Then we can know about color experiences — Harry Hindu

Well, no... but we can work out what a metamer is and a theory of protanopia and deuteranopia. I'm pretty sure the things you're looking for are somewhere in the visual cortex.

Biologists tell us about the color experiences of cats and dogs, why not humans?

@Harry Hindu

... but to simply conclude that the color experiences are the same because we're all human sounds to me more like guesswork. — InPitzotl

It doesn't approach being a proof, but, IMHO, it is a plausible hypothesis. Per my earlier post, I would not propose that everyone's experience is the same in detail, but that for most people, there is a broad degree of functional equivalence. For example, I am not a synaesthete, and so will clearly experience some things differently than those who are. I suspect, however, that these differences are at the level of iris color or hair type: for one thing, evolution is very conservative about things that are important to fitness, and our minds, and the way they interact with our environment, seem to fall into that category. For another, there is empirical evidence in a broad agreement that certain experiences are pleasant/desirable and others are most definitely not.

What arguments are there for the proposition that everyone experiences things fundamentally differently? Or do the three of us just have different intuitions about how different they are?

Biologists tell us about the color experiences of cats and dogs, why not humans? — Harry Hindu

So? We can talk about color experiences of humans too... we agree on color categories in those 95% of individuals I described earlier. But that says nothing about the experience being the same... only that the categories are.

it is a plausible hypothesis. — A Raybould

...and so is its null hypothesis.

For example, I am not a synaesthete, and so will clearly experience some things differently than those who are. — A Raybould

...color grapheme synesthetes (for example) still see color... they're just able to associate graphemes with color. Plausible explanations for color-grapheme synesthesia are here. The existence of conditions like synesthesia only seem to beef up the plausibility of the null hypothesis.

What arguments are there for the proposition that everyone experiences things differently? — A Raybould

I could think of arguments, but that's irrelevant. The only argument necessary for my strong defense of a non-position is that the null hypothesis is plausible (see below).

Or do the three of us just have diffferent intuitions about how different they are?

You misunderstand. I don't have a theory of similarity, differentness, or even commensurability of experiences of color. What I have, instead, is a standard... an expectation of what kinds of things I need to see before I start doing silly things like believing a thing, that no random guy with some strong opinion and bad arguments is going to get me to lower. And as unfair as it may seem to be, I expect others to meet this standard before I can grant them my approval of their conjectures.

So far the arguments I've heard for same experience is, we're the same species, and evolution selects for fitness. Okay, sure, we are, and it does, and yet I can't use that argument to prove we all have green eyes, so how are you going to use it to demonstrate that we experience red the same way? It's simply not good enough... you need an argument that actually argues for what you're arguing for.

:up:

it is a plausible hypothesis.
— A Raybould
...and so is its null hypothesis. — InPitzotl

And often the null hypothesis is right! We seem to be agreed, then, that this hypothesis is both plausible and possibly correct.

What arguments are there for the proposition that everyone experiences things [fundamentally] differently?
— A Raybould
I could think of arguments, but that's irrelevant. — InPitzotl

Firstly, I see that the word 'fundamentally' is missing from my words without having been replaced by ellipsis. If you want to claim that everyones' experiences may be different in ways that we may not agree are fundamental, then we can go with the alternative that I proposed, and simply agree to differ.

At this point, my working hypothesis is that you do not have any plausible arguments for that proposition. Your next sentence reinforced that view, and your final paragraph did not lead me to change my mind.

The only argument necessary is that it's plausible — InPitzotl

Coincidentally, that's what I said about @Harry Hindu's point! - but plausibility itself does not come out of nothing; it needs an argument in support. A proposition without a justification states something that is conceivable, but not anything more.

What I have, instead, is a standard... an expectation that I have of a theory before I start doing silly things like believing it — InPitzotl

In scientific inquiry, theories are preceded by hypotheses. You don't have to believe them, just consider them. Of course, if there is contrary evidence rendering a hypothesis unviable, then it is summarily rejected and science moves on, but that does not seem to be the case here.

Thanks! I am surprised by the lack of attention that Churchland's paper gets, as, in my opinion, he has identified the fundamental flaw in Jackson's argument (Terence Horgan appears to be the first person to raise the issue of equivocation, but only in passing, and narrowly.) Even though Jackson himself has changed his mind on the issue, his argument is still debated (at least two English-language books were published on the issue last year!) but that debate seems to be mostly over the "old knowledge in a new guise" responses, which to me seem to be moot, given Churchland's straightforward rebuttal.

