing red).
Some might object to me making this claim on the basis that I can’t provide a detailed map of a brain showing before and after images with differences in the neural patterns between the act of experiencing vs the act of acquiring information “about”. However, without going into a lot of detail, it seems like a reasonable and accurate impression of what is happening inside my brain. If someone does object to this line of thinking I would ask for an explanation of why they think it’s the same.
The whole Colour Scientist argument *relies *on a an expected failure of the imagination of the argument’s recipient. Dennettis right - if Mary knows (and I quote from Jackson’s original wording “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’”, then Mary is equipped to know *exactly *what to expect when seeing red. The failure is one of introspection and imagination. Mary is not a naif.
The only difference between Mary and a person with “qualia” would be the length of time that the Qualic has had to integrate the experience to the point of banal familiarity. But the MCS argument isn’t whether Mary will have *integrated *the experience into her sensorium, it’s whether it would be a completely new experience for her. Clearly, the bit I bolded makes it not so. If the argument is that Mary needs time (a form of learning, it’s true) to *integrate *the experience, that’s not an argument for qualia (as I see them). If the definition of qualia was “the ineffable way we have integrated our sense experiences in our whole psyche”, that’s be true, but qualia as used aren’t just a synonym for “integrated senses” - they’re ineffable, non-physical, intrinsic, private, apprehended and subjective units of experience, or something.
I prefer the Robo-Mary formulation where Mary has the ability to alter her own neurology to match that of the Qualic, as it presents a closer analogy to the P-Zombie thought experiment that the MCS argument is meant to address and points out its failings easier. But like Dennett, I agree that actualneurologicalexperiments have shown that it’s just failure of imagination to think the MCS argument is convincing for qualia. In order to know “all the the *physical *facts” of colour vision, you ***really ***need to know "all the neurological facts", and that’s where the imagination failure creeps in. Again, like Dennett (I’m getting sick of saying that), I agree that what riles people is that it seems counter-intuitive, but it is true - you must accept the full **implications of Mary “knowing exactly” the neurological processes leading to colour observations. It’s right there in the argument, and it’s even there in the Strong Formal Formulation: “1b Mary knows **all the physical facts concerning human color vision”.
Basically, the tl;dr version: RP, you’re right. Dennett (and I) agree with you, it’s a trivial argument.
For me, in philosophy, *whenever *an argument contains the phrase “It seems just obvious”, as the MCS one does, run screaming. Dennett spends quite a bit of Consciousness Explained breaking down why this is so seldom the case in neurology (with many real clinical examples) - I *highly *recommend reading this book (although, if pressed for time or money, a lot of the chapters find their parallel in his papers available online).
Yes. Although you should realise that Dennett originally coined the term *pejoratively, *to describe arguments like the Chinese Room or MCS which are effectively (if often unintentionally) *designed *to lead to wrong conclusions by relying on us to accept what intuitively “seems obvious” and ignore the *actual *difficult implications of an argument’s premises. The MCS is a classic argument of this form.
Dennett does stress that pumps can (in fact, should) be useful tools in explaining ideas and cutting away fat from an argument, too. I guess they’re like the Force that way.
Well, of course, I don’t believe that the thought experiment succeeds, but evidently, a lot of people do, some of them quite sharp tacks, so that needs explanation. And I must confess that my initial gut reaction is much the same as what the argument aims to elicit – the belief that, of course seeing colour for the first time must be something new to Mary, even though she has all the physical information about colour-seeing. Your gut reaction appears to be that she could not possibly have all that information – which is basically attacking the hypothetical (which I guess is fine, as it does posit a rather ludicrous situation). Perhaps I’m just more prone to accepting an idea presented to me on its own terms.
But you’re right, doing so implicitly requires making some assumptions, first and foremost that in order to acquire all the information about something, the channel by which it is acquired doesn’t matter. Now, I could see how this might be wrong, and likely is in the human case (after all, we react differently to the word ‘BOO!’ printed out on a screen than to somebody screaming it behind us), but my first reaction would be to trust this assumption – after all, any information (well, physical information at least) can be coded into a string of bits, which can be transferred via any channel that is able to transfer information at all. So it’s not that unreasonable to expect Mary – with her genius intellect, impeccable memory etc. – to be able to truly gather all the information about colour through channels other than seeing it. (However, it was this same assumption that I attempted to attack in an earlier thread on the subject.)
