Has consciousness been fully explained?

[rocketdodger]

Can we just cut to the chase and stop all this minor quibbling?

We are not talking about a system that has one constraint on it with life, nervous systems, or computers. There is the entropy issue. There is the maintenance of integrity issue. There is the ability to transmit information issue. There is the ability to transmit information maintaining its integrity issue. There is the computation issue. Etc, etc ad nauseum.

That other stuff in the world include one or the other of these constraints is not what it is important. They have to include all the important contstraints.

Haven't you picked up on their agenda, yet?

If they can show that by all scientific accounts a neuron and a rock don't really have any qualitative differences then by default the conclusion is that there must be some qualitative difference that is not accessible to science.

It isn't quibbling, it is their goal.

 

[blobru]

I'm not arguing against the use of common terms - it's just that if we use "information" to mean "stuff that is of interest and usefulness to us" - which is how we use it, normally, when talking about computers - then we have to be very careful not to extend that functional description into the physical realm.

Then you're arguing against the sloppy use of common terms, which is the business of philosophy. Fair enough. You've introduced the term "information", which is a slippery one, admittedly, and defined it as "stuff that is of interest and usefulness to us" (its normal usage, you say, though in philosophy we don't limit ourselves to normal usage; we make it explicit, then refine it).

If you stick with that definition of information, then of course it will be impossible to find "information" in mindless nature: there will be only trees falling in the forest, with no one around to hear, let alone note that the trees are part of a forest, that the forest can be considered a system whose integrity effects the integrity of every tree in it and vice versa, etc. More limiting, that "information" defined as communicated orderly sequences of state transitions ("behaviors" we tend to call them, when describing animals) which are useful ('meaningful" we tend to say, when referring to people) to the integrity of a natural system can overlap with and even embrace the narrower concept of human "information", defined as communicated, via language rather than evolution, behaviors which are useful to us: that is, useful to us as systems which rely on such meaningful information for knowledge of ourselves and our environment and how to function productively in it. That analogy seems to me, and many others, an apt comparison well worth pursuing, to see what insights it provides (who knows: it may turn out the distinction between us and the rest of nature is a culturally-inherited impediment to understanding; or it may turn out to be vital that we maintain the distinction and never let it go; or a little, or a lot, of both).

Anyway, for the purposes of discussion, to single out this thought-provoking similarity between mindless nature and mindful us, as philosophers, for curiosity's sake, let's explicitly define information as something like "communication which is relevant to the integrity of a system", and see where it leads (all the while, as philosophers, mindful how we have expanded the definition of course, and the previously observed differences between the systems to which we apply it).

In the physical realm, all the effects that one particle has on its neighbours are of equal status, and convey equal information.

Wasn't sure how you're defining "information" here. Local change in state that we observe/infer by its effects? Or 'mindless' cause-and-effect between a particle and its neighbors?

When a theory relies on the confusion between two realms, one can legitimately doubt its value.

Good. Good philosophy relies on doubt. Every analogy between realms is open to doubt. However, we shouldn't let doubts overwhelm us, imo, and dismiss analogies even as we have barely begun to explore them (nor should we assume their success until thoroughly explored; so, by all means, keep on doubting)... (tho' within reason, eh?)

 

[Ichneumonwasp]

Haven't you picked up on their agenda, yet?

If they can show that by all scientific accounts a neuron and a rock don't really have any qualitative differences then by default the conclusion is that there must be some qualitative difference that is not accessible to science.

It isn't quibbling, it is their goal.

But there is no qualitative difference; it's just different chemistry. He should be trying to show that there is a qualitative difference between life and a computer (or rock) because that would preclude the possibility of a computer doing what anything living can do.

The absence of a qualitative difference is our argument. There simply is no qualitative difference (except for the difference in kinds of chemistry involved). There are lots of constraints on the system that preserve the integrity of information. Full stop.

ETA:

Depending, of course, on what we mean by "qualitative difference".

 

[rocketdodger]

But there is no qualitative difference. He should be trying to show that there is a qualitative difference between life and a computer because that would preclude the possibility of a computer doing what anything living can do.

The absence of a qualitative difference is our argument. There simply is no qualitative difference. There are lots of constraints on the system that preserve the integrity of information. Full stop.

ETA:

Depending, of course, on what we mean by "qualitative difference".

