Cartesian Externalism

I never saw anything contradictory with the idea that we might be subjects trapped in Matrix type scenario - a brains in vats scenario. Really, given the developments of things like computer generated virtual realities, in which we immerse ourself through gaming, it is easy to imagine the possibility of the simulations being so good that they are indistinguishable from our experiences in real life. And I can’t see anything contradictory with the idea that my body when I was a baby was hooked up to some virtual reality.

Now, there are arguments like Putnam’s one against brain in vats, which are related to certain theory of meaning of the words, but the possibility of virtual reality is so clear, seems to me, that denying the possibility that we are brains in vats on base of that argument, seems to me on par with denying that there is movement based on Zeno’s argument. Certainly there is value in Zeno’s paradoxes, but the value is not in proving that motion is impossible.

Anyway, not just that I don’t see anything contradictory in me being brain in a vat, which is placed within a virtual reality controlled by machines, but I don’t see anything contradictory with the idea that my brain was put in a vat, and that what it was and is getting from the inputs are signals generated by a random process, and that only by mere chance those inputs ended up such that I’m under the illusion that I’m a subject with a life I have, with wife, with kids, with all those experiences.

I think that even this turns out to be true, and every individual thing to which I was acquainted in my life turns out to have been an illusion, I will still have idea of wives, bachelors, chairs, books, philosophy, vats, brains, language, and so on… And, if by mere chance, it also turns out that these illusions were fully inline with what is there really in the world, my intended meanings in the virtual reality, will be fully inline with the meanings in this real world. I will be able to express my previous thoughts (the same thoughts I already had) about bachelors, about books, about brains, language, and so on…

What is interesting to me is how to relate my thinking that those kinds of scenario are possible with some of my externalist inclinations.

As I said, I believe that there is no such thing as ‘phenomenal experience’, and that ‘experience’ properly (and traditionally) refers to the events in the world in which we participate, and by which we are affected or from which we learn, OR (in alternative sense) it refers to the knowledge gained in that way.

Further, related to this, I don’t think there are concepts, if by concepts we mean some constituents of our thinking which would be some things in our heads. As said, I think that words like ‘bachelors’, ‘chairs’, ‘books’ and so on, refer to multitudes of things which are part of certain (and real at that) phenomenon in the world - a phenomenon of which we are aware. (I don’t think that concepts are Platonic ideas neither.)

I would also take externalist position on words meanings also, as I think they only have meaning in the context of language as part of the practices in the society, so again, would take externalist stance on this also.

So, I guess there is some kind of tension between those views. A very interesting dialectic here.

Hidden Person

Here is an interesting visual illusion (via Spluch) …

People at Spluch say that you should move away from the screen to see the face better, but you can also try looking at the image while shaking the head left/right or squint the eyes. Of course you can also move away from the screen, squint the eyes, and shake your head left/right. (You can try few dancing moves while you are at it).

I guess this is a good example of how rod cells pick up more information than cone cells can?

Different, But Indistinguishable Experiences

Richard over at Philosophy Sucks!, posted a PowerPoint presentation of his paper Consciousness, Higher-Order Thoughts and What It’s Like. The great thing is that he has integrated his reading of the paper, so that the whole experience is much better (at least it was for me) than reading a paper.

Anyway in discussion of that post Richard gave few objections to the view of perception that I’m playing with on this blog in few previous posts (for example the posts on illusions, “appears” vs. “is”,  hallucinations and dreams, and also few follow up notes).

I don’t want to clutter comments on his blog with those issues, so I will quote Richard’s objections here, and try to answer them.
Richard says…

I just don’t see how this kind of view doesn’t fly in the face of all of brain science…for instance you say that you can have a pain in your finger and yet not feel it, but we know that there are no pains in fingers, pain is in the brain (as evidenced by phantom limb pain) so to have a pain in the finger is to be in a mental state that represents the pain as being somewhere. How do you avoid this conclusion?

