Posts not getting through again, so continued in new thread.
Continuation from
https://groups.google.com/d/msg/talk.origins/R15fNPIizH8/ebVOoMgnAwAJ
On Sunday, November 29, 2015 at 2:03:38 PM UTC, Bill Rogers wrote:
> On Sunday, November 29, 2015 at 7:58:39 AM UTC-5, someone wrote:
> > On Sunday, November 29, 2015 at 11:13:36 AM UTC, Bill Rogers wrote:
> > > On Sunday, November 29, 2015 at 4:33:38 AM UTC-5, someone wrote:
> > >
> > > > >
> > > > > If the thing did not have a first person perspective, it would have to be in a different arrangement (since in my view the first person perspective corresponds to having a certain arrangement). If the thing had a different arrangement, it would behave differently.
> > > >
> > > > Yes I understand, but do you understand that you could say that about an epiphenomenal property which only certain arrangements had. Suggesting that if the arrangement was different the behaviour was different doesn't show that the property isn't epiphenomenal in the account. What you couldn't do if the property was epiphenomenal in the account is imagine that the same arrangement lacked it and show how it would make a difference to behaviour, like you could with the strong nuclear force for example. And the point has been made that imagining the property was lacking from the same arrangement wouldn't imply that the property was epiphenomenal, since as I mentioned you could do so with the strong nuclear force. And as you've stated yourself, you could, if you wanted to, do the same with the first person perspective. What I was asking you was if you did imagine the same arrangement to lack the first person perspective would your account suggest a behavioural difference like it would if you were to imagine the constituents of the same arrangement to lack a strong nuclear force, or as a dualist account would if that arrangement were imagined to lack a first person perspective, or would it suggest a philosophical zombie (something which lacks a first person perspective but behaves the same anyway)?
> > >
> > > Whether or not something has a first person perspective is, in my view, determined by the arrangement. In order for the thing *not* to have a first person perspective, you'd have to have a different arrangement. And a different arrangement would yield different behavior. The property of having a first person perspective is part of the arrangement - you cannot take it away without changing the arrangement.
> > >
> > > I agree with you that if you assume that you *can* take it away without changing the arrangement, and if the arrangement determines the behavior, then it follows that the first person perspective is epiphenomenal. But, you can prove anything by assuming your conclusion.
> > >
> >
> > As I pointed out, the conclusion isn't reached by any assumption. You can assume that you cannot have the same arrangement but not have the strong nuclear force, but you can still imagine it to be absent, and when you do you can see the influence that the model suggests it has on behaviour. When you do the same with the first person perspective in your model, you end up with a philosophical zombie, in other words the same behaviour but no first person perspective. Like with the strong nuclear force, you aren't assuming it is possible, just using your imagination to highlight the influence the feature has in your account. If you did the same thing using a dualist account you wouldn't end up with the same behaviour, because in the dualist account the feature isn't epiphenomenal.
>
> Of course the conclusion is reached from an assumption. Let's be explicit here.
> There are several cases.
>
> Case A
> Assume 1. The same physical laws determine the behavior of both conscious and unconscious things. 2. It is possible to have two things with identical behavior, one of which is conscious and one of which is unconscious. Then it follows that consciousness is epiphenomenal.
>
> Case B
> Assume 1. The same physical laws determine the behavior of both conscious and unconscious things. 2. It is NOT possible to have two things with identical behavior, one of which is conscious and one of which is unconscious. The it follows that consciousness is not epiphenomenal. These are the assumptions I make. THis is my view.
>
> Case C.
>
> Assume 1. Something other than the physical laws that govern unconscious systems determines the behavior of conscious systems. 2.It is possible to have two things with identical behavior, one of which is conscious and one of which is unconscious. The conclusion is that consciousness need not be epiphenomenal. This corresponds pretty well to the dualist alternative you mentioned.
>
>
No, as I explained to you case B does not make it follow that having a first person perspective is not epiphenomenal. You can assume 1. that the same physical laws determine the behaviour of both things with and without the epiphenomenal feature, and 2. It is NOT possible to have two things with identical behaviour, one of them have the epiphenomenal feature and not the other. Because there can be the case for example where certain chemical arrangements have a certain epiphenomenal property. Just because they will always have it, doesn't make the feature itself not epiphenomenal. But with a property that your account suggests is an epiphenomenal property (whether you realised it or not): If you were to think for a minute what difference your account suggests it would make to behaviour there will be no difference. Whereas with a property that your account doesn't suggest is an epiphenomenal property (whether you realised it or not, e.g. the strong nuclear force): If you were to think for a minute what difference your account suggests it would make to behaviour there would be a difference. The first person perspective falls in the former category in your account, and that is because of your account, not the feature itself, because it isn't the same in a dualist account for example.
[Just for the record you made a change in another post (responded to the older one as I had already done so, but when checked the post hadn't come through):
Case C.
Assume 1. Something other than the physical laws that govern unconscious systems determines the behavior of conscious systems. 2.It is possible to have two things with identical *physical arrangement*, one of which is conscious and one of which is unconscious. The conclusion is that consciousness need not be epiphenomenal. This corresponds pretty well to the dualist alternative you mentioned.]
>
>
> > >
> > > I don't agree with you here. Whether or not something has a first person perspective is observable, in principle, from a third person perspective. Having a first person perspective is a behavior and is equivalent to being conscious. The first person perspective, itself, is incommunicable. It's Wittgenstein's beetle in a box. All you ever see about anything except yourself is a third person perspective, but it it perfectly easy to determine, from the third person perspective, that a cup has no first person perspective and a person does. All we've done in the last few threads has been to switch the phrase "being conscious" with the phrase "having a first person perspective." In my view, they are equivalent. But I think you are not interested in my view. If you were, you wouldn't persist in ignoring the straightforward version of it that I've given you many times. You're interested in trying to score points, not in understanding.