But they don't need to show a "qualitative" difference because intuitively there is one (or at least it seems that way ) -- our consciousness. The process is a sort of sick obsession with that old religious strawman "science says we are nothing but particles, so obviously science is wrong because clearly we can do all kinds of things that particles cannot." As with any strawman, efforts to get rid of it are resisted in full by those clinging to it.

At the root of the argument the two camps can be broken down like this:

On the one hand, the computationalists like you and I say "hey, there is no *strictly* qualitative difference, but if you have enough quantitative differences stacked on top of each other it ends up looking like a qualitative difference." And of course those stacked quantitative differences are exactly what you mentioned recently -- the information thing, the entropy thing, the computation thing, etc.

On the other hand, the anti-computationalists say "we know there is a qualitative difference -- human consciousness -- and if you can't account for that qualitative difference with science then there must be some non-science going on that is responsible for it."

Of course there are outliers like Piggy who want to be computationalists but still think there is a strictly qualitative difference, etc, but for the most part everyone falls nicely into the two above camps.

So the strategy employed by westprog for years now, and more recently by Malerin and all the other anti-computationalists who jumped on the bandwagon, is to oppose every effort we make to show any objective scientific difference at all between life, computers, and all the rest. They know from life experience (they aren't stupid, after all ) that yes, actually, stacked quantitative differences can appear qualitative and they know that every little quantitative difference science can show is just another step towards the computationalist position. Thus they fight every little step.

It becomes pretty obvious when you look at their posting history in totality: Well, computation exists in everything, so the brain isn't special in that regard. Well, even if computation is based on switching, switching occurs everywhere. No, transistors don't behave any differently than the rest of the universe. No, neurons don't exhibit any kind of special behavior compared to rocks. No, life is no different thermodynamically than anything else, all kinds of stuff decreases local entropy just like life does.

I have been part of this discussion for like 4 years now and the debate strategy has been crystal clear the whole time -- minimize objective scientific differences between people and rocks in order to make room for religious differences between people and rocks.

 

[Ichneumonwasp]

But they don't need to show a "qualitative" difference because intuitively there is one (or at least it seems that way ) -- our consciousness. The process is a sort of sick obsession with that old religious strawman "science says we are nothing but particles, so obviously science is wrong because clearly we can do all kinds of things that particles cannot." As with any strawman, efforts to get rid of it are resisted in full by those clinging to it.

As you well know, calling the presence of consciousness or life a qualitative (absolute) difference is begging the question, unless anyone can show that there is an unquestioned difference. That, of course, is the reason why Searle tried the Chinese Room argument and more recently the argument regarding computation.

The fact that there is meaning inherent to our nervous system belies Searle's point; and I continue to maintain that he should know better.

 

[rocketdodger]

The fact that there is meaning inherent to our nervous system belies Searle's point; and I continue to maintain that he should know better.

You have to be careful with "meaning" though because many people will take it out of context -- look how many people in this thread alone want to claim that "meaning" requires consciousness to begin with. Hello, circular logic!

People well versed in the mechanisms of natural selection can see how the meaning the activation of a neuron is objectively defined by how that neuron evolved. People without such imagination ... can't. For them, meaning is inherently linked with consciousness.

I wish there were better words for these things, since "meaning" and "purpose" and all the rest are so nested in our anthropomorphic tendencies.

 

[Ichneumonwasp]

You have to be careful with "meaning" though because many people will take it out of context -- look how many people in this thread alone want to claim that "meaning" requires consciousness to begin with. Hello, circular logic!

People well versed in the mechanisms of natural selection can see how the meaning the activation of a neuron is objectively defined by how that neuron evolved. People without such imagination ... can't. For them, meaning is inherently linked with consciousness.

I wish there were better words for these things, since "meaning" and "purpose" and all the rest are so nested in our anthropomorphic tendencies.

Agreed. Especially since I think our language, having words like consciousness be nouns for instance, creates this problem in the first place.

 

[AlBell]

People well versed in the mechanisms of natural selection can see how the meaning the activation of a neuron is objectively defined by how that neuron evolved. People without such imagination ... can't. For them, meaning is inherently linked with consciousness.

I wish there were better words for these things, since "meaning" and "purpose" and all the rest are so nested in our anthropomorphic tendencies.

