EXACTLY!
So you get why the title was funny, then?
Because not only can you not make a robot that feels pain, you can't make a human that feels pain, because pain doesn't exist (eliminativist).
Or, you do what Chiro just said, make it the "whole thing". Where, when Dennett says he or others or soldiers are in pain, it refers to the process in this diagram.
But if you are reductionist now, then you can make a computer that feels pain.
Hahaha. The old 'mood' function. Classic.
So...
Either: a) pain is a top level node in this diagram, or b) pain can not be a top-level node in this diagram.
So we ask what "pain" is. And we say things like, maybe it has something to do with reacting uncomfortably, or fear, or stress, or this chemical release, or that brain process, or that desire created or thwarted, of that uncomfortably memory had.
And we say, okay, so whatever "pain" is, it is associated with {a, b, c, ..., x, y, z}. And we can not think of anything else to put into that set, because we just exhausted the exercise.
So, it has been revealed that (((pain))) is no where to be found as a top-level part in the diagram. Let's call this PainL. Heh.
Then yeah. If you think PainL is no where on this diagram, but it is something not satisfied by anything in the process diagram with you. Then you know what, you can't make a computer that feels PainL.
But you know what else? I would then be a PainL eliminativist. Because I don't think PainL exists. I don't think any human has PainL. I don't think PainL is a coherent concept. Etcetera. I think it is stupid to speak of pain in terms of PainL.
Okay.
So you make the reductionist move. And you say, fine fine, pain is not a top-level node. Pain is just the process described in this diagram.
Let's call that PainM.
Then yes, you can make computers that feel PainM, obviously. And what it means for a human to be in pain is PainM. And no I'm not a PainM eliminativist. And yes I am a PainM reductionist. Yes I, people, soldiers, feel PainM.
I think both yourself and Dennett miss the point.
Either: a) pain is a top level node in this diagram, or b) pain can not be a top-level node in this diagram.
Why a top-level node? I'm not disputing this, but I'm curious why you stipulate that. Dennett's description seemed clear that pain has an origin in what he calls the 'old low path' (OLP) through the limbic system. The example of different effects of analgesia provided either prior to, or after, the painful stimulus showed that you either prevent the afferent information from being processed in the OLP, or the information is processed. In the latter case, the individual says that they indeed feel pain, but that they don't mind it, such as is the case with morphine.
That says to me that there is something essential to the kind of information that can be experienced as pain which originates in the OLP. But I'm not sure that it matters, and here's why:
So we ask what "pain" is.
To me, there is no answer to this question. I take Wittgenstein's side. It's up to the individual's interpretation whether they are feeling pain. In the case where someone who is doped up claims to feel pain, yet also says they don't "mind it", I'd just argue that the claim cannot be denied, but we might think there is a categorical error being made. If they don't 'mind it', then this contradicts something that we'd take to be an essential factor in pain, because pain is a word that describes something which just is something we mind very much.
Pain is an interpretation of a certain contextualized information content. It is something which has the potential to be pain in the right context. Consider the pain someone feels who has been in a coma and opens their eyes for the first in perhaps many weeks. There is a pain with the initial light stimulus. Is this because some real thing (that is, with a real ontic status) has been created? Or is the stimulus so sensitized from lack of use that some threshold for interpretation causes it to be experienced as unpleasant? The exact same information on a regular day would not be painful.
I would then be a PainL eliminativist.
This is the part that I really want to get across. What pain is, is not the important question. The important question is why there is something that it is like to be in pain.
Let's say Dennett is right, and functionalism is the case as a reductive explanation for the cause of pain. Great. I've got no real issue here (despite some of the important problems I've mentioned about his diagram once we get past the point of perception).
This doesn't explain the troubling part about pain. The causal mechanism of pain could be a hundred different things; what we want to know is why there should be something that it is like to be in pain.
I'm with Chalmers here. It is logically possible for the entire functional apparatus to be in place and working, yet where there is nothing that it is 'like' to be in pain.
To see why that's the case in Dennett's functional diagram, consider his only real intentional nodes in the process: belief and desire. There is only a vague way that these two things have experiential content. Like, if I say I believe that Sammy Sosa had 600+ career home runs, it would be strange to say that having that belief is an experience. It is the same with desire in many ways. What we could say is that these aren't associated with qualia, proper, but something below qualia which is psychological. Learning is another example of something that doesn't quite meet the threshold of a quale. Learning can be understood in totally psychological terms, just like belief and desire can.
It's easy to see where Dennett's functional map of pain could account for all of the causes of pain and the associated effects, but it would not have to account for there being something it is like to be in that pain. It could result in someone pulling their hand away from the hot stove without there ever being the qualia associated with being burned.
If it is logically possible for an atomically identical system in some possible world to function identically without there being something it is like to experience the pain, then pain supervenes on the physical properties of the brain, but it is not identical with those properties.
(post is archived)