• jkg20
    382
    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].
    One can circumvent objections by changing the argument, which is effectively what I did: it is certainly not Jackson's original argument. It is an argument that essentially states that if you accept that Mary gains anything epistemologically, you are obliged to give an account of what she gains, and the account given must not result in Mary getting something that she already had or could have obtained without seeing the red tomato.
    I agree that there is some unclarity about what physicalism is, you mentioned Tim Crane, and one of his earlier papers was on exactly that point. I tried to encapsulate a rough approximation of physicalism, and let me try to be less circular by saying that Mary having complete physical knowledge would give her the apparatus to predict every measurable event in the world, given knowledge of the measurable starting points. Measurable here means measurable by apparatus available to anyone. So, when Mary sees, at time T+x, the red tomato, she is in some overall measurable state. At time T she is able to predict accurately the measurable state she will be in. Let us suppose she does this, i.e. at time T calculates all the measurable effects on her and her environment of her coming to see a red tomato without the aid of optical equipment. She will write down a number of statements such as "At T+x my cones are excited in such a such a way", "At T+x+dx my optical fibres are excited in such and such a way", "At T+x+dx+dx my frontal lobe will be excited in such and such a way" and so on and so forth. At T she has this entire set of time indexed propositions covering the moment she is confronted by the red tomato up to, let us say, two seconds afterwards. We make the assumption also that any abilities she has at T+dx are also included in this set of propositions, perhaps in the form of conditional statements of the kind, if at time T+dx+dz I am again confronted by a red thing, then such a such a measurable effect will be observable. She needs no confirmation of the truth of these time indexed propositons, because her theoretical apparatus is complete and true. I.e: she knows all these propositions.
    The question now is, given that Mary is in that position, does she still gain anything epistemologically at all when she sees the red tomato? That she does is based on the intuition pump, but we do not have to accept intutions, which are after all not much more than guesses. It is a question for the idealist and dualist as much as it is for a materialist monist. The materialist monist has the luxury of responding "no", and thereby rejecting the intuition, but I suppose then has to deal with the persistence of the "illusory" intuition that Mary would gain something. The idealist and dualist must really respond "yes" and, as you hint, need to say something about what she does gain, which usually leads them into hand gesturing about qualia and so forth. However, the materialist monist who responds "yes" also needs to say something about what Mary gains epistemologically and what she gains must meet some criteria. The first of these is, of course, that whatever is proposed should not already be included in the set of all the time indexed propositions previously mentioned, since otherwise it is not something she gains epistemologically. The second is that whatever is proposed is genuinely something epistemological in nature. This is where, presumably, a materialist monist might want to try suggesting that Mary gains some measurable abilities. The problem then is to state what those abilities are in terms that do not quantify over anything not quantified over in Mary's theoretical apparatus, and is also an ability she genuinely gains that is not included in the set of propositions covered earlier. Faced with that, the materialist monist might point out that the assumption was that the set of propositions included all conditionals covering all the abilities that Mary has at time t+x, so the challenge to provide an new ability she gains cannot be met. Fair enough, so either the materialist monist back tracks and says that Mary does not gain anything epistemologically when she sees the red tomato, and so face the question of accounting for the illusory intuition that she does and its persistence, or the materialist monist tries to come up with some other kind of epistemological gain Mary makes that is not based on her gaining abilities or propositional knowledge. FOLLOWING SENTENCES ADDED: Or perhaps the materialist monist might try to deny that abilities can be reduced to conditional propositions the truth of which can be determined by taking measurements. But that looks like admitting that there are abilities that we have that fall outside of the scope of Mary's theoretical apparatus.
  • A Raybould
    86


    Do you see what I'm getting at?InPitzotl

    I think I do, and I think we are talking at cross-purposes. You are saying that these differences exist, and I am saying that they are second-order effects, modifying an underlying commonality. We could both be right! (Or wrong.) Alternatively, we could just agree to disagree in our opinions and wait for neuroscience to achieve a more fine-grained picture than fMRI and related technologies has delivered so far.

    BtW, I recall seeing some fairly convincing evidence (some variant of the Ishihara test), a few months ago, that a small percentage of women are functional tetrachromats, having two different yellow-detecting pigments, but I do not recall where I saw it.


    If I experience red a different way than you experience red, wouldn't we still both call red things red?InPitzotl

    Not necessarily - it seems quite possible that, while we might agree that certain things are red, we might not agree on others. Of course, people do sometimes disagree over subtle distinctions in hue, but that is what I would call a second-order effect.
  • A Raybould
    86


    ...if you accept that Mary gains anything epistemologically, you are obliged to give an account of what she gainsjkg20

    The question of who bears the burden of proof (or accounting) is not always clear-cut, but in this case, I think Churchland's response, especially when put in terms of what is discursively learnable, puts the burden squarely in the dualists' court. I think this is most clear when we adopt Crane's third premise, again put in terms of what is discursively learnable: "After her release, Mary learns a discursively-learnable fact about color." A dualist could simply assert that this is so, but doing so would not put us under any logical obligation to accept it, especially as we could ask the dualist why, if it is discursively-learnable, Mary could not have learned it before her release. If she could have done so, then the premise that she must learn something new does not hold.

    Also, as I have pointed out elsewhere, no dualist has ever given an account of what non-physical whatever-it-is Mary supposedly gains.

    As it happens, however, I think physicalists can give a very straightforward account of why she could not learn, beforehand, whatever she learns from seeing colors. As you propose, Mary, knowing everything that could possibly be known about the physics of color perception, will be able to figure out what changes will occcur in the physical state of her brain as a consequence of any particular color-vision experience. According to the physicalist premise, these changes, and these changes alone, bring about Mary's new knowledge, so all she would have to do, while isolated, is to make those changes happen to her brain state, and she would then know, say, what it is like to see red (or whatever she would learn from the specific experience for which she has worked out what it will do to her brain.)

    Knowing what those changes are, however, is not enough to make them happen. Our brains are simply not connected in such a way that we can address individual synapses and make specific changes to their state. Mary has all the knowledge she needs, but lacks the means to put it to use.

    In this view, what she learns from seeing colors is physical knowledge in one sense: knowledge in which the information content is in physical form. It is not physical knowledge in the other sense of being a true proposition about the physical world (anything that is, she already knows.) The physical knowledge, of color vision, that Mary can learn while isolated, is limited to things of the latter form, so there is no problem for physicalism in Mary gaining physical knowledge in the former sense afterwards.

    This is actually a contingent fact, as it is possible that, at some time in the future, we will develop the technology to make those targeted changes to our brains. With such a device, and Mary's physical knowledge, she could come to learn what color vision is like, without having experienced it (the experience of using such a device might be quite unusual, perhaps like the experiences of those people who must be conscious as brain surgery is performed on them.)

    Update:
    As far as I can tell, these objections work just as well as a response to your argument as to Jackson's. If I am not mistaken, the major difference between them is that you say that Mary's complete knowledge means that she can infallibly predict the future. This takes us into the murky waters of determinism and causality: if she can predict what she will know in the future, why can she not know it now? (Especially as, by premise, it is propositional.) It would be begging the question to say that she cannot do this because it is nonphysical knowledge.
  • InPitzotl
    310
    BtW, I recall seeing some fairly convincing evidence (some variant of the Ishihara test), a few months ago, that a small percentage of women are functional tetrachromats, having two different yellow-detecting pigments, but I do not recall where I saw it.A Raybould
    https://en.wikipedia.org/wiki/Tetrachromacy#Humans
    I have to say though... it's a bit interesting seeing this called a yellow-detecting pigment.
    Not necessarily - it seems quite possible that, while we might agree that certain things are red, we might not agree on others.A Raybould
    I'm a bit lost. You're now speculating that there are behavioral differences?
  • A Raybould
    86

    I think it is sometimes called 'yellow' (if I am recalling that usage correctly) because it is a variant of the more common green opsin, with a spectral response shifted towards longer wavelengths.