I at least am having trouble parsing "People well versed in the mechanisms of natural selection can see how the meaning the activation of a neuron is objectively defined by how that neuron evolved.".

Could you explain that in a bit more detail?

 

[westprog]

Can we just cut to the chase and stop all this minor quibbling?

We are not talking about a system that has one constraint on it with life, nervous systems, or computers. There is the entropy issue. There is the maintenance of integrity issue. There is the ability to transmit information issue. There is the ability to transmit information maintaining its integrity issue. There is the computation issue. Etc, etc ad nauseum.

That other stuff in the world include one or the other of these constraints is not what it is important. They have to include all the important contstraints.

Well, the best thing to do is to list off the constraints that apply to life, list off the constraints that apply to computing, match the ones that are in common, and verify that these constraints only apply to life and computing. Then, it might be possible to conjecture about what that means.

The entropy thing just doesn't apply.

 

[Ichneumonwasp]

Well, the best thing to do is to list off the constraints that apply to life, list off the constraints that apply to computing, match the ones that are in common, and verify that these constraints only apply to life and computing. Then, it might be possible to conjecture about what that means.

The entropy thing just doesn't apply.

Of course the entropy thing applies. It is one of the constraints -- the influx of energy to counteract the increase in entropy locally. That there are other constraints is also fairly obvious. The only reason I brought up the "entropy" thing is that it cuts out the example you provided as a potential counter. But it is very wrong to suggest that local influx of energy is not important both to computing and to life; that in some what "it does not apply". Life doesn't exist without it and neither does a computer unless you can show me one that spontaneously manifests from spare parts.

 

[westprog]

But they don't need to show a "qualitative" difference because intuitively there is one (or at least it seems that way ) -- our consciousness. The process is a sort of sick obsession with that old religious strawman "science says we are nothing but particles, so obviously science is wrong because clearly we can do all kinds of things that particles cannot." As with any strawman, efforts to get rid of it are resisted in full by those clinging to it.

At the root of the argument the two camps can be broken down like this:

On the one hand, the computationalists like you and I say "hey, there is no *strictly* qualitative difference, but if you have enough quantitative differences stacked on top of each other it ends up looking like a qualitative difference." And of course those stacked quantitative differences are exactly what you mentioned recently -- the information thing, the entropy thing, the computation thing, etc.

On the other hand, the anti-computationalists say "we know there is a qualitative difference -- human consciousness -- and if you can't account for that qualitative difference with science then there must be some non-science going on that is responsible for it."

Of course there are outliers like Piggy who want to be computationalists but still think there is a strictly qualitative difference, etc, but for the most part everyone falls nicely into the two above camps.

So the strategy employed by westprog for years now, and more recently by Malerin and all the other anti-computationalists who jumped on the bandwagon, is to oppose every effort we make to show any objective scientific difference at all between life, computers, and all the rest. They know from life experience (they aren't stupid, after all ) that yes, actually, stacked quantitative differences can appear qualitative and they know that every little quantitative difference science can show is just another step towards the computationalist position. Thus they fight every little step.

It becomes pretty obvious when you look at their posting history in totality: Well, computation exists in everything, so the brain isn't special in that regard. Well, even if computation is based on switching, switching occurs everywhere. No, transistors don't behave any differently than the rest of the universe. No, neurons don't exhibit any kind of special behavior compared to rocks. No, life is no different thermodynamically than anything else, all kinds of stuff decreases local entropy just like life does.

I have been part of this discussion for like 4 years now and the debate strategy has been crystal clear the whole time -- minimize objective scientific differences between people and rocks in order to make room for religious differences between people and rocks.

The above I think shows the fundamental difference in this discussion.

The particular point about entropy is a fairly typical example. The Wasp suggested that entropy reversal was a characteristic exhibited by life and computing, and that such thermodynamic uniqueness was evidence in favour of the computationalist approach.

I pointed out, with references, that this was not, in fact, the case. Entropy reversal of this kind is extremely common. I also pointed out that this is quite a common misunderstanding, and that it has historically been misused by creationists intent on demonstrating the special position of life.

That should really have been the end of it. The point was pretty well indisputable. However, as can be seen from the above, the evidence has little to do with it. Maybe I was right about this one point - but my motives were suspect. What's the point in being correct if you're following the wrong cause?