I tend to favour the ability hypothesis mentioned in the Wikipedia article, which is that Mary gains knowledge-how rather than knowledge-that, namely, knowledge of how to use the word 'red'. However, I would add the proviso that Mary does not actually gain any knowledge (knowledge-how) at all until she learns that the colour she is seeing is called 'red', or in other words, until she learns how to connect the red sample with the word 'red'. This cannot be expected to be learned automatically.

I think that most proponents of that view would say that she would be able to remember colors, compare colors, and even perform Hume's task of imagining an intermediate hue between two that she is seeing, before ever learning any names for them. It is mildly interesting to wonder if she would divide up the spectrum as we do (at which point, I must say that I have always regarded the distinction between indigo and violet as questionable!)

I prefer the Wittgensteinian view that colour samples have a function similar to (or equivalent to) that of grammatical rules. The ability to remember and know colours without names seems a little too close to a private language.

,
I just thought of something. Let's go back to this:

Then we aren't talking about knowing red, rather we are talking about knowing what it is like for Harry to see red
— Harry Hindu
Correct. — InPitzotl

If its possible that every one of us could experience a different color when looking at the same thing, then the scribble, "red", doesn't refer to an actual color, rather it refers the the experience an individual has of some color when looking at something. So, what would it mean for Mary to understand "red", if not that she understands that it is merely a scribble that refers to a particular experience, not a particular color, when looking at a certain object.

In this sense, apples are actually red, or wavelengths are red, in that particular wavelengths give rise to a particular experience that is consistent. So when we refer to red wavelengths, we aren't referring to a color, rather a common cause for our experiences.

Nobody has ever seen a brain state, furthermore, whether it is possible to establish correspondences between so-called brain states and first-person experience is exactly the point at issue. — Wayfarer

several years ago I watched a documentary on the BBC in which a man was put into a fMRi machine and presented with two cards to choose from. the computer analysing his brain was consistently able to predict the choice he would make, before he made it.

for the purposes of the documentary the timing was the issue, that his brain knew before his mind knew that he would choose the cross not the circle. for the purposes of this thread, the point is that an analysable brain-state equated to a specific experience. I'm not overly familiar with the particular quandary being debated in this thread, but that Mary has a brain-state that corresponds to her experience of seeing Red does not seem to be an issue.


Kaarlo Tuomi

Let's try to circumvent Churchland's attack without buying into the theoretical apparatus that Churchland and Jackson buy into:
1. Mary at time T is in possesion of a complete theoretical apparatus that allows for the prediction of all future events, or for the QM theorists lurking, allows for the prediction of the probability distribution of all future events
2. As a matter of definition only, anyone with that kind of knowledge knows all the physical facts.
3. Lemma: from 1 and 2 Mary at time T is in possession of all the physical facts.
4. Mary, up to time T, has never seen any colours other than white and black and the 50 or so shades of grey in between.
5. Mary at time T+x sees with her own eyes a red tomato, and not by seeing a black and white image of a red tomato, nor through any other kind of technological apparatus.
6. Mary gains something at T+x that she did not possess at T.
7. Given Lemma 3 and premise 6, what Mary gains is not knowledge of a physical fact.

Premise 6 must either be denied or accepted, whether one be a physicalist an idealist or some wishy washy compromise between the two. How do we motivate that it must be accepted? "Well, put yourself in Mary's position" is the standard technique to do that. So, we imagine ourselves living in a world which really has just the colour range of old film noir, and suddenly we stumble across a red tomato, or perhaps a pair of red shoes. That might pump the intuition that our lives are suddenly richer than they were, and so we have gained something.