I don’t think that it is problematic to say that the pain we feel is in the finger just because there is possibility of illusion. We don’t say that the rabbit we see is in the brain, just because there is a possibility of illusion. Both cases seem analogous to me, and as one can accept that the objects we see are not in the brain, one can accept that the pain we feel is also not in the brain. I had more detailed analysis of this, in the post about cyborgs sharing the pain.

The other objection Richard gives is the following…

Also there is the obvious problem of dreams and hallucinations. I know you have addressed this issue, arguing that it is the imagination that has something to do with it, but this answer is no good because we know from brain imaging studies that when you dream about things the actual visual cortex is active but this is not the case for imagining. We also know that we can stimulate the visual cortex directly and generate visual experiences in the absence of objects, so how can objects themselves be the constituents of experiences if we can have experiences in the absence of objects?

I want to thank Richard for pointing to the issue with connecting imagination and dreams. Adding imagination there was wild speculation on my part (as it turns out - wrong), but I don’t think it was essential to the argument. The general idea was that some kind of *seeing affector* can be used to explain the possibility of hallucinations and dreams, and still leave open the possibility that when we see real things it is the things themselves that are constituents of the experience.

The idea is that the word “experience”  should be read in an externalist manner, and not as something private and internal to the subject. I don’t think this is problematic, and that even aligns better with the everyday usage of the word “experience”. I guess I will argue this in separate post. So, “experience” being read in this way, we can say that while two experiences are different (say seeing a box and pyramid from certain side), they might be indistinguishable by the subject. Same can be said about two different experiences, e.g. seeing a rabbit and hallucinating one - that those are different experiences, but they might be indistinguishable by the subject. In this way, we can say that the possibility of illusions doesn’t implicate that the objects themselves can’t be constituents of the experience. The same reading can be used for the cases of stimulation of visual cortex.

Of course lot of things in this view depends on issue of how to read “experience”, and what this thing philosophers call “experience” is supposed to be. That’s why I asked few posts ago for some clarification of what representational theories mean by that word. (The only ‘official’ answer I got was Pete Mandik’s - who said that experience is supposed to be a theoretical concept, and not something of which we are directly aware).

Few Notes on my Position on Perception, Ending with a Rant

I will try shortly to explain further my position on perception that I wrote about in few previous posts, and how it stands in relation to some other positions…

First to start negatively - this position isn’t representationalism - it doesn’t say that experience represents the world as being somehow, nor for that matter that there is veridical and falsidical experiences which would depend on the issue if the experience represents the world as it is, or not. Even less it is qualia (or sense-data) view. It is negated that any such thing as “phenomenal seeming”,  in any sense in which it might remind of Cartesian theater.

Instead in this view the objects in the world are constituents of the appearance (or experience). However the appearance is also constituted not just by those objects and their characteristics (towards which e.g. the seeing is directed) but also by the act of perception, and the characteristics of that act (e.g. the presence of fog, the distance from the object, the angle, subject wearing glasses, and so on).

So, according to this there are no mental states at all which represent the objects of our perception and which would somehow give rise to the phenomenon of “phenomenal experience”. Instead the “experience” and “appearance” is to be read as something not in the head, but constituted by the objects and their characteristics, the act of perception and its characteristics, and any other things and their characteristics which contribute to the appearance because of its nature. Such is the thing with a fog, or with the glasses for example. They are not some necessary constitutive part of the seeing something, but because of the nature of seeing, the objects put between the eyes and the object affect the appearance of that object. Because of adding all those things to the appearance, this view can talk about such things as illusions and hallucinations both:

• Without a need for representation which would be veridical or not. Instead this view sees the illusions and hallucinations matter of appearing-same of different things given the variation of constituents of the appearances and their characteristics.
• Without denying the transparency of experience (which would be case for sense-data theory), and even because of the immediacy of the appearance, removing the need to talk about “phenomenal experience” as some kind of entity, which would be connected to what-is-it-like.

What about the what-is-it-likeness then? Where do we put it in this view? Well, it is so to say everywhere…For sure IT IS like something to see a rolling ball from three meters distance while wearing glasses, and looking through fog. What is it like? Well, it is like seeing a rolling ball from three meters distance while wearing glasses, and looking through fog.