> >
> > I'm interested in breaking your little story, and already it is now at the point that it is no longer any good for you to be able to claim that in your theory certain behaviour would have a first person perspective. Because your claim has now changed from your theory assumes that certain behaviour will have a first person perspective to it is observable from a third person perspective that it does. A claim that the burden of proof is on you to prove. And basically it was rubbish,
>
> When you were a child, you learned to identify conscious things from their behavior. It's how we decide how to use the word conscious. We can refine our definition, but it still comes from observed behavior.
>
> >but nevertheless I have a simple thought experiment that shows that you can't in principle observe it, so I don't even need to bother about who the burden was upon, in case you dispute that you'd need to prove your statement of fact.
> >
> > Take a robot for example, it can behave in a way which you can claim is conscious behaviour, but you can't observe whether it has a first person perspective, or if it had what it was like. That is simply a fact, but if you don't want to accept it,
>
> Of course it's not a fact. In my view, if that robot behaves in a way that I consider to be conscious, then of course it has a first person perspective.
>
It doesn't matter whether "in your view" if it behaves in a certain way that you would consider it to be conscious, because although you might think it would have a first person perspective, you didn't actually observe that it did, you just assumed it did because you assumed your view was correct. But a dualist could assume your view is wrong, and their observations be compatible with their view, because they didn't observe the first person perspective that they assumed the robot didn't have for example. And it is a fact that they didn't observe it, and not a case that they weren't denying what they observed. You would be lying if you stated in the robot's case that you'd actually observed it though, rather than admitting it was just something that you were happy to assume it had.
> >then I'll illustrate the point using a simple thought experiment. Imagine some scientists build a robot, the Mark 1, which gives the type of behaviour that *you* would classify as consciousness. It has cameras for eyes, with each pixel having three 8-bit intensity values (so a range from 0-255) each represent a colour intensity. These come through 3 channels A, B, and C. Internally, in software ("version 1"), the robot holds a table so to speak of linking words to the colour values. E.g. if channel A has an intensity of 255, and B & C intensities of 0, the robot will use the word "red" to describe the colour if B is 255 and A & C are 0 it will use the word "green" and if C is 255 and A & B are 0, it will use the word "blue" to describe the colour. You can imagine that all its reactions to colour are based off the ABC channel values.
>
> If all of its reactions to colors are based on the ABC channel values then I already know that the robot does not behave as I do. Our human response to color is far more complex than simply identifying wavelengths. Just based on that alone, I might decide the Mark 1 robots did not exhibit conscious behavior.
>
> >
> > Then imagine that the Mark 1 can take two types of eye camera. The RGB eye camera which, for each pixel, uses channel A for the red light intensity data, channel B for the green light intensity data, and channel C for the blue light intensity data. Or the BGR eye camera which uses channel the A channel for the blue light intensity data, channel B for the green light intensity data, and channel C for the red light intensity data.
> >
> > Then consider four Mark 1 robots, two with RGB eye cameras, and two with BGR eye cameras. From each pair, one is in a red room with a blue table, and the other is in a blue room with a red table. Not presumably because the robots behave in a way which *you* define as conscious, you are going to claim that not only do they have a first person perspective, but that you can observe it from a third person perspective. So in such a scenario how could you observe whether both the Mark 1s (the one with RGB camera eyes, and the one with BGR camera eyes) in the red room with a blue table experience the room with similar red qualia as you would, and the table with a similar blue qualia as you would? Explain what type of experiment you could do to tell, or what other method of observation you'd use.
>
> I can't even tell if you experience the same qualia I do, much less a robot, but I'm reasonably sure you have a first person perspective nonetheless.
>
Is what it is like for the robot not something you can observe from a third person perspective then, even in principle? Because if not, as I stated a few posts ago, and has been snipped:
---
The reason I ask is that if my assumption was correct and all behaviours are in principle observable from a third person perspective, then the first person perspective itself, the feature that a cup doesn't have, isn't a behaviour, because it isn't observable from a third person perspective. And if it isn't a behavioural feature, then it cannot be reduced to other behavioural features.
---
Similarly, if what it is like for the robot isn't something that can in principle be observed from a third person perspective, then it isn't a behavioural feature, and therefore cannot be reduced to other behavioural features.
> If I take the robots outside and they behave consciously, make long term plans, avoid danger, tell me what they think about politics, tell me how they feel about each of the other robots, respond appropriately to conversations, etc, then I'll conclude they have a first person perspective. I won't be able to *have* their individual first person perspective, but I'll be happy enough to conclude they have one.
>
> The whole inverted spectrum thought experiment is old news.
>
> You can find it here:
>
>
http://plato.stanford.edu/entries/qualia-inverted/#BehUndSce
>
> Section 3.1 seems to be closest to your argument and, when used as an argument against physicalism it is pretty much equivalent to the zombie argument. From the linked text....
>
> "Although there is some controversy about just how science-fictional anti-functionalist inversion scenarios need to be, anti-physicalist inversion scenarios are usually supposed to be very remote from actuality, and their possibility is hotly disputed. Almost uncontroversially: if an inverted spectrum argument against physicalism works at all, then a simple zombie scenario will equally serve the purpose."