RD simply can't leave well alone. We saw it when he misunderstood a simple point about relativity. So when I make an unarguable point about entropy, he comes back with "No, life is no different thermodynamically than anything else, all kinds of stuff decreases local entropy just like life does". Can he deny that this is true? No, the references are very clear - except from the usual creationist crazies, who would entirely support his interpretation. This would be a good time to abandon the latest half-baked theory thrown together on the spur of the moment, tying together computation and thermodynamics. Instead he accuses me of fighting "every little step".

In real science, done by real scientists, they are actually grateful to have holes in a theory pointed out. Testing ideas rigorously doesn't harm them - it makes them stronger. If a computationalist theory were really resting on the idea that life has some unique thermodynamic status, then it would quickly collapse. I'm pretty sure that the theory doesn't, in fact, require any such thing. So why defend the error and attack the correction?

 

[blobru]

... I wish there were better words for these things, since "meaning" and "purpose" and all the rest are so nested in our anthropomorphic tendencies.

"Philosophy is a battle against the bewitchment of our intelligence by means of language" -- one of Wittgenstein's more famous quotes. Anthropomorphism certainly plays a large part in that bewitchment. And convention. And subjectivity. Ignorance. Arrogance. Metaphysics. Failure to note context (is a big one). And language itself: a toolbox of thousands of words with which to label billions of phenomena. Words end up doing double, triple, n-tuple duty; confusion, frustration, and/or hilarity ensues. Philosophy, done right, tries to sort it out, and keep a straight face; or failing that, throws its hands up and laughs at itself.

In the context of discussions of combined natural and human senses of "information", for meaning something like "relevance to the integrity of a system" might work; while for purpose maybe substitute "integrity" or "function" (keeping the system going, functioning optimally, etc.).

 

[Ichneumonwasp]

The above I think shows the fundamental difference in this discussion.

The particular point about entropy is a fairly typical example. The Wasp suggested that entropy reversal was a characteristic exhibited by life and computing, and that such thermodynamic uniqueness was evidence in favour of the computationalist approach.

Now what a second. What I said was that there are multiple constraints on what we call life and on computing and that localized reversal of entropy is one of them. I also said that other examples showing this not to be absolutely unique would actually help the argument and not destroy it.

Sure, we can easily point to other chemical reactions and identify localized reverses in entropy. The formation of any macromolecule is an example. But the formation of macromolecules is also an important step in the formation of life -- which depends on those localized reverses in entropy.

I initially said that the only things I could think of that show those localized reverses are man made things and life, but sure there are other chemical reactions that do the same if we only focus on the creation of order locally (which is the proper approach).

I am not the one claiming uniqueness here -- you are the one pushing for it. I don't think for a second that there is anything absolutely unique about life.

So, if we want to start with a list of constraints, let's start with a simple one:

1. Localized reverses in entropy
2. Complexity (throw out simple repeating monmers that make polymers)
3. Multiple different interacting parts.
4. The ability for some of the multiple parts to reset and perform the same activity again and again.

That's a simple start to the list of constraints that include computers and life. We can add new constraints as anyone thinks of things in nature that have all four of these.

 

[Meed]

Now what a second. What I said was that there are multiple constraints on what we call life and on computing and that localized reversal of entropy is one of them. I also said that other examples showing this not to be absolutely unique would actually help the argument and not destroy it.

What is the argument?

I don't get what the "computationalist" position is other than "consciousness is SRIP" and "a simulation of a conscious entity would actually have consciousness". Neither of these seem to have been supported.

Does defining "computation" in a way that captures life and computers but excludes everything else relate to these claims somehow? Of course I'd expect it's theoretically possible if you throw together a hundred constraints, but I don't see what it accomplishes.

 

[rocketdodger]

The above I think shows the fundamental difference in this discussion.

The particular point about entropy is a fairly typical example. The Wasp suggested that entropy reversal was a characteristic exhibited by life and computing, and that such thermodynamic uniqueness was evidence in favour of the computationalist approach.

I pointed out, with references, that this was not, in fact, the case. Entropy reversal of this kind is extremely common. I also pointed out that this is quite a common misunderstanding, and that it has historically been misused by creationists intent on demonstrating the special position of life.

That should really have been the end of it. The point was pretty well indisputable. However, as can be seen from the above, the evidence has little to do with it. Maybe I was right about this one point - but my motives were suspect. What's the point in being correct if you're following the wrong cause?