    If two people differ in their experience of a class of stimuli, is there any reason to be certain that they will be able to come to an agreement on how to categorize them? (Though it would be easier for them to do so if they were merely second-order differences.) We don't learn what 'red' means by memorizing a canonical chart of all the hues that are red. And there are behavioral differences: for one thing, tetrachromats behave differently than most of us in the tests that demonstrate their particular talent.
  • InPitzotl
    310
    I think it is sometimes called 'yellow' (if I am recalling that usage correctly) because it is a variant of the more common green opsin, with a spectral response shifted towards longer wavelengths.A Raybould
    That actually describes erythrolabe (the L opsin). (I'm not saying anything's wrong here BTW, just that it's a bit interesting to hear talk of yellow-sensing cones)
    If two people differ in their experience of a class of stimuli, is there any reason to be certain that they will be able to come to an agreement on how to categorize them?A Raybould
    Is certainty an appropriate burden for some purpose here?
    We don't learn what 'red' means by memorizing a canonical chart of all the hues that are red.A Raybould
    The color categories are in large part shaped by cone properties. Metamerism for example is an expression of different spectra that have the same effects on the cones.
    And there are behavioral differences: for one thing, tetrachromats behave differently than most of us in the tests that demonstrate their particular talent.A Raybould
    Now I think you're lost. We're talking about two trichromats with similar cones potentially or not as the case may be having different kinds of color experiences. Tetrachromacy comes into play here only by comparative speculation that there might be different modes of experiential color than the ones trichromats have, which suggests different potential experiences.
  • jkg20
    382
    As far as I can tell, these objections work just as well as a response to your argument as to Jackson's.
    It depends. It seems to me the materialist monist would have to reject the premise of my argument that Mary does gain epistemologically when she confronts, or is confronted by, the red tomato. That's an option of course, and one which towards the end of your post, you hint at. However, I am still unconvinced that the materialist monist can accept the premise that Mary gains something and offer a response to the question what. You write:
    "According to the physicalist premise, these changes, and these changes alone, bring about Mary's new knowledge,"
    You seem to agree that what Mary knows prior to the event includes all the changes referred to, so what is Mary's new knowledge that is brought about by them? Is it an ability, is it some new set of propositions, is it some other kind of knowledge? That question does not seem to be answered in the paragraph of yours I extracted that quotation from. You mention only that the physicalist just needs to point out that it is information encoded in physical form, but what is that information?

    As to your burden of proof point, I was attempting to give the basis of an argument that avoids tossing the burden ball. As I say, there is a burden just as much on dualists to give a substantial account of what is learnt by Mary. Idealists as well would need to say something, but would probably opt for denying that Mary gains anything and instead say that the intuition that she does gain anything is based on a false understanding of what it is to know all the physical facts, something that the materialist monist could also try. In any case, idealists have different tricks up their sleeves and don't generally, at least as far as I am aware, use Jackson's argument, or any variant of it, at all.

    Regarding the definition of physicalism I gave, you are correct that it requires Mary to have the theoretical apparatus to predict, with appropriate statistical distributions where necessary, what events will occur in the future. For sure this raises issues about determinism, but that one area of metaphysics should have a dependence on or be related to another area is no big surprise: it is probably only in professional analytic philosophy circles that anyone pretends, or has to pretend, that metaphysics can be cut into bite sized chunks. I suppose it might be said that my argument requires that the timeindexed propositions that Mary knows at T can only involve statistical distributions of event occurrences, and that at time T+x the distribution curve collapses to a point and so Mary comes to know a proposition without any statistical hedging, which she could not have known before. Not clear to me if that would help the materialist monist though, I'd have to think about it.
  • A Raybould
    86

    Is certainty an appropriate burden for some purpose here?InPitzotl

    Feel free to reply at the greatest confidence level you think can be justified.
  • InPitzotl
    310
    Feel free to reply at the greatest confidence level you think can be justified.A Raybould
    "Justified" isn't the point. Purpose of holding this burden is. Usually when I see the certainty burden it's an indicator of a double standard of burden of proof; one holds some opinion A and then one adopts a "prove me wrong" view, with the naive idea that they're open minded because they will consider other opinions, but the pragmatic idea that they hold opposing opinions to A to a much higher standard than opinion A itself. What I'm questioning isn't what burden can be met, but why you are applying this burden.

    FYI, I already listed (response to very next quoted snippet) reason to suspect agreement; I'll back fill this with more detail for you some time later when I get time if you like.
  • A Raybould
    86


    I'll back fill this with more detail for you some time later when I get time if you like.InPitzotl

    I eagerly await your explanation.
  • A Raybould
    86


    It is clear that before we can complete the discussion of your argument, we have to resolve the issue of what would count as an effective counterargument. I do not think I am misrepresenting you when I paraphrase your position as being that a physicalist must either reject the premise that Mary learns anything on first seeing colors, or must give a full (or almost complete) physical accounting of what she does learn.

    This is simply not so. A physicalist can alternatively accept that Mary does learn something, but show that it is beside the point in one way or another, and one way to show that is to make a plausible case that there could be some explanation for why Mary had not learned it beforehand, that does not challenge physicalism - and that's what most of Jackson's opponents do seek to show.

    This sort of thing is common in all fields of incomplete knowledge. For example, in astrophysics, the rates of rotation of galaxies seems inconsistent with what we know of gravity. There are at least two putative explanations: MOND (gravity works differently at long range) and dark matter (there's something we haven't seen yet adding to the galaxies' gravitational field.) Suppose someone says that the rotational data proves MOND. When a skeptic says that dark matter is an alternative possibility, simply saying "so prove me wrong" would not be a reasonable response from the first person, and demanding that the skeptic must say exactly what dark matter is would amount to that, and would be a clear case of burden-shifting. The rational response to the issue is to say that we have these hypotheses, and have not yet figured out what is going on. Jackson is not saying this about the mind; that is not even a point that needs arguing. In his initial paper, he literally claims to have proof of the falsity of physicalism (a position that he subsequently backed slightly down from, and then reversed entirely.) If you are merely saying that physicalists have not yet presented an explanation of the mind, then there are much easier ways to make that point, if you can find anyone who disputes it.

    Let's look at it another way: one of your premises is that Mary knows enough to predict the future in every physical detail, but what, exactly, does she know? It would be inconsistent to consider a lack of a complete explanation to be an objection to the premises of responses, but not to your original premises.

    This is not just my opinion; this is how philosophy (and science) works. This is what Churchland said: 

    "The view sketched above is a live candidate for the correct story of sensory coding and
    sensory recognition. But whether or not it is true, it is at least a logical possibility.
    Accordingly, what we have sketched here is a consistent but entirely physical model (i.e.,
    a model in which Jackson's conclusion is false) in which both of Jackson's premises are
    true under the appropriate interpretation. They can hardly entail a conclusion, then, that
    is inconsistent with physicalism. Their compossibility, on purely physicalist assumptions,
    resides in the different character and the numerically different medium of representation
    at issue in each of the two premises. Jackson's argument, to re-file the charge, equivocates
    on 'knows about'."

    This is far from the only thing I have to say about your previous reply, but I think we need to deal with this one first.
  • InPitzotl
    310

    Okay, so since we keep going back and forth over this point, I've cobbled together the following illustration. Here's a color optical illusion based on Beau Lotto's famous cube illusion:
    Reveal
    fauxlotto.png

    Here, the two color swatches on the large cube indicated by the corresponding color swatches on the gray cube are the same color.
    fauxlotto-cf.png
    This diagram illustrates three different layers of "color"... layer L0 indicates the physical spectrum. Layer L1 is what the cones in your eyes resolve (what I'm calling here colorimetric). Color experiences occur in this diagram at layer L2. The reason for using this illusion is to highlight that L2 colors are distinct from L1 colors, because certainly since these two indicated swatches are the same color spectrum (L0), the cones in your eyes would respond the same way (L1), so any difference you see (experiential) is upstream (in our discussion so far I keep getting the impression that you're confusing L1 with L2).

    L2 colors, mind you, cannot distinguish something if L1 cannot distinguish it (under the presumption that our eyes are actually serving some sort of purpose... that we're not just psychically sensing color independent of our cones). We should for example expect the same color illusion works the same way if we replace the L0 color here with a metamer (another spectral color that our cones cannot distinguish). Likewise, it appears to be common that whatever color experiences we actually have, behaviorally we tend to sense that the L1-colors (in say CIELAB space) seem to vary naturally according to variances in our experiential L2 colors.

    Now get back to the hypothetical tetrachromat. A tetrachromat's L1 color space should be expected to be 4 dimensional just as us trichromat's L1 color spaces are three dimensional. Under the presumption that tetrachromacy works similar to trichromacy, we would expect some 4 dimensional CIELAB-like L1 color space; some trichromat metamers would be distinct L1 colors for such tetrachromats. In this 4D-CIELAB space we might presume L2-colors to vary roughly according to the L1 colors, but there would also be more L2-colors. Presume then as a rough model that we take a 3D slice of this 4D L2-color space that roughly corresponds to one trichromat's L2-space. Would we not be able to take a different 3D slice of this 4D L2-color space and have a second trichromat have this as his L2 colors, and still have the property in both individuals of having each of their L2 colors vary continuously in correlation to the 3D L1 color spaces?
  • jkg20
    382
    This sort of thing is common in all fields of incomplete knowledge.
    But the arguments, Jackson's and mine, are based on the idea that Mary's knowledge is complete, at least complete insofar as to make any examples of epistemic gaps drawin from the current state of physics/chemistry etc irrelevant. Of course the current state of science is far from complete. One way of expounding on what that incompleteness means is to home in on the idea there are observable phenomena that our uncontested theoretical apparatuses do not allow us to predict as being observable. Hence my definitions of what it would be for Mary to have the complete knowledge required of her.

    I do not think I am misrepresenting you when I paraphrase your position as being that a physicalist must either reject the premise that Mary learns anything on first seeing colors, or must give a full (or almost complete) physical accounting of what she does learn.