I guess lot of people might not buy this as any kind of explanation, and require some reduction. Physicalists and dualists will say that we have a physical world, made from molecules (or maybe quantum foam), and that what-is-it-likeness should metaphysically or at least nomologically supervene on that.

But to me it seems this is looking at the things from wrong direction. Why can’t we instead ask this? - What will we find on physical/chemical/neurological level if we analyze the situation in which I’m looking at a rabbit. Now, we can analyze that situation, and say that in such and such case, there are photons bouncing off the rabbit’s fur, coming in the direction of our eyes, getting focused by our eye’s lenses, fall on the retina, where they affect the rod/cone receptors, and so on… , after some time resulting with specific movements of my lips tongue, jaw, synchronized with specific changes of tension of the vocal chords and controlling flow of air through them.

But, this won’t be explanation of the first situation, it is merely another description, in different terms. Now the terms are not “see”, “rabbit”, “5 meters from me”, all of which carry implicitly some what-is-it-likeness, but is speech in terms of different entities, which carry fully different what-is-it-likeness (what is it like to detect a photon, what is it like to base a belief there are photons on base of the explanation of photoelectric effect, etc…; what is it like to understand different kinds of organizations and emergent information processing of the neurons in a neural network, or to observe neurons through microscope, etc…).

So, on this view, the what-is-it-likeness is, so to say - just the consequence of the being a person which can see a rabbit, and which is seeing a rabbit. Of course, physics in its approaches ignores those things (justifiably) because of its nature, that only things which can be objectively evaluated count. However the consequence of it is that it can’t produce them back  from the impoverished picture of the world that it produces.

The picture of the physics is merely a picture which shows just an aspect of this world. It is an abstraction, and not a ground of the world.

Hallucinations and Dreams

In talking about illusions few posts back I said that the first thing to consider about illusions, is that two things can appear same depending on the characteristics of those things, but also depending on the characteristics of the act of seeing. That is, depending on the things like how far are the things, the angle from which we look at them, if there is maybe a fog, if we look at the things through some colored (or distorting) glass. One can also add to this facts about the person looking, like that the person is tired, that he has poor vision and has no glasses, and so on…

Once one accepts that two things (or situations, events, etc…) can look the same depending on all those characteristics even the things (situations, events, etc…) are different, we have easy explanation of what illusions are. They are phenomena where one of those situations is taken as standard, and the other one requires some complications, so that we use the “looks like X”, or “seems like X”, to explain how this second thing under the specific circumstances appears as the standard situation X. (Of course this isn’t isolated on appearances of single things, we can for example that it seems that someone is home, and that usage can be covered with explanation given here, we just say that the things in the house appear as they would if the person is at home. For example the person’s shoes are there, or his coat, or there is a sound coming from the TV, etc…)

The idea here is that we don’t need to suppose that there are such things as “visual fields”, or “phenomenal experiences” or “phenomenal appearances/seemings” which would be used to explain the possibility for illusions. But, what about hallucinations and dreams. After all, we can say that in the case of illusions, there is really two things that under certain circumstances appear same. But in the case of hallucinations or dreams… There is nothing. So, is this a proof that one can’t do away with those assumed “visual fields”/”phenomenal experiences/appearances/seemings”, and that we need them to explain dreams and hallucinations?

I think not. Imagine that we have some semi-transparent glasses with some (semi-transparent) picture painted on them. When we see through this kind of glasses, the picture from the glasses might contribute to the seeing in such way, that what we look at appears mixed with the picture on the glasses. For example when we look at a blank wall, it might appear as if the picture on the glasses is on the wall. Now imagine that we have something analogous to semi-transparent glasses “behind” the eyes. This “seeing affector” will affect the seeing in similar way to semi-transparent painted glasses, i.e. when we look at the wall, it will appear same as when we are looking at the wall which has picture on it.