If we accept that intuition, and intuition is all it really is, we then face the problem of describing what exactly it is that we, and Mary, have actually gained. Here dualists/idealist and so on have the upper hand, since they have a more or less unrestricted ontological menu to choose from. Physicalists, though, need to be careful. They need to express what it is Mary gains, but they need to do that in such a way that the description falls outside of the complete theoretical apparatus mentioned in premise 1. Just to give a taste of the problem here, suppose the conversation between a phyiscalist and an early incarnation of Frank Jackson goes like this:
Physicalist: Mary gains an ability.
FJ: OK sport, what ability?
Physicalist: The ability to recognise red things.
FJ: She could alread recognise red things. Tomatoes are red things and when she was given black and white pictures of vegetables and asked to pick out the tomatoes, she had no problems.
Physicalist: That's not what I meant. I meant she gained the ablity to recognise red things on the basis of their colour, not their size or shape.
FJ: But she could already recognise red things on the basis of their colour, not their size or shape. Mary has some remarkably sensitive measuring equipment and given a room of coloured objects, on the basis of the wavelengths of light they are reflecting, and accounting for the ambiant conditions, she is able to pick out those reflecting in the 700–635 nm range, and those are usually red things.
Physicalist: That's not what I meant either. I meant she gains the ability to pick out red things just by looking at them.
FJ: Are you sure about that? What if the next tomato she sees is an unripe one and she is just using her eyes? You and I would call it green. Unless Mary can rush back to her equipment, she's probably going to lump it in with the red things.....hell of an ability she's gained there, sport.

Obviously, the physicalist will have a retort.

Let's cut to the chase, and start with one issue that seems to render everything else moot - your conclusion:

7. Given Lemma 3 and premise 6, what Mary gains [is] not knowledge of a physical fact. — jkg20

This seems to come back to what I said earlier about equivocation over the phrase "physical knowledge":

In Jackson's usage, he is referring to knowledge of physics, but physicalism does not imply or depend on all knowledge being about physics; what it does imply is that knowledge cannot be gained without there being physical change within the entity doing the learning. — A Raybould

I am unable to read your lemma 3 in any other light than you are making the same equivocation as Jackson here, and taking "physical knowledge" to be "knowledge of physics", as opposed to "knowledge which has its information content encoded in physical form." If this is not so, could you clarify your argument on this issue?


By the way, in 1927 Bertrand Russell penned a concise aphorism that captures what you are saying here (if I have not misunderstood you), and also the essence of Jackson's argument:

It is obvious that a man who can see knows things which a blind man cannot know; but a blind man can know the whole of physics. Thus the knowledge which other men have and he has not is not a part of physics.

As far as I know, Russell never regarded this as a refutation of materialism, and rightly so, as to mistake it for such would also require the same equivocation.

Essentially yes, but...

Firstly, we must recognize that people have been talking about colors for millenia, and, for the most part, doing so without having any theories about how the perception of color works (or, in some cases, demonstrably false ones).

As I mentioned elsewhere, language depends on a common vocabulary, and common vocabularies ultimately depend on shared experiences. We do not have direct access to what is going on in other peoples' brains, and not even to what is physically going on in our own. Nor can we directly sense the wavelength of the light we see or the energy of its photons. About the only shared experience of color comes from people looking at a given thing (say, a ripe tomato) and agreeing that a certain aspect of that experience should be given a name, such as 'red'.

For this to work, several conditions must be met, such as:

[1] Most people should agree that there is such an aspect to a tomato.
[2] Most people should agree on what else is red, such as blood or the embers of a fire.
[3] Most people should agree on things that are not red, such as the sky.
[4] Most people should agree that, nevertheless, the sky also has an aspect that is, in some useful sense, in the same category as the redness of a tomato.
[5] Peoples' color categories should be stable over time, barring any change in the thing being observed.

...and so on. Given that not everyone is able to join in this agreement, it seems that our being able to come to a broad agreement over these issues is a contingent fact, and that it is conceivable that this might not have been so.

This is where my database analogy comes in. It is completely implausible that most of our brains are identical at the neuron level, yet still we meet the conditions of the above list. In my opinion, we are like the ordinary users of the two databases, who can agree that their databases contain the same information, even though the two databases have almost no physical state in common. For the purposes of the ordinary users, who use the databases in performing tasks in the external world, the physical state is irrelevant. Only the system admins have any reason to be concerned with that state, and when we debate the nature of the mind, we are, as it were, logging on as administrators.

---

The takeaway from this is that almost nothing about how the mind works can be clarified or settled by arguing over the definition of the words we use to label our experiences, because those words were coined operationally, in ignorance of the underlying processes.

At this point, my working hypothesis is that you do not have any plausible arguments for that proposition. — A Raybould

Your working hypothesis doesn't work.

Your next sentence reinforced that view, and your final paragraph did not lead me to change my mind.