    I think I allowed myself to be misrepresented, since if you are saying that the two premises of my argument concerning what it would be to have complete physical knowledge are up for grabs as well, sure, I'll concede that. None of the premises are a priori truths. In fact I think I suggested that some idealists would home in on challenging what I am saying would count as complete physical knowledge. I think I can see how an idealist might try responding, but what of a materialist monist? Would the objection be that physics and the other special scienses will never reach completion?

    The point is that if a materialist monist responds along the lines that Mary gains something, then when asked "what" just replies something that is physically encoded but we do not yet know, then the fact that we do not yet know it is irrelevant, since the argument rules out Mary having such gaps, or attempts to. If, on the other hand, a materialist monist says Mary now knows what it is like to see red, and that knowledge is physically encoded, then that means there is some set of propositions about Mary's physical state which describe what is encoded, and so expresses what it is she comes to know. But by hypothesis Mary had the tools at T to predict that those very propositions would be true, with a given range of liklihood perhaps, at time T+x, which puts strain on the claim that she has gained any knowledge. I suppose there is the avenue to explore that she gains extra confirmation of the truth of the time indexed propositions, although she did not need that confirmation. Althernatively, as I put it in my previous post, she gains knowledge that the probability distribution has collapsed onto a specific member of the set of possible propositions.

    Anyway, either complete physical knowledge is possible or it is not. If it is not, and the admission of this has been forced on the basis of arguments involving people coming to know things when they see things for the first time, then dualists have it. If it is not possible for other reasons, then dualists still have work to do. If it is possible to have complete physical knowledge, then I suppose we need to agree on how exactly to spell out what that would mean. I proposed glossing it in terms that provide the tools to predict all observable events, but maybe there are better alternatives.

    Editing my own post and specifically concerning this remark

    " then that means there is some set of propositions about Mary's physical state which describe what is encoded, and so expresses what it is she comes to know. "

    I suppose there is a riposte here that what is encoded can only be stated as knowing what red looks like, and that there is no alternative way of expressing that content, but that admitting this does not entail that the content is not encoded physically. Of course, accepting this might involve having to buy into a representational theory of mind, which personally I do not, regardless of any issues about this knowledge argument. However, regardless of that, is that riposte really available for a materialist monist? If the content encoded can only be expressed as "knowing what red looks like" then there could be no identity expression of the kind "knowing what red looks like is identical to ........ " where ........ must be precisely an alternative way of expressing that same content. You might want to say that materialist monists do not need to make such identity statements, but their monism, in the end, probably does require them to do so. One could be eliminative, but in which case there is no such thing as "knowing what red looks like" anyway, so the eliminativist would simply be rejecting the idea that Mary gains anything when she see a red tomato.
  • A Raybould
    86


    But the arguments, Jackson's and mine, are based on the idea that Mary's knowledge is completejkg20

    Jackson (and also you, I believe) are actually arguing that Mary's knowledge, at the time of her release, is incomplete - but any response along these lines is missing the point of my analogy. It is a simple fact that no-one (neither monists nor dualists of any type) have complete knowledge of how minds work (if any faction did, the issue would be settled.) Given this, then not only is it is inconsistent to rule out one side of the debate on account of its proponents' inability to give a full accounting of how minds work, it is doubly so to accept the view of side that claims to have proof, despite their not being able to give such an accounting, while rejecting the claim of those who merely say that it has not been proven, on the grounds that they cannot provide a full accounting. Until we can get agreement on what would constitute an acceptable response, there is no point in discussing the merits of any argument.

    It would be really useful if, in your next reply, you could address this issue alone, as I am not going to re-state the counter-arguments to any other point you make, only to have them once again rejected for a bogus reason.
  • A Raybould
    86


    Nevertheless, I think we can trim the range of issues to be covered.

    I think I allowed myself to be misrepresented, since if you are saying that the two premises of my argument concerning what it would be to have complete physical knowledge are up for grabs as well, sure, I'll concede that.jkg20

    Personally, I don't think your concept of complete physical knowledge is incoherent, or inconsistent with respect to the way you define it (and the same goes for Jackson's); my point is that "physical knowledge" is an ambiguous phrase, and your definition is not the one that leads to a challenge to physicalism, should Mary learn something more on release.

    As Crane points out, this is a thought experiment, so improbable things are allowed, and if you object to the impracticality of achieving this knowledge, you are simply not engaging with it.


    If, on the other hand, a materialist monist says Mary now knows what it is like to see red, and that knowledge is physically encoded, then that means there is some set of propositions about Mary's physical state which describe what is encoded, and so expresses what it is she comes to know.[my emphasis.]jkg20

    This is a very common and deep-rooted misconception. It simply is not the case that knowing all the facts, about the physical state of one's brain when one has some particular knowledge, is the same as having one's brain in that state, and physicalists can reasonably propose that unless one's brain is in that state, one does not have that knowledge.

    Of course, if that knowledge is not propositional/factual, then there is obviously no way that facts about that knowledge are that knowledge, but even if we limit ourselves to factual knowledge alone, propositions stating facts about something we know are not the propositions comprising that knowledge itself - but I do not think we need to consider the propositional case further, anyway:

    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.A Raybould
  • A Raybould
    86

    Firstly, thanks for taking the time to put together a well-argued case.

    I cannot imagine why you think I am confusing L1 and L2, but there is nothing to be gained by following that any further.

    Last things first...
    Would we not be able to take a different 3D slice of this 4D L2-color space and have a second trichromat have this as his L2 colors, and still have the property in both individuals of having each of their L2 colors vary continuously in correlation to the 3D L1 color spaces?InPitzotl
    Possible? Certainly. Likely? I was already of that opinion, based broadly on the sort of evidence and argument you are presenting here. Necessary? No; if we are going to suppose that everyones' experiences of color are different, without any constraint on how different they might be, then we cannot assume continuous variation, let alone any isomorphism between individuals, or even any stability within a single individual.

    You had asked "If I experience red a different way than you experience red, wouldn't we still both call red things red?" and my reply was, and still is, "Not necessarily." The point here is that the assumption, that we would both agree about the redness of all (or most) things, is predicated on assumptions about our experiences being similar. I am not saying that these assumptions are unreasonable (quite the contrary, in fact; I am making the same assumptions in my own point of view); I am saying that these assumptions are being made.

    And what are we basing those assumptions on? The observed behavior of people (including statements about their experiences) and what knowledge we have of the physics of visual perception! It seems we can make informed guesses, about how the experiences of different people are likely to be similar, from these sources.
  • InPitzotl
    310
    I cannot imagine why you think I am confusing L1 and L2, but there is nothing to be gained by following that any further.A Raybould
    It's because you keep talking about behavioral responses and disagreements on whether all people would agree that particular things are red if they simply have different L2 colors but share L1 colors.
    Possible? Certainly. Likely? I was already of that opinion, based broadly on the sort of evidence and argument you are presenting here.A Raybould
    But hang on... that, too, is the wrong takeaway. There are arguments for same-experience that you have not yet given. There are also counterarguments to those arguments. But in the end you still wind up at my position... that we simply need more study.
    Necessary? No; if we are going to suppose that everyones' experiences of color are different, without any constraint on how different they might be, then we cannot assume continuous variation, let alone any isomorphism between individuals, or even any stability within a single individual.A Raybould
    The how-different is irrelevant. There are two vector spaces, and you can map them up linearly or non-linearly (pre-adjustment of the sort we see in the cube). Our L2 colors appear to align more or less linearly. There's an implied hypothesis that the mapping would be linear. I get the impression that you somehow think that the very linearity of the L1 to L2 mappings critically depends on what the basis vectors in the L2 space represent.
    The point here is that the assumption, that we would both agree about the redness of all (or most) things, is predicated on assumptions about our experiences being similar.A Raybould
    Wrong. For us to agree about the redness of all (or most) things, the only thing that is necessary is that we form the same categories of L2 colors that vary in the same way to L1 color spaces; it is entirely unnecessary that the L2 colors themselves be the same. The presumption that our experiences are similar is not yet warranted. Given Jane is a tetrachromat, we could have j-red, j-green, j-blue, and j-c4. Your "red" could be j-c4; my "blue" could be j-c4 (incidentally, these are just simplified illustrative mappings). Also, obviously, we don't actually agree that our L2 colors are similar (your phrasing, "our experiences being similar")... otherwise, philosophers wouldn't brandish about terms like "inverted spectrum".
  • A Raybould
    86

    There's an implied hypothesis that the mapping would be linear.InPitzotl

    That's my point about your argument - you are making assumptions - assumptions that say "our experiences have these similarities".

    But in the end you still wind up at my position... that we simply need more study.InPitzotl

    Equally for your assumptions. You are being inconsistent. You are also "straw-manning" me in suggesting that I think otherwise.

    ETA:
    Also, obviously, we don't actually agree that our L2 colors are similar... otherwise, philosophers wouldn't brandish about terms like "inverted spectrum".InPitzotl

    Please explain how the conclusion follows from the premise.