Another assumption is that all information is equally useful to the brain, no matter how it was acquired, i.e. that the information gathered by reading about a subject doesn’t just sit in the ‘reading-centre’ of the brain afterwards, but that it can be referred to by any part of the self. Without this assumption, the argument again breaks down: all the physical information is present in Mary’s brain before seeing colour, it’s just not in the vision part of the brain. It can only enter there through actually seeing it, at which point her vision centre will indeed learn something new, but that ‘something new’ will be a copy of information already present in other parts of the brain, and hence, only boring old physical information.
This assumption however seems superficially borne out by everyday experience. First of all, we certainly seem to ourselves to be a cohesive whole; everything that’s part of me is part of me, not just part of part of me. Also, we generally have conscious access to both visually and textually received information in much the same way – whether you tell me ‘Bobby plays ball’, I see Bobby playing ball, or I read ‘I’m playing ball’ on his twitter, I have, if asked what Bobby does, equal access to the knowledge item gathered, and can accurately report his playing ball. So one might expect Mary to have similar universal access to the knowledge gathered.
But there’s good evidence that we are far less integrated than we generally seem to ourselves (and hopefully, others), the classic example being blindsight – a condition in which, through neurological damage, a person looses all or part of their sight, but can, if interrogated, nevertheless ‘guess’ what is present in their field of view with better than chance accuracy.
However, I think the real reason behind the success of Mary and other thought experiments like it is that it just seems so obvious (but see MrDibble’s post!) that our conscious experience, the way it is to me to be me, to experience what it is like to be me, is irreducible to brute physical facts. After all, any detailed account of our brain, of our neurons, their states, the overall electrochemical activation patterns, makes no reference to any ‘self’ anywhere, so how could it contain the information about how it is to that self that doesn’t seem to be there to, say, see the colour red? We think – more: are convinced – that our selves are irreducible, because obviously, to us, it seems that they are – we can not experience, and hence, not imagine (as all imagination is to imagine what it is like to experience something), a world without our selves, as all our experience is predicated on the self. So it is fundamentally beyond our grasp that at some level, there exists a description of the world that makes no reference to selves at all, and yet is the same world!
This is basically the reasoning behind Leibniz’ Mill, which has its modern analogues in arguments like the Chinese room experiment, the Chinese nation (those poor Chinese, unwittingly drafted into philosopher’s thought experiments!), Blockhead, and others:
](http://plato.stanford.edu/entries/leibniz-mind/)So this is the root of the debate, of which Mary is just another, recent elaboration: that any materialist, reductionist account of consciousness seems fundamentally insufficient. Whether or not it actually is, is the question that is being debated when one discusses things like Mary’s room etc.; this is a question that is perhaps not so obviously to settle, and (if one is interested in such things) a rather important one, as well. So viewed in its proper context, it’s perhaps not all that surprising that learned people may want to waste their time on it…
Your objection here is what I would consider to be the whole point of the argument.
The Mary argument is trying to show that there is additional “information” beyond the objective information. That’s all. But this is actually a very significant conclusion.
In studying this subject I would urge caution. Easily 50% of people who come at it fail to grok the problem of consciousness and then declare that the whole thing is a misconception.
I don’t just mean the Dennetts of the world who seem to appreciate the problem but (basically) believe that other philosophers are coming at it from the wrong angle. I mean people who just don’t follow what the problem is even supposed to be.
(Actually some would put Dennett in the latter category…personally, I suspect he gets it…but there is a constant temptation in philosophy to declare “I can’t solve it, therefore it must be an illusion”).
I’m not sure the OP is saying that there is additional non-objective information (which would indeed be the suggested conclusion of the thought experiment); the way I read him, he’s merely claiming that not all objective information can be obtained by ways different from experience, i.e. seeing colour directly, or that it matters for the ‘interpretation’ of the information how it is obtained, i.e. possessing all the objective information obtained directly through experience is different from possessing all the objective information obtained indirectly, through study. In both cases, the argument would indeed fail to support its conclusion, that objective (or physical) information is not sufficient to describe conscious experience.
The Mary argument is trying to show that there are two different kinds of information: objective and subjective.
The argument that Mary does not have all the information if she has not actually seen colour is the whole point.
The OP mentions particular brain pathways needing to be stimulated before she can have all the information. But brain pathways are not information, and I still think this “refutation” is actually just reiterating the point of the argument.
As I have no cones for detecting microwaves, the implication is that I can never have all of the information about microwaves. And if I can’t, science can’t.
That’s a pretty profound conclusion, and an example of the hard problem of consciousness.