RD simply can't leave well alone. We saw it when he misunderstood a simple point about relativity. So when I make an unarguable point about entropy, he comes back with "No, life is no different thermodynamically than anything else, all kinds of stuff decreases local entropy just like life does". Can he deny that this is true? No, the references are very clear - except from the usual creationist crazies, who would entirely support his interpretation. This would be a good time to abandon the latest half-baked theory thrown together on the spur of the moment, tying together computation and thermodynamics. Instead he accuses me of fighting "every little step".

In real science, done by real scientists, they are actually grateful to have holes in a theory pointed out. Testing ideas rigorously doesn't harm them - it makes them stronger. If a computationalist theory were really resting on the idea that life has some unique thermodynamic status, then it would quickly collapse. I'm pretty sure that the theory doesn't, in fact, require any such thing. So why defend the error and attack the correction?

1) Speaking to people in the 3rd person is bizarre.

2) You neglected to actually defend yourself against my claim regarding your motives. Does this mean your motives are what I claim they are?

3) I asked you how time dilation fits with your claim about "real" processes having "time dependence," and to this day you still haven't given an answer. Instead you just wave your hand and state that I don't understand relativity. lol

 

[rocketdodger]

What is the argument?

I don't get what the "computationalist" position is other than "consciousness is SRIP" and "a simulation of a conscious entity would actually have consciousness". Neither of these seem to have been supported.

Does defining "computation" in a way that captures life and computers but excludes everything else relate to these claims somehow? Of course I'd expect it's theoretically possible if you throw together a hundred constraints, but I don't see what it accomplishes.

The issue is whether or not the abstractions used by computer science, such as SRIP, describe any system at all in nature or only systems that satisfy certain constraints.

Why is this an issue?

Because the debate strategy used by Westprog and Malerin is to argue that if every system in nature can be described by the abstractions of computer science, such as SRIP, then every system in nature is conscious according to the computationalist position and thus the computationalist position must be wrong.

To move past this stupidity we are trying to show that um no, not every system in nature can be described using the abstractions of computer science. Only certain systems can, for example life and computers. Thus saying that consciousness is a type of SRIP does not imply that everything is conscious.

 

[Ichneumonwasp]

What is the argument?

I don't get what the "computationalist" position is other than "consciousness is SRIP" and "a simulation of a conscious entity would actually have consciousness". Neither of these seem to have been supported.

Does defining "computation" in a way that captures life and computers but excludes everything else relate to these claims somehow? Of course I'd expect it's theoretically possible if you throw together a hundred constraints, but I don't see what it accomplishes.

I have no idea what the argument is. Westprog is the one who is pushing me on this and I am replying. You'll have to ask him. All I know is that it will be easy to find some set of constraints that includes man made things and life and excludes everything else in nature. I think it's a stupid exercise, though.

The point that I think is important is that there are multiple constraints on any system that that we would call conscious or that we would say 'computes'. It is the constraints on the system that create the ability to compute. Falling rocks don't compute, even if they can be seen by another person who defines them as adding together. For a system to compute it must be constrained in several ways so that information is preserved, can be manipulated, etc. I think the more interesting question, and one that I can't fully answer, is what constraints are necessary to make some function into computation.

I'm the wrong person to ask about computationalism because I don't define myself as a computationalist. I'm still very much moer a physicalist in my outlook (I know brains can do it and one of the things they clearly do is modulate inputs into neurons, something that computers can't do at their most basic level -- computers would have to simulate that activity). But I am very willing to argue against what I can see as misguided arguments against the computationalist position.

The more I have thought about the simulation issue, the more convinced I am that a simulation should be able to create consciousness. I don't see a clear argument against it. I think the best argument is something along the lines that the movement of electrons through gates or switches is not that close to the ways that neurons interact, but I don't see why that should matter all that much. That does not sound like a knock out argument to me, especially considering the fact that computers can do all sorts of things that we used to think impossible for any man made thing to accomplish. I still don't see why a simulated person, simulated down to every detail that we could abstract from the real world could not be said to be conscious if we made it possible to interact with 'him/her'. The outputs from actual humans (on a neuronal level) are relatively simple, so there shouldn't be that much problem linking up the stuff going on in the computer responsible for the simulation to some sort of audible/visual output. (ETA: Let's say for example that we simulate this person and have someone in the simulation throw a stick at his head without there being any line of code that says -- move if a stick is thrown at your head -- and I think if the simulation were well done he would duck to avoid the stick.)