    I cannot imagine why you think I am confusing L1 and L2, but there is nothing to be gained by following that any further.
    — A Raybould
    It's because you keep talking about behavioral responses and disagreements on whether all people would agree that particular things are red if they simply have different L2 colors but share L1 colors.
    InPitzotl

    While you are about it, please explain how that follows, also.
  • InPitzotl
    310
    That's my point about your argument - you are making assumptions - assumptions that are more consistent with experiences being similar than differentA Raybould
    Try to pretend for a second that you understand math. Abstract the nature of experience out and let's talk about pure vector spaces. One example vector space would have a certain amount of salt mixed with a fluid; a certain amount of glucose, and a certain amount of alcohol (note that we're not concerned with infinite vector spaces, since color space is restricted to a range, so this works perfectly well). We could encode something like CIELAB (x,y,z) as distinct concentrations of salt, glucose, and alcohol mixed in the fluid. We could also swap out glucose with maltodextrin. (NOTE: There's no proposal here that experiences are solutions... I'm just showing you just how wild and arbitrary you can get when creating a vector space; outside of this parenthetical, precisely because it's arbitrarily wild, I'll come back to solution-theory of experience as a proxy). As another example, we might encode color as an oscillating function of some value varying in time, like a sound wave... x's basis could be assigned to 10kHz oscillation, y to 8kHz, and z to 7kHz, such that we have a color encoding of f(t/2pi) = sin(10000t)x + sin(8000t)y + sin(7000t)z. We might consider swapping x's basis from a 10kHz frequency to a 12kHz one (along these lines, we can also choose phase encodings or any of a number of things). I could invent vector spaces here all day long.

    You've invented some sort of abstract vector space where linear correlations are no longer linear correlations, such that something would be "more consistent with [vector bases] being similar than different", but that is incoherent gibberish to me. The linearity of our mapping to saline-glucose-alcohol solution has absolutely nothing to do with the fact that we used glucose here as opposed to maltodextrin as the basis vector. The linearity of our sound encoding has absolutely nothing to do with the fact that we assigned x's encoding to 10kHz as opposed to 12kHz.

    So what pray tell are you even talking about?
    Equally for your assumptions. You are being inconsistent.A Raybould
    I'm more than willing to grant that variability on experience (among people) needs study; the point of going over these examples and objections isn't to present an opinion, but rather to present reasons not to form a particular opposing one until we get more information.

    But I don't think your objections about the linearity of a vector space depending on what the basis vector represents are even coherent. It appears to me that you're trying to support a conclusion that if everyone agrees this is orange, their solutions representing the color must all use glucose, because we would have to make an assumption that maltodextrin solutions can also represent consensus of orange before concluding that it could possibly be maltodextrin. That is the very kind of thing I'm arguing against... garbage arguments for an opinion (in this case same-experience), which push comes to shove are double standards of burden of proof (too low justification required to conclude same-experience but more than what should reasonable required to reject it). You should not in this case claim that you have adequately justified same-experience until you have measured the glucose in the solutions of various people.
  • A Raybould
    86

    Ha! this is just a long-winded way of attempting to hide that you are failing to refute what I have been saying: people generally agreeing on categories implies a degree of commonality in the mapping of sensory input to experience. We also have examples of second-order variance from this commonality, as demonstrated by various forms of color-blindness, which need not necessarily arise at L1.

    At this point you are reduced to attempting to pin straw-man opinions on me, with some motte-and-baileying thrown in.

    By the way, you may have missed some additional points that I added to my previous post after first submitting it.
  • InPitzotl
    310
    that people to generally agree on categories implies a degree of commonality in the mapping of sensory input to experienceA Raybould
    There's a difference between a mapping being equivalent and the thing being mapped to being the same. This is how you originally engaged me:
    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.A Raybould
    What arguments are there for the proposition that everyone experiences things fundamentally differently?A Raybould
    I read "in detail" and "fundamentally differently" as describing that what's being mapped to being more or less the same. Additionally, you posted the first thing in response to this quote:
    but to simply conclude that the color experiences are the same because we're all human sounds to me more like guesswork.InPitzotl
    So I'm not quite sure "straw man" is the right term for it... it's pretty easy from this reading to get the interpretation that you're claiming that if the mapping is the same then the things being mapped to should be the same. But let's just clear that up right now.

    Are you arguing that if the functional mappings are equivalent (say, in the sense that if the behaviors of ranking things into categories are the same), that this implies that the experiences are the same?
    ETA:
    We also have examples of second-order variance from this commonality, as demonstrated by various forms of color-blindness, which need not necessarily arise at L1.A Raybould
    I have no idea what you mean by "not necessarily" (are you envisioning some universe where established science is wrong?), but it's already established that protanopia/protanomalies are associated with L cones, deuteranopia/deuteranomalies are associated with M cones, and tritanopia/tritanomalies are associated with S cones. We know that the former two are common in men due to the fact that the genes for L and M cones are on the X chromosome and absent on the Y chromosome, so females get an extra copy of them (note that S cone encoding is not on this chromosome). For similar reasons it's females who tend to be tetrachromats because the rarer OPN1MW2 gene could be present on one of these X chromosomes, and OPN1MW1 on the other, presenting as two different encodings of "the M" chromosome. These genotypes encode for differences on the opsins, and differences of the opsin proteins modulate the sensitivities to particular frequencies of light that these proteins react to when they photoisomerize (fold based on absorbing a protein), which is the primary biophysical mechanics triggering color vision. (Note: There are also L cone alleles). The effect is that these two variants of M chromosomes in effect become two distinct photoreceptors. So according to these established modes of color blindness (and tetrachromacy), they are in fact at the L1 level.

    Also, obviously, we don't actually agree that our L2 colors are similar... otherwise, philosophers wouldn't brandish about terms like "inverted spectrum". — InPitzotl
    Please explain how the conclusion follows from the premise.
    A Raybould
    It's almost a direct translation. L2 colors are our experiences. An inverted spectrum philosophically is by definition an inversion of the experience of colors, ergo, it would be an inversion of the L1 color space mapping to the L2 colors.
    It's because you keep talking about behavioral responses and disagreements on whether all people would agree that particular things are red if they simply have different L2 colors but share L1 colors. — InPitzotl
    While you are about it, please explain how that follows, also.
    A Raybould
    If there's an inverted spectrum, the mappings from L1 to L2 colors would be inverted, but the continuity of said mappings need not be affected. The only behavioral argument on the table so far is: "evolution is very conservative about things that are important to fitness, and our minds". That reasoning presumes there's a fitness advantage, but you would need to explain an actual advantage to make this argument solid (again, we don't all have green eyes by this argument).
  • A Raybould
    86
     

    This is how you originally engaged me:
    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.
    — A Raybould
    What arguments are there for the proposition that everyone experiences things fundamentally differently?
    — A Raybould
    InPitzotl

    And in quoting me, you left off the very next line:

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

    Later on, I wrote:

    Do you see what I'm getting at?
    — InPitzotl

    I think I do, and I think we are talking at cross-purposes. You are saying that these differences exist, and I am saying that they are second-order effects, modifying an underlying commonality. We could both be right! (Or wrong.) Alternatively, we could just agree to disagree in our opinions and wait for neuroscience to achieve a more fine-grained picture than fMRI and related technologies has delivered so far.
    A Raybould

    so it is not as if I am imputing straw-man arguments to you.


    Also, obviously, we don't actually agree that our L2 colors are similar... otherwise, philosophers wouldn't brandish about terms like "inverted spectrum". — InPitzotl

    Please explain how the conclusion follows from the premise.
    — A Raybould

    It's almost a direct translation. L2 colors are our experiences. An inverted spectrum philosophically is by definition an inversion of the experience of colors, ergo, it would be an inversion of the L1 color space mapping to the L2 colors.
    InPitzotl

    The only invocations of inverted qualia that I am aware of are in modal metaphysical arguments against functionalism. I may be mistaken, but I do not think that even a majority of the philosophers who invoke these arguments have much commitment to the proposition that such cases occur in the actual world, and the philosophers who do seem to be a minority of all philosophers.

    More generally - and I think it covers all the points you have raised here - you are saying that there are cases where disparate experiences would have no observable behavioral effects, and I am saying that nevertheless, there are other cases in which they would (plus that I am of the opinion that commonality of experience, at a functional level and as a first-order effect, is a plausible hypothesis, for reasons that I will get into below.) Inverted spectrum scenarios are not the only way in which two people might differ in their experiences, and not all of those alternatives would result in  them agreeing over the categorization of colored objects.If you can invoke inverted spectra, surely I can invoke these others!

    Of course, if you summarily rule out all scenarios, in which the two parties would not agree on color categories, as "incoherent gibberish", then of course they could differ over the full range of remaining possibilities and never disagree on the categorization of color!