I guess it is clear where this is going - the idea is that we can look at the back of our eyelids (or whatever we are looking when we have our eyes closed), but that the “seeing affector” can affect the seeing in such a way, that that “darkness” in fact appears as when thing appear in normal situations. (As argued in the ‘Appears as a Red Ball’ vs. ‘Is a Red Ball’ normal situation is that in which we learn the concepts and where we are ignorant of possible complications). So, anyway the idea is that something like this happens in the case of dreams - that what we look at in those dreams is actually nothing (taken in the sense how in the situation where there is no light, we can’t see anything), but that some “seeing affector” makes it appear as the cases in which we are seeing some particular things. Hallucinations would be similar case, just in those cases the “seeing affector”, affects seeing in such way that the situations in the world appear as it would appear when there is something there, which in fact in those cases isn’t. I’m not sure if in the hallucinations the hallucinated things are mixed with what we see in the real world, but if it does, then the analogy with semi-transparent glass might make even more sense…

Anyway, what is this “seeing affector”? I think that it is connected to what we call “imagination”. In the experiments done by Perky in 1910,published in the text called An Experimental Study of Imagination, she found out that the subjects were failing to distinguish banana imagined on the screen from a projected imagine of a banana on the screen (they instead e.g. reported that their imagined banana started rotating). So, the imagination there seems as a perfect candidate for this “seeing affector” (of course seeing is juts one perception, and there is no reason why it would be special, and why we couldn’t talk about general affect on the senses), as the way it affects what is seen seems very similar to the semi-transparent painted glass analogy.

Also one can point to the case of Zoltan Torey, which was presented on the All In The Mind radio show some time ago. Zoltan eyes were hurt in an industrial accident, so he was permanently blinded. But this is what he says in the interview:

Evidently the visual cortex, far from going blank and atrophying, it has picked up in acuity, and it is now totally under my command, so I virtually live in a visual space that I constantly produce myself. It is not really a canvas that I’m looking at, it is really visual space, so that I am—in the middle of which I find myself. So if I turn around, for example, I see what’s behind me, and as I turn my head around in the room where I am, so I orientate into the objects and furniture which I’m facing. It’s a completely technicolour, textured, visual world which apparently I continuously produce.

This points even better of the potential role of imagination as a “perceptual affector”, which I assume as an explanation of dreams and hallucinations.
Of course, this is not an argument against the idea of phenomenal experience, however I think it shows that one can argue about perception being directed to the outside things, or so to say that we are aware of the things themselves, and that we don’t need any middle entity which would be used to explain phenomena like illusions, hallucinations and dreams.

The Web of Belief - A Reading Note 2

Continuing from the paragraph quoted in previous post, Quine further writes (my italics):

In an early page we asked what sorts of things were the objects of belief. Then we gratefully dropped that question, noticing that we could instead talk of sentences and of believing them true. Now a similar maneuver conduces to clarity in dealing with the notion of observation: let us ask no longer what counts as an observation, but turn rather to language and ask what counts as an observation sentence.

What makes a sentence an observation sentence is not what sort of event or situation it describes, but how it describes it. Thus I may see the dean of the law school mail a birthday check to his daughter in Belgium. Saying so in these terms does not qualify as an observation sentence. If on the other hand I describe that same event by saying that I saw a stout man with a broad face, a gray moustache, rimless spectacles, a Homburg hat, and a walking stick, putting a small white flat flimsy object into the slot of a mailbox, this is an observation sentence. What makes it an observation sentence is that any second witness would be bound to agree with me on all points then and there, granted merely an understanding of my language. The witness would not be bound to agree that it was the dean, whom he or she might not know, nor expected to know anything about the check or a daughter in Belgium.

About agreement on observational sentences… As I’ve said in the past posts about illusions, it seems to me an obvious fact that different things, (and there I mean completely different things), can appear same way, depending on…

• characteristic of things
• which kind of sensual modality (or access) is in question
• the characteristic of that sensory access. For example in case of seeing - where we are seeing the things from, from what angle, do we wear glasses, are we sleepy, is there a fog, has anybody tampered with our brain, etc…

So, I think, that instead of “second witness would be bound to agree with me on all points then and there, granted merely an understanding of my language”, it would be better to speak that in case of observational sentences the second witness would be bound to agree that the situation seems such and such (granted merely an understanding of the language).