 

[Meed]

I have no idea what the argument is.

You said "I also said that other examples showing this not to be absolutely unique would actually help the argument and not destroy it." I was asking what argument you were referring to.

The more I have thought about the simulation issue, the more convinced I am that a simulation should be able to create consciousness. I don't see a clear argument against it. I think the best argument is something along the lines that the movement of electrons through gates or switches is not that close to the ways that neurons interact, but I don't see why that should matter all that much. That does not sound like a knock out argument to me, especially considering the fact that computers can do all sorts of things that we used to think impossible for any man made thing to accomplish. I still don't see why a simulated person, simulated down to every detail that we could abstract from the real world could not be said to be conscious if we made it possible to interact with 'him/her'. The outputs from actual humans (on a neuronal level) are relatively simple, so there shouldn't be that much problem linking up the stuff going on in the computer responsible for the simulation to some sort of audible/visual output. (ETA: Let's say for example that we simulate this person and have someone in the simulation throw a stick at his head without there being any line of code that says -- move if a stick is thrown at your head -- and I think if the simulation were well done he would duck to avoid the stick.)

I agree there's no knock out argument against it (at least that I've seen), but I haven't seen any particularly strong arguments for it either. I think we agreed that a simulation of water wouldn't be wet, a simulation of an orange wouldn't be an orange, a simulation of a ball rolling wouldn't be a rolling simulation of a ball, etc. It seems consciousness is treated uniquely as something that would actually be real if simulated, while everything else wouldn't.

 

[Meed]

The issue is whether or not the abstractions used by computer science, such as SRIP, describe any system at all in nature or only systems that satisfy certain constraints.

Why is this an issue?

Because the debate strategy used by Westprog and Malerin is to argue that if every system in nature can be described by the abstractions of computer science, such as SRIP, then every system in nature is conscious according to the computationalist position and thus the computationalist position must be wrong.

I agree with the argument. If you don't have a definition of SRIP that excludes things you don't consider conscious then it's useless as a definition/explanation of consciousness. The onus is on you to properly define the terms you propose being relevant.

Of course, most people would not consider electronic toasters or computers conscious, so even if you manage a necessary and sufficient definition that includes toasters, computers, certain other machines and animals with brains, while excluding everything else then you will not have come up with a definition capturing what most people consider consciousness to be.

 

[Ichneumonwasp]

You said "I also said that other examples showing this not to be absolutely unique would actually help the argument and not destroy it." I was asking what argument you were referring to.

Ah, sorry, I misunderstood. That argument would be that there is nothing absolutely unique about life. I don't hold that there is something that is absolutely unique about life -- it is very complex chemistry. So, other examples of locally reversed entropy help to show that this, like everything else, is just part of physics.

That argument has more to do with ontology, specifically with monism. The idea that there is something absolutely unique about life or about consciousness implies a dualistic perspective.

I agree there's no knock out argument against it (at least that I've seen), but I haven't seen any particularly strong arguments for it either. I think we agreed that a simulation of water wouldn't be wet, a simulation of an orange wouldn't be an orange, a simulation of a ball rolling wouldn't be a rolling simulation of a ball, etc. It seems consciousness is treated uniquely as something that would actually be real if simulated, while everything else wouldn't.

Simulated water would not be wet in this world and a simulated orange would not be a 'real' orange. Within the simulation, however, the water would have all the properties that water has in this world and the orange in the simulation would be the same in the simulation as an orange in this world. A simulation of a rolling ball would not be a ball rolling in 'reality' but the 'rolling' would be identical -- only the ball would not be real. Simulated action is still action; as long as it occurs for the same reasons (in other words, as long as it isn't just some code saying 'put a pixel here and make it make the pixels change to make it look like a ball is rolling on a screen). The action of rolling or the action of consciousness still occurs in the simulation just as it does in the real world. I don't see a way of separating the two actions (except for where they occur). There is nothing special about consciousness in this regard. The difference, at least to my mind, is between objects and actions. Objects can only be isomorphic, but I think actions should be identical because they only consist in a relation of parts.