    Are you arguing that if the functional mappings are equivalent (say, in the sense that if the behaviors of ranking things into categories are the same), that this implies that the experiences are the same?InPitzotl

    This is an issue you raised, not me:
    If I experience red a different way than you experience red, wouldn't we still both call red things red?InPitzotl

    And I replied with the response that I have not yet seen any reason to change, "not necessarily."  I'm not making a point here, I am merely answering your question. I assumed your question was the prelude to an argument on your part, but I do not believe we have seen the follow-on yet. And if you object to "not necessarily", are you not arguing for "necessarily?" What else is there?

    As to what my position is:
    It is my impression that the idea, that we cannot and could not tell if our experiences are the same, is the majority view, and it is one I used to hold, but I have come to think of it more likely than not that they are, up to the level of first-order effects. My reasons for suspecting this to be so are these:

    Empirically, neurophysical studies using multiple subjects work - they produce quite specific results that can reasonably be taken as general, applying to the majority of the population. In study after study, some aspect of mental functionality is found to be expressed in the same small region of all the subjects' brains, so the evidence is that our brains are quite similar up to an architectural level, and at least to some functional level.

    It would be a fair point to say that these studies have only gone so far in figuring out how brains work, though this strikes me as being like a 'God in the gaps' argument. It is at least plausible that this commonality will continue to hold as we advance our understanding of how brains make minds.

    A dualist might say that it is possible for two people to be identical in every possible physical way, and still have different experiences (that's what inverted spectra arguments claim), but I am not a dualist and I doubt this is possible; I reckon that if we could acheive a complete, causal, physical model of the brain, we would find that's all there is, and that there isn't anything beyond that for experiences to differ over.  
     
    You have attempted to dismiss this as my attempt to impute attitudes to researchers, but it is actually based on the empirical fact that these multiple-subject studies work, and that it is uncontroversial to generalize the results from the studies' subjects to the general population.

    Secondly, the human brain is a network of neurons trained over hundreds of millions of years, and at each generation, the information accumulated by that training is squeezed into DNA and reconstituted. It is certainly possible that, while externally our minds function similarly in many ways, under the hood, each brain is working quite differently than any other. A more parsimonious expectation, however, is that this multi-generational training has produced brains that function alike, to a first approximation. If they did not, how likely is it that they would have, so to speak, 'learned the lessons' imparted by evolution? If a child's high-level brain function can vary markedly from that of its parents, how likely is it that it would nevertheless still be behaviourally similar enough that the child has approximately the same level of that part of its fitness that comes from its mental abilities? Alternatively, what sort of mechanism would be needed to conserve the external behaviour in the face of internal variation? We cannot depend on evolution doing that, as evolution is itself dependent on the conservation of fitness traits from parent to offspring.

    As for your objection that we don't all have green eyes, here is where the point about evolutionary conservatism (selection results in tight constraints on variation in matters that are strong determinants of fitness) comes in: we can have various colors of our irises because the particular hue of their pigment does not strongly determine fitness and so is not strongly selected for. I do not think it is very speculative to say that our mental abilities are strong determinants of fitness (unless, of course, one thinks our experiences are epiphenomenal.)

    To be fair, I have not made these arguments in this detail before, though they have all appeared in abbreviated form. One advantage of this discussion is that I have given the issue more thought.

    So this is all speculative, but I wrote this earlier, and you said at the time that you accepted it:
    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.A Raybould

    Furthermore, what are the counter-arguments, other than that it is speculative, which isn't a fatal flaw in a mere hypothesis? You appear to have started into an argument based on us agreeing on color categories, but you do not appear to have finished it yet. Earlier, you wrote:
    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.InPitzotl

    OK, like my views, it is an opinion, but to me this looks like 'guilt by association': it seems, to you, like other arguments, ones that suffer from "hasty generalization" and "personal incredulity" - but how are these flaws manifest in this particular argument? What do you think I am being incredulous of?
  • InPitzotl
    310
    The only invocations of inverted qualia that I am aware of are in modal metaphysical arguments against functionalism.A Raybould
    See section 3: https://plato.stanford.edu/entries/qualia-inverted/
    I may be mistaken, but I do not think that even a majority of the philosophers who invoke these arguments have much commitment to the proposition that such cases occur in the actual world, and the philosophers who do seem to be a minority of all philosophers.
    I'm not so concerned with consensus among philosophers or any group of people for that matter... a million guesses is still not evidence.
    More generally - and I think it covers all the points you have raised here - you are saying that there are cases where disparate experiences would have no observable behavioral effects, and I am saying that nevertheless, there are other cases in which they wouldA Raybould
    I agree on that part, but I think you're confused. You specifically asked me for arguments to support that our color experiences were fundamentally different... I provided them.
    It is my impression that the idea, that we cannot and could not tell if our experiences are the same, is the majority view, and it is one I used to hold, but I have come to think of it more likely than not that they are, up to the level of first-order effects.A Raybould
    Great! This is certain to be a more useful analysis. But going in, I disagree that the premise "we cannot and could not tell if our experiences are the same" is justified.
    Empirically, neurophysical studies using multiple subjects work - they produce quite specific results that ...A Raybould
    ...okay, so I'll call this the architectural argument.
    It would be a fair point to say that these studies have only gone so far in figuring out how brains work,A Raybould
    It's true that we share general architectures; this is the foundation for being able to describe such things as the visual cortex, areas V1 and V4, the ventral and the dorsal stream, and such. But it's still a bit of a mystery how colors are encoded; so far, color analysis seems to be a bit distributed. So yes, we can't quite claim that we've figured out how the brain works. But, yes, this is the most promising area of study, and this is the type of thing to appeal to in such arguments, but there's a bit more work to do. You still can't quite say though the architectures are the same therefore the experiences should be. You have to call apples apples and oranges oranges. You should talk about the candidates for the experiential correlates being the same. Just because our brain shares an architecture doesn't mean it'll share everything.
    A dualist might say that it is possible for two people to be identical in every possible physical way, and still have different experiencesA Raybould
    Sure, but I'm not sure dualists can claim adequate justification.
    You have attempted to dismiss this as my attempt to impute attitudes to researchersA Raybould
    Not exactly... I've pointed out a situation where you're basically guessing researchers share your view and then appealing to the researchers having your view, which is basically an appeal to authority as a fallacy.
    Secondly, the human brain is a network of neurons trained over hundreds of millions of years, and at each generation, the information accumulated by that training is squeezed into DNA and reconstituted. It is certainly possible that, while externally our minds function similarly in many ways, under the hood, each brain is working quite differently than any other. A more parsimonious expectation, however, is that this multi-generational training has produced brains that function alike, to a first approximation.A Raybould
    The description here in terms of evolution sounds a bit Lamarckian. Also, this is quite hand wavy. There's "information" that we share based on evolution. From there, you go to specifying that particular kinds of "information" must be shared, because the alternate theory is that "information" is not shared.
    If they did not, how likely is it that they would have, so to speak, 'learned the lessons' imparted by evolution?A Raybould
    Too vague and hand wavy still. Evolution isn't a teacher teaching individuals lessons. It's just a blind process that happens to do what it does. Natural selection tends to keep the genes around that tend to stay around; sexual reproduction tends to shuffle genes around; genetic mutations of specific types tend to just happen at particular rates. Brains don't just behave... they learn; ANN's aren't quite the same thing as a brain, but they're success stories (to the degree that they are) in mimicking how brains learn at some level. Behaviors important to survival are significant, but brains come with hormone systems, body sensations, drives, and instincts for such things. Also, it sounds pretty expensive to code entire brains into a genome, in terms of genomic cost, and evolution is known for being "good enough" (it's why we tend to die of old age... we simply don't need to live that long to pass our genes into the gene pool so the selection pressure dips below what's needed to keep up with mutation rates).
    If a child's high-level brain function can vary markedly from that of its parents, how likely is it that it would nevertheless still be behaviourally similar enough that the child has approximately the same level of that part of its fitness that comes from its mental abilities?
    We're not talking about high level brain functions here. We're talking about the experiential correlate of a color.
    Alternatively, what sort of mechanism would be needed to conserve the external behaviour in the face of internal variation?A Raybould
    Any isomorphic mechanism would do the trick. Instead of using high and low circuits for 1's and 0's, we could use magnets being in the same or different directions. It doesn't really matter, so long as a change is a change. Instead of encoding y as glucose in the solution we could encode it as maltodextrin.
    We cannot depend on evolution doing that, as evolution is itself dependent on the conservation of fitness traits from parent to offspring.A Raybould
    Evolution is only selecting for fitness where it matters. If a variation does not affect fitness where it matters, evolution would not care about that variation. You're in effect just begging the question; you're assuming that the variations would have an effect on fitness and then arguing that evolution would select those out:
    we can have various colors of our irises because the particular hue of their pigment does not strongly determine fitness and so is not strongly selected for.A Raybould
    So green versus blue eyes don't seem to matter much to fitness.
    I do not think it is very speculative to say that our mental abilities are strong determinants of fitness (unless, of course, one thinks our experiences are epiphenomenal.)A Raybould
    In what way?
    So this is all speculative, but I wrote this earlier, and you said at the time that you accepted it:A Raybould
    ...not quite, but the disagreements are in the weeds (e.g. it's possible to formulate a theory before a hypothesis)... and not quite useful for this discussion.
    Furthermore, what are the counter-arguments, other than that it is speculative, which isn't a fatal flaw in a mere hypothesis?A Raybould
    Counter-arguments fall back to debate mentality. What we're really interested in is the truth. So the analysis to be done on a hypothesis is to explore the ways in which the hypothesis could reasonably fail. That's what I've been doing here.
    but how are these flaws manifest in this particular argument?A Raybould
    It's very simple. The burden aligns with the purpose. Let me start with a summary of what I see you as doing and get back to this.