However, when we give up the possibility of sense-data language, I don’t think there is much sense to talk about infallibility of this kind of observational sentences. As there is not some factual state to which we have infallible access.

1. This IS NOT going back to sense-data description, or to some
   “phenomenal seeming” which would be divorced from the world, and about which we could give some statements.
2. According to the given analysis, the usage of “it seems that T” doesn’t mean that T is the case, but that among other things if T is the case, it would appear like whatever is the case appears now.
3. Seeming or Appearance are always to someone. Something can’t appear except to someone. (Notice the symmetry with e.g. One can’t see without seeing something) So implicitly ’seems’ is ’seems to me’. And this further acknowledges that there might be limits to my perception, or limits to what I have focused on, or what I’m ignorant of, because of which it appears to me that T.

So, the report is then not a infallible report of some phenomenal fact, but just report of a judgment that we do. Something like “That might be car, but maybe it is not.”. It is hard to see that as some kind of infallible report (though of course, it is hard to see how it can be wrong too).

The Quine’s position on this issue is not like the one I’m presenting here though. He seems to acknowledge that there is some kind of “private experience” and sense-data, and that there are such things as “I’m in pain” and “I seem to see blue”, which would be reports about that “private experience”…

We remarked that some philosophers have identified observations with events of sensation. It is thus not to be wondered that in some philosophical writings the title of observation sentence is reserved for sentences very different from observation sentences as we have defined them. It is reserved for introspective reports like “I am in pain” and “I seem to see blue now.” Such reports also have been rated as infallible. It must be conceded that they tend to be incontestable, because of the speaker’s privileged access to his or her private experience. But on this very point they differ diametrically from observation sentences in our sense. The situations that make them true are not ones to which multiple witnesses could attest.

What Are Illusions?

Chris at Mixing Memory has a post discussing few visual illusions, and also pointing to the announcement of the Best Visual Illusion of the Year.

As illusions came out also in the discussion we had with Richard in the comments, I thought I write a post about them.

What are illusions?

I guess the first intuitive answer would be: “the illusion is when something appears differently from how it is”. For example… two lines are same length, but they appear as having different lengths. Or… some color is gray, but we see it as yellow (check under Colour Perception illusions here). Or, there is no painted skull on the wall, but it seems to us that there is a skull. And so on…

But saying that “things appear differently from how they are”, has certain problems…

To point to one of the problems, I will point to a simple illusion. Take a white ball, for example. The ball is white and seems white. Now shine a red light onto it. The ball seems red, right? But what does “seems red” means there? We can say that white ball when under red light appears same as red ball under white light. But why put priority to how the red ball appears under white light? Why not say that red ball under white light appears as white ball under red light. What is so special about white light?

It seems to me, all we can say here, is that “red ball under white light” and “white ball under red light” appear the same. But if we don’t put priority on the one of those situations, where does the illusion come from?

Let me now propose the different description of what illusions are about, and then try to discuss it further:

Desc*: Illusion:One thing can appear as some another thing even if the things are different. In the course of our lives, we are encountering one of those situations more often, and the other situation usually requires a deliberate setup, or unlikely conditions which rarely occur. So, when we encounter the second situation we tend to judge it to be the first situation.

Illusions are then, not inherent in perception but are problem of the judgment, and usually of our ignorance of the complications in the situation, including the ignorance of the limits of our senses of perception.

So, how this differs from that first explanation that was proposed… Here is I think main differences:

1. That first explanation seems to imply that there is some such entity as “an appearance of a thing”. The logic is, if two things appear as same or similar, there is something identical in both cases, and that which is identical is “an appearance”, then this “appearance” is reified as independent thing, and most likely located in the mind/brain of the subject. (I will explain why in the next point). And because the appearance is something in the mind, to which we have direct access, if there is some mistake it has to be in the appearance itself. Instead of this kind of thinking, in this other way of looking at the illusions, we can negate that there are such things as appearances. Instead we speak about things appearing some way to us, that is, “appear” is used just as a verb and points to a relation between the thing and a person, and not to some other specific entity.