Did you bother to try? At the highest level of abstraction this seems incredibly simple to me. We just postulate that there's a large number of potential experiences and that the brain, while learning to see colors, simply picks any of these to bind to the interesting categories. Synesthesia demonstrates that there are different kinds of experiences to pick from, and that they can at least be useful in some people as category tags even cross-sensually. My natural response to this is, if there are this many kinds of experiences, how many can there be? Incidentally, what made you make up your mind in the first place such that it would need to be changed?

but plausibility itself does not come out of nothing; it needs an argument in support. — A Raybould

But the argument here is, at least IMO, trivially made. The arguments given for same-experience to me sound like classic textbook hasty generalization. It seems you're describing an approach that is particularly vulnerable to argument from personal incredulity, and is way too quick on the belief button for my tastes.

(FYI, there's a lot of other stuff you're saying I've no problems with; it's just that I'm only going to pick on where we conflict).

At this point, my working hypothesis is that you do not have any plausible arguments for that proposition.
— A Raybould
Your working hypothesis doesn't work. — InPitzotl

All the evidence to that point (notably, the absence of any argument) supported it.Whether the situation has improved with your latest post remains to be seen.

But the argument here is, at least IMO, trivially made. The arguments given for same-experience to me sound like classic textbook hasty generalization. It seems you're describing an approach that is particularly vulnerable to argument from personal incredulity, and is way too quick on the belief button for my tastes. [my emphasis] — InPitzotl

That's not an argument, it is an opinion. Furthermore, it is an opinion of an argument, not an opinion about how minds work. You are as entitled to your opinions as the rest of us, but don't expect anyone to find anything like this, doubly removed as it is from being an argument for a substantive claim, to be persuasive.

Synesthesia demonstrates that there are different kinds of experiences to pick from, and that they can at least be useful in some people as category tags even cross-sensually. — InPitzotl

I am amused by the fact that your only substantive argument in this whole reply makes use of one that I first introduced to this discussion! Note that I was prepared to introduce a counterpoint to the position I favored, and that is because I am interested in getting to the truth, not in scoring points in some minor debate. Your question, "did you bother to try?" is best addressed to yourself.

We just postulate that there's a large number of potential experiences and that the brain, while learning to see colors, simply picks any of these to bind to the interesting categories. — InPitzotl

What does 'binding' mean? How are 'interesting' categories determined?

Furthermore, with "a large number of potential experiences" to pick from, unless the picking is done in some principled manner, one should expect that utter chaos would result, with people unable, for example, to agree even on what sorts of sensory perception there are. In practice, we find broad agreement, with only a few corner cases, such as the synesthesias that I mentioned, to hint of the potential chaos. How, in your theory, do you resolve this issue?

Conventional epistemology makes a distinction between propositional/knowing-that knowledge and various other forms, such as knowledge-how, (e.g. knowing how to ride a bicycle, which you don't learn from a book), and knowing a person or a place by acquaintance. For a while, the leading response to Jackson was the ability hypothesis, which claims that, on her release, Mary gained, not propositional knowledge, but certain abilities, such as how to recognize, recall and compare colors. These are generally considered to be examples of knowing-how, not knowing-that/propositional knowledge. These days, the favorite form of reply seems to be "old knowledge in a new guise": when Mary sees colors, she does not gain any new knowledge, but sees her existing knowledge of physics in a new way. To me, this seems both implausible and unnecessary, an attempt to explain away something that already has a straightforward explanation, so I am disinclined to describe it further. — A Raybould

I'm in way over my pay grade here. I'll just say that I'm a big believer in the importance of qualia to any theory of mind, and opposed to those who say you can have intelligence without self-awareness. I'm sure I'm muddling many philosophical issues.

That's not an argument, it is an opinion. — A Raybould

You're really missing the point. "Evolution is conservative therefore we all experience redness the same way" is also an opinion. Opinion dismissal is no justification for another opinion; all opinions should be held to the same standard, even if you happen to hold one of them.

Furthermore, it is an opinion of an argument, not an opinion about how minds work. — A Raybould

Not... exactly. This is a guideline to applying relevant standards. You're arguing about differences and similarities of experiences between humans, under the presumption of a physicalist interpretation. So let's start here... is there reason for you to disagree that the key factor to look at here is the neural correlates to experiences?