    You are advancing an argument that evolution theory suggests same-experience... the burden on that is fairly high, and has not been met. To adequately justify this, you need to show how evolution is inconsistent with different-experience. The first argument you advanced failed very quickly with a sanity check; green-eyed people not being universal suggests that evolution doesn't always produce universal traits. What was really missing from this argument was relevancy. You did a little better with the second argument but you're still missing the relevancy... you seem to beg the question by thinking backwards about it, something along the lines of "if different-experiences were had, there would be differences in fitness". I claim this is backwards because you're trying to start at evolutionary fitness and then conclude same-experience, so the real challenge here isn't fitness-affecting differences, it is fitness-invariant differences.

    That's where the arguments I'm providing come into play. The purpose of these arguments is different, and along with this comes a difference in burden. There is no argument on the table that evolution supports the different-experience theory, so there's no need to demonstrate that it does. The argument rather is that evolution is compatible with different-experience theory. If it's simply compatible, then we can reach the intended conclusion, which is not that we have different experiences... but, rather, that you have yet to adequately justify that we have same-experiences. Now were the argument on the table that we actually have different-experiences of color, the burden would shoot up matching the burden I'm expecting of you. But the actual stated position I'm taking is a non-position.
  • jkg20
    382
    It would be really useful if, in your next reply, you could address this issue alone,
    Sorry, I am not clear on exactly what issue it is that you want me to address. The paragraph preceding the one from which the above quotation is extracted makes it look like the point you want to make is that if current science is incomplete, then there should be no philosophical arguments brought forward to eliminate a given avenue for its advancement. Is that the issue you want me to address? Or is it that there should be no such philosophical arguments brought forward which contain as a premise that current science is incomplete?
  • jkg20
    382
    This is a very common and deep-rooted misconception. It simply is not the case that knowing all the facts, about the physical state of one's brain when one has some particular knowledge, is the same as having one's brain in that state, and physicalists can reasonably propose that unless one's brain is in that state, one does not have that knowledge.


    That there is a distinction between, on the one hand, being in a physical state P, and, on the other, knowing all the facts associated with being in a physical state P, does not seem to be a distinction over looked either by Jackson or by me. The issue concerns in what that distinction consists. I can know all the facts about the physiology of sitting in a chair. What knowledge or other epistemic quality do I gain by being in that state for the first time that I did not already have before being in it? Here the reply "nothing" seems to have as much going for it as "something". The intuition on which Jackson's argument and mine rely is that "something" has a stronger pull when the example in question concerns seeing something red for the first time. As I have already admitted, it is just an intuition. However, if you accept the intuition, then something substantive should be said about what it is that is gained, and that is not a burden that falls on the materialist alone. The dualist also has to give us something to work with. For what it is worth, I do not believe either side has a particularly coherent idea about what it is that is gained that would pin it down sufficiently to provide grounds for further investigation.
  • A Raybould
    86
     
    I think we should reset this discussion, and start over. 

    Firstly, let's get rid of the ambiguous phrase "physical knowledge", and replace it with "knowledge of physics", and similarly, replace "physical fact" with "fact about physics", "physical proposition" with "proposition about physics", etc. If we reach a point where this substitution does not express what we want to say, we will find an alternative, non-ambiguous substitution. In particular, if "physical information" comes up, we will use either "information about physics" or "information coded in physical form", according to which we mean.

    Secondly, let's put aside what the skeptic believes. If you like, imagine the skeptic is someone who leans towards dualism but still doubts that your argument succeeds.

    Thirdly, let's use Churchland's 'equivocation' objection as modified by Torin Alter and Tim Crane, which uses the concept of discursively-learnable knowledge. This simply means any knowledge that can be gained through reading, or by attending lectures, seminars, tutorials, etc., or watching or listening remotely - i.e., any knowledge that can be conveyed language backed up by illustrations and demonstrations (though not, of course, in this case, any visual that needs to be in color!). This looks very much like propositional or factual knowledge, but by tightening the argument by using  'discursively-learnable' instead, we can simply avoid getting mired in the question of whether the intellectualists are right in claiming that all knowledge is propositional: even if they were right in some sense, Churchland's rephrased objection is unaffected, as it does not rely on making any distinction between propositional and non-propositional knowledge.

    In your definition (and Jackson's), all knowledge of physics is clearly discursively-learnable (even when one deduces a fact from others, the line of reasoning that led to the deduction could be written down and learned by others from those words, so it, too, would be discursively-learnable.)

    What (if anything) Mary learns from seeing colors is either discursively-learnable or it is not. We can consider these two cases separately.

    If it is not discursively-learnable, then physicalism is not challenged by Mary not learning it until she was released, as she could have learned all the discusively-learnable knowledge that could ever possibly be known, and still learn something more. It is irrelevant whether what she knows before being released is knowledge of physics or of something else.

    Alternatively, if her new knowledge was discursively-learnable, then why could she not learn it from her studies? You cannot simply assert, without justifying the claim, that it was not knowledge of physics, and therefore ruled out by the premise, as that would be begging the question.

    Whatever else you say about your argument, you must also address this issue if you want to claim that it avoids Churchland's charge of equivocation. Therefore, I urge you to confine your reply to anything that is germane to this particular issue, so that we do not get mired in side-issues.
  • A Raybould
    86


    The SEP is an excellent resource, and in section 2 we have a discussion showing that it is not straightforward to find an inverted spectrum scenario that is clearly behaviorally-invariant (and as we will see, my argument is not defeated by the existence of some scenarios that are.)


     "Too vague" seems to have become your default response,  but by itself, it is too... vague? Are there any non-vague metrics of vagueness, and more-or-less objective thresholds? The falsifiability criterion of science provides one (despite not being directed specifically at vagueness alone), and it is an appropriate one here: Vague claims (such as of vagueness exceeding some unspecified threshold) are not falsifiable.

    What I am saying here could be falsified; a (not-yet-achieved) causal model of how the human brain produces a mind would reveal whether, to a first-order approximation, most humans are undergoing the same physical functions when they experience color. That's good enough for a hypothesis.

    For physicalists, where there is no physical difference, there is no difference simpliciter. As I mentioned previously, dualists disagree, but so long as dualism is speculative, it is not enough to rule out my hypothesis: at most, I can restrict it by adding the premise of physicalism, which, despite all efforts, has not been ruled out. And while this is a side-issue, there is nothing more vague than dualists' concepts of the non-physical; there are no hypotheses about how the allegedly non-physical or extra-physical aspects of minds work. As someone who rejects all real and imagined vagueness, you presumably have no truck with dualism.


    There are a number of  misunderstandings in your post:

    More generally - and I think it covers all the points you have raised here - you are saying that there are cases where disparate experiences would have no observable behavioral effects, and I am saying that nevertheless, there are other cases in which they would
    — A Raybould

    I agree on that part, but I think you're confused. You specifically asked me for arguments to support that our color experiences were fundamentally different... I provided them [your ellipsis.]
    InPitzotl

    More specifically, I asked was what objection you had to my reply "not necessarily" to your question "If I experience red a different way than you experience red, wouldn't we still both call red things red?", and your replies only demonstrate possibility where, to be a reply that addresses the evolutionary question, arguments for necessity (or at least predominance) are necessary As far as I can tell, you never did finish the claim you had launched into with that original question.


    You have attempted to dismiss this as my attempt to impute attitudes to researchers
    — A Raybould

    Not exactly... I've pointed out a situation where you're basically guessing researchers share your view and then appealing to the researchers having your view, which is basically an appeal to authority as a fallacy.
    InPitzotl

    ...And it is a claim that you have already been corrected on, and, moreover, right in the part of the quoted sentence you cut out. The full quote is this:

    You have attempted to dismiss this as my attempt to impute attitudes to researchers, but itis actually based on the empirical fact that these multiple-subject studies work, and that it is uncontroversial to generalize the results from the studies' subjects to the general population.A Raybould
     

    - no appeal to authority there.


    The description here in terms of evolution sounds a bit Lamarckian.InPitzotl

    That puzzled me, but I see that I was not being clear here. I am not saying that a child inherits from its parents' adult minds, complete with what they have learned (that obviously does not happen.) I am only using a completely conventional concept of evolution, and it might help if I go through them one by one:

      [1] The brain is predominantly a network of neurons, that has evolved in accordance with current evolutionary theory.
      [2] Consequently, and just like any other organ. its architecture, and therefore its function, is not accidental; it is the end result of billions of natural-selection experiments.
      [3] At each step of this process (i.e. each generation), the initial state of the organ, and the instructions for its development, are encoded into the genotype and then reconstituted from it.
      [4] For evolution to work, there must be some variation between parents and their offspring, but if it is too large and commonplace (beyond second-order), then adaptation will not occur: either the offspring is not viable, or adaptation will be restarting anew with each generation. Empirically, it is not the case that only a few children in each generation have mental abilities (or any other trait) comparable to those of their parents.