2.When talking about “the appearances” as entities in the previous case, they are imagined as simple, and things to which the subject now has some kind of direct and infallible access. It has to be infallible in this model, because if it is fallible then we haven’t solved the problem at all. Because of this fallibility it will be possible for the “appearance” to appear differently to the subject. And then we need to assume another level of “appearance” and so on…

3.Instead of that, in Desc*, we can include in the picture different properties of the subject’s access to the thing. To be more precise, in one case of appearance,  we can talk about:

• the intentional content (that to which we access through our senses - it can be a dog, cow, box, etc…)
• the type of intentional access (e.g. seeing, hearing, etc..)
• the different characteristics of the intentional access (for example, when we talk about seeing, a person can have glasses, the object might be put under different light or context, there can be fog, can look at the thing from different angles and distances, his eyes might be tired, someone might have rewired some things in the brain, etc…)

All those complications now become possible variables which might be tampered with, in order for something to look like something else.
For example we can say things like “a red ball in normal context appears like a white ball when one sees it through red glasses (or other way around)”, or “two lines with equal lengths drawn on paper with added arrows at the ends, look like lines with different lengths seen in perspective”, or that “gray circle in certain context seems like purple circle in normal context”, or that “a wall after we tamper with our eyes by fixing our eyes to certain color, appears like wall with a circle on it”, and so on…

So, in general there is nothing problematic here that is not present in simple case where a box, and a pyramid can appear same, if we look them from proper side. Things appear same because of limits of our perception, and some of those require a deliberate setup and are rarely encountered, so we will tend to judge such situations as others.

Take for example, a case of afterimage illusions. It requires first fixing your eyes at a specific place for certain time (say 20-30 seconds), and then looking at some blank wall, and not moving the eyes, in order to trick us into making wrong judgment. But we usually don’t make wrong judgment, and say that it “seems” because we are aware that the situation just seems like that other situation (as both situation seem same), and not that it is that other situation.

If you by chance glance towards the sun, it is pretty easy after that to tell that you have problem with the vision. It is not that we are thinking that world has gone weird and some green patch is moving over it.

Examples of Neural Adaptation - Visual Illusions

If you look at one of the white X-es in the colored circles (just look at the X, nothing else) on left side for 15-20 seconds, and then look at the X on the right side, you probably will see an afterimage, which will be in what Paul Churchland calls chimerical colors. Those are colors which can not appear to you in the normal looking at the things (for explanation of why afterimages happen, you can check this post at Mixing Memory, or check the page with videos on this blog, for a link to a presentation of Paul Churchland). Anyway the short story is that there is neural adaptation, a process where the neuron after given period of being excited with certain input, will “get tired” and go, all by its own, to its “normal” state. If after that the excitatory input goes away, there will be aftereffect such that the neuron will go to inhibited state, even that there is no input which inhibits it. Or the other way around, if inhibitory input is present for a longer time, and then removed, the neuron will go into excited state even there is no excitatory input.

As explanatory model it explains the afterimages (though of course doesn’t explain the phenomenal “look” of the colors, and taken by itself doesn’t explain some other visual illusions). But what is interesting is that this kind of aftereffects can be seen not just in case of color, but also other “more complicated” features. In the following picture there are faces of prototypical female on left side, prototypical male on the right, and a female/male morph image in the center. Try and concentrate on the left picture first (for e.g. 30 seconds), and then look at the center picture. Does it look as male or female? Then concentrate on the right picture for 30 seconds, and then look at the center picture. How about now?

If all is successful, this should show gender adaptation phenomenon, and if we connect this to the theory about how aftereffects in colors happen, should mean that probably there are also neurons in the brain which are used for recognition of gender. (also check Mixing Memory post for more).

BTW, I took the pictures for the second example from the Beauty Check site. It is about how morphing of multiple faces affects attractiveness of the face. Check it out here.