You are entitled to your opinions, but don't expect anything like this, doubly removed as it is from any substantive issue, to be persuasive. — A Raybould

But convincing you isn't supposed to be the goal. The goal is supposed to be applying a valid truth criteria. In this case, the criteria of truth should have something to do with neural correlates to experience... would you agree?

ETA:

What does 'binding' mean? How are 'interesting' categories determined? — A Raybould

Binding in this case simply means creating some sort of stable state at some layer in the network; "interesting categories" can be self-discovered by features of the net analogous to how deep neural networks work. I'm not too interested in fleshing out this theory since I don't particularly subscribe to it... I'm more interested in gating the justifications to the things that they're supposed to be including (here is an ancient video including presentation of self-discovered categories using deep neural nets; but here is a link to a paper discussing the kinds of things I think should be included in any speculation).

The only thing that was to the point in your previous-to-last post was that you finally put up an argument (or, rather, the outline of one) in support of the position you have been taking: viz., your insistence that it is simply not plausible that sensory experience functions in a similar manner for most of us.

We just postulate that there's a large number of potential experiences and that the brain, while learning to see colors, simply picks any of these to bind to the interesting categories. — InPitzotl

It is not clear to me what you have in mind here, though it seems to be suggesting a chaotic process - but whether that is so may depend on what you have in mind when you say "a large number of potential experiences" (do you mean the totality of an individual's sensory inputs at any given time? Are emotional states included? Conscious thoughts?), on what basis you envision the picking to be decided, and how the categories are chosen for binding. At least until you explain these matters, it is not clear to me that you have a viable argument for your position.

I'm not overly familiar with the particular quandary being debated in this thread, but that Mary has a brain-state that corresponds to her experience of seeing Red does not seem to be an issue. — Kaarlo Tuomi

Indeed. That is achieved by measuring something called 'voxels' which depict oxygen consumption and other parameters of brain activity. It's heavily dependent on predictive algorithms which are used to validate the presumed correspondence between various cognitive acts and the associated neural activity. So it's not difficult to see issues with confirmation bias and the like in this kind of scenario.
See Do you believe in God, or is that a software glitch?

I think that most proponents of that view would say that she would be able to remember colors, compare colors, and even perform Hume's task of imagining an intermediate hue between two that she is seeing, before ever learning any names for them. It is mildly interesting to wonder if she would divide up the spectrum as we do (at which point, I must say that I have always regarded the distinction between indigo and violet as questionable!) — A Raybould

To add to my previous rushed response, even if we admit that one gains the ability to remember and compare colours, it is unclear how or whether this is a gain of knowledge of any sort. What knowledge-that or knowledge-how does one gain from this purely mental ability? How might one prove that they are able to remember or compare colours?

I defined what I meant by being in possession of all the physical facts: it is to have the theoretical apparatus sufficient to predict every physical event or fact. There is no equivocation in the argument I gave, which does not even end with the conclusion that physicalism is false The argument simply points out that if one accepts that Mary gains something the question arises, for everyone: what does she gain? She does not come into possession of a new physical fact, since by hypothesis she was already in a position to predict, and thus be in possession of, all the physical facts that take place when she sees the tomato. If you are a physicalist, and you accept that Mary gains something, which as I pointed out you do not have to accept, you are constrained to provide a description of what she gains that i. does not introduce anything new that is not already quantified over by the theoretical apparatus Mary already has, and ii. nevertheless correctly accounts for what Mary now has or is now able to do.

It is not clear to me what you have in mind here, though it seems to be suggesting a chaotic process — A Raybould

Why would you need to appeal to chaotic processes?

but whether that is so may depend on what you have in mind when you say "a large number of potential experiences" — A Raybould

Start simple. Imagine we develop a fourth cone type and manage to develop tetrachromacy.

At least until you explain these matters, it is not clear to me that you have a viable argument for your position. — A Raybould

You're making the wrong appeal to the wrong person. I'm not playing the debate game; in fact, I'm skeptical of that entire game. Convincing you isn't the prize I'm after. Tossing that away, the only reading of that statement left is that you're declaring dependence upon me to question your premises for you, which I read as a bad thing.

So it's not difficult to see issues with confirmation bias and the like in this kind of scenario. See: Do you believe in God, or is that a software glitch? — Wayfarer

that was an extremely interesting read. the dead salmon is both very funny and extremely frightening at the same time.

thank you

Kaarlo Tuomi

Purely by chance, I noticed that you added a postscript to your last-but-one post after I had replied to it, in which you offered some sort of response to my requests for clarification of something you had postulated. I'm prepared to put that down to concurrent editing, but there are some unusual aspects to that response.