    If they did not, how likely is it that they would have, so to speak, 'learned the lessons' imparted by evolution?
    — A Raybould

    Too vague and hand wavy still.
    InPitzotl

    That might seem to be so, to someone unaware of what I wrote about it immediately afterwards. Once more we see the pattern of quoting out of context. 

    As for the paragraph you followed this with, it is all addressed in the above list, which is just a selection of the relevant positions taken in conventional evolutionary theory, with regard to the evolution of any organ, not just the brain. If you like, I could go through how it does so sentence-by-sentence, but I think it would be unnecessary clutter to belabor each point in turn.


    If a child's high-level brain function can vary markedly from that of its parents, how likely is it that it would nevertheless still be behaviourally similar enough that the child has approximately the same level of that part of its fitness that comes from its mental abilities?

    We're not talking about high level brain functions here. We're talking about the experiential correlate of a color.
    InPitzotl

    And what are those correlates?


    Alternatively, what sort of mechanism would be needed to conserve the external behaviour in the face of internal variation?
    — A Raybould

    Any isomorphic mechanism would do the trick. Instead of using high and low circuits for 1's and 0's, we could use magnets being in the same or different directions. It doesn't really matter, so long as a change is a change. Instead of encoding y as glucose in the solution we could encode it as maltodextrin.
    InPitzotl

    I think you are missing the point here. As we have seen, it is not the case that no matter how a child's "experiential correlates" of sensory input differ from those of her parents, it could not have any behavioral consequences. Given the empirical evidence that children and their parents do, to a first approximation, behave similarly with respect to their sensory stimulation (generally agreeing on the categorization of colors, for example (your example, as it happens)), then it seems that, whatever internal correlates they have to those stimuli, they are generally from the subset of possibilities that preserve this commonality of behavioral responses. The question then is, if childrens' internal correlates of stimuli are free to differ markedly from those of their parents, what genetic and ontological mechanism might there be to restrict the variation in the experiential correlates such that the behavioral commonality is preserved?


    We cannot depend on evolution doing that, as evolution is itself dependent on the conservation of fitness traits from parent to offspring.
    — A Raybould

    Evolution is only selecting for fitness where it matters. If a variation does not affect fitness where it matters, evolution would not care about that variation. You're in effect just begging the question; you're assuming that the variations would have an effect on fitness and then arguing that evolution would select those out:
    InPitzotl

    Not at all - you are arguing against your misunderstanding of my argument. My point is based on empirical fact: if variation is routinely producing children that have markedly different functional responses to color stimuli than their parents, then how come we only very rarely see those variations that do not result in observable differences?


    we can have various colors of our irises because the particular hue of their pigment does not strongly determine fitness and so is not strongly selected for.
    — A Raybould

    So green versus blue eyes don't seem to matter much to fitness.
    InPitzotl

    So we agree on something. Is it too much to think that we also agree that this lack of fitness is because the primary function of the iris' pigment is to block the transmission of light, and so what it reflects is not consequential?


    I do not think it is very speculative to say that our mental abilities are strong determinants of fitness (unless, of course, one thinks our experiences are epiphenomenal.)
    — A Raybould

    In what way?
    InPitzotl

    The alternative is so lacking in plausibility that I will not bother to reply until specific arguments for it are presented. Extraordinary claims require extraordinary evidence.


    So this is all speculative, but I wrote this earlier, and you said at the time that you accepted it:
    — A Raybould

    ...not quite, but the disagreements are in the weeds (e.g. it's possible to formulate a theory before a hypothesis)... and not quite useful for this discussion.
    InPitzotl

    (By the way, I'm curious about what was, at some point, a theory but not a hypothesis.)

    For reference, let's see what you are still not disagreeing with in any material way: "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." I will have more to say on that below.


    Furthermore, what are the counter-arguments, other than that it is speculative, which isn't a fatal flaw in a mere hypothesis?
    — A Raybould

    Counter-arguments fall back to debate mentality. What we're really interested in is the truth...
    InPitzotl

    Argument and counter-argument are the principal methods of philosophy, and are important in science; these are both generally regarded as attempts to discern the truth. What is not acceptable are various ploys often seen in other arenas of debate, such as the repeating of a claim after it has been refuted, as if it had not been, or quoting out of context.


    ...So the analysis to be done on a hypothesis is to explore the ways in which the hypothesis could reasonably fail. That's what I've been doing here.InPitzotl

    To the extent you have been on-point in that respect you have been helpful, and I thank you for that - but your work here is not complete.


    but how are these flaws ["hasty generalization" and "personal incredulity"] manifest in this particular argument?— A Raybould

    It's very simple...
    InPitzotl

    The last time you said that, you shortly afterwards backed away from the claims that followed... What you wrote here did not actually throw any light on what I am being incredulous of, which is fine if you are no longer making this claim.


    To adequately justify this, you need to show how evolution is inconsistent with different-experience.InPitzotl

    As you have misunderstood the evolutionary argument, your judgement of its efficacy is suspect.

    ...The first argument you advanced failed very quickly with a sanity check; green-eyed people not being universal suggests that evolution doesn't always produce universal traits.InPitzotl

    As we have seen above, you did not follow that argument, and, as a consequence, replied with a non-sequitur.


    you seem to beg the question by thinking backwards about it, something along the lines of "if different-experiences were had, there would be differences in fitness". I claim this is backwards because you're trying to start at evolutionary fitness and then conclude same-experienceInPitzotl

     As we have seen above, you did not follow my arguments, and, as a consequence, you are arguing against your own mistaken attempt to paraphrase my position. Furthermore, any scenario, in which the different-experience hypothesis has a difficulty explaining observations and the same-experience hypothesis does not, is an argument (or evidence, if it is an empirical fact) for the latter over the former.


    The argument rather is that evolution is compatible with different-experience theory.InPitzotl

    I don't recall that argument being made - could you point me to where it was?


    But the actual stated position I'm taking is a non-position [my emphasis.]InPitzotl

    While your stated position, being indefeasible (though trivially so) is a strong one to sustain during a debate, its usefulness in the search for knowledge is wholly dependent on other people looking for answers. 

    As soon as a thesis is mooted, both it and its antithesis exist as hypotheses, regardless of your personal attitude to them. While your stated goal, apparently, is to show that the notion that our experiences are similar to a first approximation does not achieve hypothesis-hood, the closest option you have to that is to summarily dismiss it. There are several ways you could achieve that: You could show it to be incoherent or internally inconsistent, or you could show it to be ruled out by established facts. You could show that it is unfalsifiable. Failing any of these, you could, if you had an alternative view, argue that it is so much more plausible, than the alternative, then the latter is barely worth considering - but, absent that differential plausibility, implausible hypotheses cannot so easily be dismissed (plenty of previously implausible ideas have become theories.) By simply presenting ways in which new knowledge could defeat it, you are already treating it as a hypothesis, and will continue to do so up until the point where you achieve one of the above.


    BTW, in reply to an earlier ETA:
    So according to these established modes of color blindness (and tetrachromacy), they are in fact at the L1 level.InPitzotl

    I am guessing that your emphasis on 'established' indicates that you were aware of this, but I had in mind cerebral and congenital achromatopsia and dyschromatopsia, color agnosia, and any related pathologies.
  • InPitzotl
    310
    The SEP is an excellent resource, and in section 2 we have a discussion showing that it is not straightforward to find an inverted spectrum scenario that is clearly behaviorally-invariant (and as we will see, my argument is not defeated by the existence of some scenarios that are.)A Raybould
    You're confusing your opinion with your argument. The stuff in section 2 is a different argument than what you've presented. Regarding that, some of the arguments in that section are in fact decent and relevant, but they do also presume things about color processing for which we really need more detail. For example, take this from SEP:
    As noted in the previous subsection, there are more perceptually distinguishable shades between red and blue than there are between green and yellow, which would make red-green inversion behaviorally detectable. And there are yet further asymmetries.Inverted qualia
    ...which does seem compelling (more generally, qualitative "inter-qualia" comparisons yield subjective asymmetries like this), but there are potential L1 reasons for this (via color opponency):