For one thing, you say "I'm not too interested in fleshing out this theory since I don't particularly subscribe to it", which is rather an odd thing to say about something you had postulated only a few hours before as if it were a strong argument for your position. I will assume that you were not disenchanted with it at the time you first postulated it, as to put forward a view that you did not actually hold, without making that clear, would be quite deceptive, so it seems that you changed your mind during the time when I was probing its relevance. How long, I wonder, would you have left me under the impression that you still held that view?

You also added links to a video and a paper, and the latter, at least, does seem to have some relevance (Categorical encoding of color in the brain)

Here we have an experiment using functional MRI to study the responses of a number of subjects to sensory stimuli. This is a very common sort of experiment, yet If it were the case that everyone's responses to stimuli were more-or-less unique, such studies would fail to produce consistent results. Not only do they do so, but the whole tenor of the paper (and others like it) is infused with the tacit assumption that this commonality in function is a fact of how brains work - there is no hand-wringing over whether the subjects will be similar enough in their responses to justify the approach, and I would bet that the possibility that it would not be is rarely an issue in the funding of such experiments.

This, of course, is commonality at the level of abstraction probed by functional MRI, which does not prove that the commonality must hold at higher levels of functional abstraction, but it certainly does not suggest the contrary. You have already accepted that hypotheses do not have to meet the standards of evidence needed to be classed as theories, and that you don't have to believe them to consider them.

As for your examples of synesthesia and tetrachromacy, it is not unreasonable to consider them as second-order effects. I don't think many papers reporting this sort of work contain the caveat that their results cannot be extended to these cases.

Given the results of this experiment, perhaps you consider color categorization to be a counterexample to the hypothesis that most people's responses to stimuli are broadly similar?

There is no equivocation in the argument I gave, which does not even end with the conclusion that physicalism is false [my emphasis] — jkg20

Well, if you haven't reached the conclusion that physicalism is false, then you have not completed your task of, as you put it, circumventing Churchland's attack [1]. Jackson makes no bones about the goal of his argument: it is to demonstrate that physicalism is false, and Churchland's response says that it fails to do so.

Churchland accepts the premise that Mary has all the knowledge that Jackson claims she has, and also that she learns something new after her release, but this presents no challenge to physicalism, because what she learns afterwards is not within the set of physical knowledge as defined by Jackson (and you), and physicalism neither depends on nor implies that it should be.

Churchland uses Jackson's equivocation over "knows about" to make this point: all the knowledge Mary gains from her studies must be propositional (she is being denied any opportunity to learn about color by any other means, such as experience), while it is entirely plausible that what she learns afterwards is not propositional. In the paper, which I urge you to read, he makes a biological argument for the plausibility of this new knowledge being non-propositional (one of the things he points out is that there are a great many species of animal having trichromatic vision, and for which, therefore, there is presumably something that is "what it is like" to see colors, but do not have language; presumably, they are not believing in propositions of any sort.)

Churchland also makes a point that people often get wrong: he is under no logical obligation to provide a complete and factually correct account of exactly what Mary would learn on her release. While that would be one way to refute Jackson, there are others, and showing that the argument is invalid is how Churchland does it here. As Jackson is claiming to have proven that physicalism is false, he has assumed the burden of showing that there is no way it could be true (to this, I would add the point that no dualist has ever given an explanation of how their alleged non-physical aspects of minds work.)

As Tiim Crane has pointed out, the issue of validity can be turned into an issue of soundness, by adding a further premise, such as "after her release, Mary gains propositional knowledge about color vision." Unsurprisingly, Jackson, who overlooked this issue, made no argument for it.

In retrospect, Churchland was not as tightly constraining as he could have been, and by calling Jackson's "all the physical knowledge" propositional, he encouraged some dualists to claim that Mary might learn something propositional on her release. As they have not been able to say what that is, despite the fact that there have been billions of people who are both language-capable and know what it is like to see colors, I doubt it.

This is moot, however, as, whatever else Mary's physical knowledge is, it must be something one can learn from books or lectures (Alter calls this 'discursively learnable.') The additional premise, then, becomes "after her release, Mary gains discursively learnable knowledge about color vision." This puts dualists on the horns of a dilemma: Unless what she learns on her release is discursively learnable, Churchland prevails, but if it is, then how come she did not learn it from her studies? It would be begging the question to just assert the premise that it must be nonphysical discursively-learnable knowledge.

Note that Jackson never equivocates in the sense of saying, in one place, that "learns about" means this, and in another place, that it means something contradictory to that: it is, instead, implicit in how he uses the phrase. He does essentially the same, or so I propose, with the phrase "physical knowledge." He (and you) define it esentially as "knowledge about physics", and if Mary learns something on her release, it cannot be physical knowledge, therefore physicalism is false. It is in the "therefore" that we have a false implicit premise, which is that, as this new knowledge is not knowledge of physics, physicalism must be false - but physicalism neither implies nor depends on all knowledge being knowledge of physics. The English phrase "physical knowledge" is ambiguous, as, in addition to this reading, it might also be read as "knowledge, where the information content is encoded in physical form", but Jackson's argument cannot just be restated to use that definition consistently, as one would be begging the question to assert that what she allegedly learns after her release is not in that category.


[1] Update: As soon as I posted, I realized that you probably meant that your argument went beyond proving that physicalism is false. Unless it succeeds at that, however (and I do not think it does), then anything else that follows is moot.

Even if we admit that one gains the ability to remember and compare colours, it is unclear how or whether this is a gain of knowledge of any sort. — Luke

If it is not knowledge, then Jackson's argument fails, because it depends on her gaining knowledge that is not in the set of "all the physical knowledge." More conventionally, it is regarded as a different sort of knowledge: knowing-how instead of knowing-that. As Jackson's concept of physical knowledge is of the knowing-that sort, physicalism is not challenged by her new knowledge not being part of it.

How long, I wonder, would you have left me under the impression that you still held that view? — A Raybould

For one thing, you say "I'm not too interested in fleshing out this theory since I don't particularly subscribe to it", which is rather an odd thing to say about something you had postulated only a few hours before as if it were a strong argument for your position — A Raybould

Ah, I see why you were confused now. But I think maybe you want to read this post a bit more carefully before suggesting that I might have misled you.

so it seems that you changed your mind during the time when I was probing its relevance. — A Raybould

Nope, I did not change my mind. I still say that "evolution is very conservative about things that are important to fitness, and our minds" is a bad argument for same-experience theory.

there is no hand-wringing over whether the subjects will be similar enough in their responses to justify the approach, and I would bet that the possibility that it would not be is rarely an issue in the funding of such experiments. — A Raybould

This sounds a bit fishy to me. It sounds like you're guessing that the researchers secretly agree with you, "betting" that they do, and then appealing to this secret agreement.

As for your examples of synesthesia and tetrachromacy, it is not unreasonable to consider them as second-order effects. I don't think many papers reporting this sort of work contain the caveat that their results cannot be extended to these cases. — A Raybould

But hang on, that's the wrong take-away. Let's focus on tetrachromacy a bit more... if a human develops tetrachromacy, would you think it reasonable that said human would have more color experiences? I mean either it's that, or somehow the same color gamut just gets "redistributed", or we just happened as human trichromats to max out on the number of potential color experiences (or, supply your own other?) If tetrachromacy does lend to more color experiences though, then there are indeed more potential color experiences to be had. So the next question to ask is, do we need to add more stuff to the brain in order for it to be able to have these additional color experiences, or would the brain somehow wind up with them if we had four cone types giving us richer information? We could ask the same question on lower levels (e.g., how the ganglia forming color opponent channels in the retina would organize), and that leads to questions about, gee, how do they organize anyway, in us trichromats?

Do you see what I'm getting at? None of this actually argues for different-experience, but you certainly can't just handwave same-experience without understanding how this stuff works in the first place. All of these are good questions, and they really require more study.

Given the results of this experiment, perhaps you consider color categorization to be a counterexample to the hypothesis that most people's responses to stimuli are broadly similar? — A Raybould

I think that question is making assumptions that are not yet justified. A response to a stimulus is behavioral. What effect would having different kinds of experiences of similar stimuli have on the behavior? If I experience red a different way than you experience red, wouldn't we still both call red things red?

If it is not knowledge, then Jackson's argument fails — A Raybould

Yes, I agree.