    Red/green and blue/yellow are opponent color processes involving respectively some L-M and L+M-S function, with brightness being roughly an L+M+S function so close to just M that it's modeled that way in some color spaces. So for real colors, variance from green to yellow has L-M going from balanced to low without changing L+M-S balance much, whereas variance from red to blue has the same L-M variance going from high to balanced while also the L+M-S process goes from high to low. So in this case, the variation of real colors from yellow to green is expressed in the variation of just one color channel going from neutral to green, but the variation of real colors from red to blue is expressed in the variation of two color channels transitioning inversely from yellow+red to blue+neutral. Similarly with yellow versus brown being quite different versus blue and dark blue, yellow has an incredibly high brightness whereas blue starts out at fairly low brightness (relative luminance). Since these are L1 like properties, inverting the mappings from L1 to L2 colors may still carry these properties; IOW, an inverted spectrum might not quite be as much like flipping an RGB image's colors as one might presume.
    "Too vague" seems to have become your default response, but by itself, it is too... vague?A Raybould
    It's not exactly a default response so much as it is prompted:
    A more parsimonious expectation, however, is that this multi-generational training has produced brains that function alike, to a first approximation. If they did not, how likely is it that they would have, so to speak, 'learned the lessons' imparted by evolution?A Raybould
    ...in relation to the topic at hand, this is indeed too vague. Your expectation is more parsimonious than what exactly? Function alike in what ways? What "learned lessons"? I'm perfectly happy to say that human brains evolved in human like ways, but that does not really imply same-experience unless you can connect the similarity of human evolution to the similarity of color experience, which I've yet to see. Other than that, it's yet another nature versus nurture debate. Truth is, both nature and nurture make brains, especially human brains.
    My point is based on empirical fact: if variation is routinely producing children that have markedly different functional responses to color stimuli than their parents, then how come we only very rarely see those variations that do not result in observable differences?A Raybould
    Because, for example, predators aren't examining your brain with fMRI to see if you represent redness on this spot or that spot or using this average frequency of pulses or that frequency? The important thing from a fitness perspective is that you run away, hide, or fight the predator appropriately.
    The alternative is so lacking in plausibility that I will not bother to reply until specific arguments for it are presented.A Raybould
    That's your choice, but in effect, given that you're the one claiming to be supporting a proposal that the experiential correlates have a fitness advantage, not answering the question absolutely equates to not addressing the very thing your proposal is supposed to be about. If your proposal is about a tie between fitness and particular experiential correlates of color, it is backed if and only if you can demonstrate what it is about... i.e., tie fitness to experiential correlates of color. This is what I mean by relevance.
    For physicalists, where there is no physical difference, there is no difference simpliciter.A Raybould
    Sure. So let's focus on the correlates, since that's where the difference would be.
    And what are those correlates?A Raybould
    I don't know, but I think that's the key question. In terms of L1 colors, there are reasonable explanations of development that are works in progress but involve self organization; these generally produce opponent color processes. Example:
    https://www.frontiersin.org/articles/10.3389/fncir.2014.00016/full
    This itself requires a bit more study. In terms of color perception in the brain, as I said before, we know there are multiple places of analysis. But there's also the possibility of potential signal space (such as these toy models of potential Poisson rates)... and since the signals potentially start this way there may be transference from these signal rates to positions, or not. But "qualia" may be more complex than many philosophers tend to treat it as well; e.g., there's such a thing as pain that doesn't hurt, suggesting that qualia aren't necessarily singular in the first place. This might be explained as pain in practice actually having separable pieces that typically are co-associated but not essentially so, and the same could possibly be true of color. What we're looking at developmentally for different-experience would be any sort of theory that starts with this brain in a semi-random developmental state and just "settles into" the "nearest corners". The entire question here (that has not yet actually been addressed) is how many "corners" are there, why are there that many, and how do the L1 colors map into them? Assuming opponency-sized "corners", a same-experience theory would postulate four to six of them, and some sort of specific developmental pathway whereby particular opponent signals latch into the specific four (or six) that they're supposed to map to.
    What I am saying here could be falsifiedA Raybould
    ...I'm fine with that, but "adequately justified" is more akin to what you're saying being verified.
    Argument and counter-argument are the principal methods of philosophyA Raybould
    Yes, but counterarguments should not have to require an opposition taking a side to provide the counterarguments, and most certainly should not require the opposition to hold the countering view as an opinion. Also, two people discussing a thing need not necessarily each pick a corner and box; it's entirely possible, and may even be more productive, for the two to simply walk hand in hand from corner to corner.
    Furthermore, any scenario, in which the different-experience hypothesis has a difficulty explaining observations and the same-experience hypothesis does not, is an argument (or evidence, if it is an empirical fact) for the latter over the former.A Raybould
    So, let's get back to tetrachromacy. Let's suppose we introduce a new gene in the human gene pool, call it OPN1MW3. OPN1MW3 expresses in people who have it by producing an M cone with spectral sensitivity shifted towards blue by the same amount (measured in frequency) that M shifts L spectral sensitivity towards blue; let's call this a N cone. This gene is an allele for the M on the X chromosome. So suppose we have: (a) Adam, who has L, M, and S cones; (b) Bill, who has L, N, and S cones; (c) Cindy, who has L, M, N, and S cones. So here are some questions. (a1) Is Adam likely to be a trichromat? (b1) Is Bill? (c1) Is Cindy more likely to be a tetrachromat or a trichromat?

    (c1) is the interesting question... but regardless of its answer we still have followups. If the developmental process is such that Cindy's likely to become a tetrachromat, then there are questions about what exactly causes the L2 colors that Bill sees; your hypothesis suggests it's some built-in gene, but if self organization suffices to establish L2 colors this is questionable. But if Cindy is only a tetrachromat if she has some other specific gene, some BN (brain-gene-N analog; here, this is just a generic referent... it could be more than one gene), that pre-structures her brain, then we must also ask whether Bill can have that gene as well, or under what conditions precisely Bill sees what Cindy sees that Adam doesn't see.

    Your hypothesis seems to require a "BM" gene of sorts. Okay, we have the human genome mapped out... so which gene is BM?
    While your stated position, being indefeasible (though trivially so) is a strong one to sustain during a debate, its usefulness in the search for knowledge is wholly dependent on other people looking for answers.A Raybould
    The point isn't to simply maintain some position with unreasonable standards though. The point is to require relevance. The thing being talked about here is the actual stuff happening between our ears in our soft pink squishy warm brains, that has to do with our subjective conscious experience of colors. Some discussion of and/or constraints on how that subjective experience's correlates develop is necessary to provide a theory of how much the subjective experience's correlates can vary. Without having that discussion or addressing what those constraints are, you're just plain not having the required conversation.
    While your stated goal, apparently, is to show that the notion that our experiences are similar to a first approximation does not achieve hypothesis-hoodA Raybould
    I've no objection to same-experience as a hypothesis. My objection is claiming that the hypothesis is adequately justified prematurely.
    I am guessing that your emphasis on 'established' indicates that you were aware of this, but I had in mind cerebral and congenital achromatopsia and dyschromatopsia, color agnosiaA Raybould
    Achromatopsia and dyschromatopsia are the same modes of L1 level color deficiencies previously discussed (though there are acquired forms). Color agnosia as far as I'm aware is a defect of the ventral stream, which is particularly interesting for awareness of L2 colors at all (if not L2 modes of color at all). I would be interested in an L2 specific defect.
  • jkg20
    382
    What (if anything) Mary learns from seeing colors is either discursively-learnable or it is not. We can consider these two cases separately.
    OK, so this argument puts off the question of accepting whether or not Mary learns anything.
    I guess against this argument, the skeptic would have to push the investigation of the supposed distinction between knowledge being discursively learnable and knowledge being physically encodeable. What does "physically encodeable" mean? Given the representationalist background of most versions of physicalism these days, I suppose one proposal for a gloss of "physically encodeable knowledge" would be knowledge the content of which is entirely representational in or through a physical form. So, a premise of the Churchland/Crane/whoever argument is that there is a distinction between knowledge that is discursively learnable and knowledge, the content of which is entirely encodeable in a physical medium. So what is content? Content is propositional. Even if one accepts the idea of non conceptual content, that is not to accept the idea that there is content that cannot be expressed in terms of propositions, it is just to accept that the person having the mental state with that content need not have the resources to state those propositions. So if all content is propositional, then not only is all discursively learnable knowledge physically encodeable, all physically encodeable knowledge is expressable as propositions and so is discursively learnable. Thus, perhaps, one way in which the skeptic might try to deal with the Crane/Churchland/whoever argument by collapsing the distinction required by one of its premises.
  • jkg20
    382
    Or to put things a little differently, and perhaps more clearly, the Crane/Churchland line seems to force the skeptic to a position in which they will have to defend the following modus tollens
    1 If some item of knowledge is physically encodeable, then it is discursively learnable
    2 The item of knowledge Mary gains is not discursively learnable
    _________________________
    The item of knowledge Mary gains is not physically encodeable

    The support for premise 1 would of course have to be independent of any issues concerning Mary, and I gave an indication of one way that might be done in the previous post.
bold
italic
underline
strike
code
quote
ulist
image
url
mention
reveal
youtube
tweet
Add a Comment

Welcome to The Philosophy Forum!

Get involved in philosophical discussions about knowledge, truth, language, consciousness, science, politics, religion, logic and mathematics, art, history, and lots more. No ads, no clutter, and very little agreement — just fascinating conversations.