Chapter 6: On the Probability of Panexperientialism

[hypnagogue]

If we accept that panexperientialism is a logically coherent worldview (see chapter 5), we are left with the question of how likely it is to be true of our world. This chapter is devoted to answering that question.

Panexperientialism is roughly the view that subjective, phenomenal experience outruns cognition-- that is, that experience is not limited to cognitive systems. Before we can proceed to evaluate panexperientialism and its opposition, we need an idea of what is meant by the term "cognitive." It is difficult to pinpoint exactly what should count as cognitive and what should not; ideally, we would like the word to pick out systems that are similar to human brains in certain sophisticated regards. Rosenberg nominates Robert Kirk's "basic package" as a good example of the sense of the word "cognitive" that he adopts in this chapter. The basic package is a group of information processing capacities that includes the following:

* The possession of needs and goals.
* Self-directed behavior.
* The capacity to acquire information which is for the system in the sense that it can pursue goals by modifying its behavior in response to the information.
* The ability to store information.
* The ability to assess information.
* The capacity to make decisions. (pg. 105)

The arguments to follow are not predicated on following Kirk's sense of the word "cognition" to the letter. Rather, Kirk's basic package is to be taken as a right-spirited attempt at capturing the essence of the word in a way that corresponds to relevant high-level characteristics of human brain function, without being so general as to lose its intuitive meaning.

In order to assess the likelihood that panexperientialism is true, Rosenberg does not provide direct arguments making the case for panexperientialism, but rather argues against rivaling theories that restrict experience only to cognitive systems (call these cognitive theories of experience). On any such theory, there must be some class of base properties of cognitive systems that accounts for experience. Rosenberg addresses three of the most popular proposals for such base properties-- complexity, functionality, and biology-- although the moral behind his critiques could probably be extended to any cognitive theory of experience.

Recall arguments from chapters 2 and 3 that p-consciousness is not entailed by physicalism. If this is the case, then p-consciousness cannot be constructed from existing physicalist ontology, and we need some sort of new fundamental objects/processes/laws to account for it. In the absence of hard empirical data about p-consciousness beyond our own first person cases, we can evaluate any such fundamental laws by way of comparison to already established fundamental laws, such as fundamental physical laws that govern gravitation or electromagnetism. All known and widely accepted fundamental laws share striking properties of "simplicity, clarity, objectivity, and elegance" (pg 113). To the extent that one regards these features as universally descriptive of all of nature's fundamental aspects, one can judge the plausibility of new candidates for fundamental laws by evaluating them along these dimensions.

The general form of the critiques Rosenberg mounts against cognitive theories of experience, then, is that it is highly unlikely that their proposed base properties for experience-- e.g. complexity or functionality-- would feature in fundamental laws of nature. In particular, he argues that the nature of the concepts they employ, the simplicity of the laws featuring these concepts, and their possible empirical consequences are varyingly so complex, ambiguous, interest relative, and ad hoc as to be unlikely candidates for fundamental natural laws. By contrast, Rosenberg regards the theory of consciousness he will develop in the second half of the book to meet the criteria of simplicity, clarity, objectivity, and elegenace, even though it has panexperientialist consequences. Rejecting this theory in favor of a cognitive theory of experience just to avoid panexperientialist consequences, then, would be in violation of accepted standards of good theory construction. As such, Rosenberg's analysis here is in keeping with the Liberal Naturalist emphasis on methodological conservatism-- we should follow established standards of good theory construction, even if it leads us to become more ontologically radical than we might have liked.

 

[Canute]

I must say I struggle a bit with the idea making a clear distinction between cognition and consciousness. I see the reason for making that distinction, and in some contexts the sense of it, but still I wonder what an experience is in the absence of cognition.

To have an experience we must know that we are having it, and to me knowing implies cognition, (or perhaps 'proto-cognition'). This is not really an objection to Rosenberg's argument, more an uncertainty on my part, since pan-experientialism is still coherent (I think) even if one accepts what might be called 'pan-proto-cognition'.

I suppose then the question becomes how minimal a cognitive process has to be before it ceases to be a cognitive process and leaves just a non-cognitive experience. This suggests a sorites/boundary type problem as to how, where and whether a line can drawn with non-cognitive experience on one side and cognitive experience on the other. I'm unsure at the moment whether such a line can be drawn without raising some difficult questions.

One that comes to mind is that of self-awareness. Is it necessary to be aware of 'self' in order to be have an experience? I'd say not at an absolute level. But at the level of discrete experiencing objects and entities, where the experience is being had by an individual part of the whole, it seems to me that some awareness of a discrete self is required. Can one have an awareness of self without cognition? Can one have experiences as a discrete and individual entity in the absence of a sense of being a discrete and individual entity? Does pan-experientialism become incoherent if one allows that some minimal cognitive process is associated with the having of experiences?

One more - is pan-experientialism equivalent to microphenominalism?

 

[Tournesol]

I must say I struggle a bit with the idea making a clear distinction between cognition and consciousness. I see the reason for making that distinction, and in some contexts the sense of it, but still I wonder what an experience is in the absence of cognition.

I don't think having a quale in the absence of experience is much of a problem. In the old days, people used to think that colours-as-we-see
them were no different from the actual properties of objects; thus
(in terms they would not have used), the surface of an object
was coated with colour-qualia, and if no-one looked at it they
went unexperienced.

Can you have experience without qualia ? Well yes, thought of an abstract kind is generally held to be without phenomenal content.

Can you have experience without cognition ? Well, animals
infants and sleepwalkers presumably do

 

[Steve Esser]

One that comes to mind is that of self-awareness. Is it necessary to be aware of 'self' in order to be have an experience? I'd say not at an absolute level. But at the level of discrete experiencing objects and entities, where the experience is being had by an individual part of the whole, it seems to me that some awareness of a discrete self is required. Can one have an awareness of self without cognition? Can one have experiences as a discrete and individual entity in the absence of a sense of being a discrete and individual entity? Does pan-experientialism become incoherent if one allows that some minimal cognitive process is associated with the having of experiences?

Here, I would say one can have experience without being reflectively or introspectively self-aware. In fact, I would argue we are in a pre-reflective mode of self-consciousness much of the time. Our capacity for reflection is part of the distinctively human cognitive package, but experience can take place without it, and it is still experience "for-me" (pre-reflective self-conscious). I think you can likewise picture stripping away other features of cognition and still have experience.

Can you have experience without cognition ? Well, animals infants, and sleepwalkers presumably do

I agree. Thinking about the likely nature of first-person experiences had by progressively "lower" animals also helps us conceptually separate experience from cognition.

 

[hypnagogue]

I must say I struggle a bit with the idea making a clear distinction between cognition and consciousness. I see the reason for making that distinction, and in some contexts the sense of it, but still I wonder what an experience is in the absence of cognition.

See the section of chapter 5 titled "Is Panexperientialism Coherent?," particularly the third observation. Basically, what experience without cognition comes down to is the claim that even for some non-cognitive systems, there is something it is like to be such systems. Rosenberg does not claim that we can directly conceive of what an experience in the absence of cognition would be like, but we can understand what such a thing would be in an abstract way, by way of analogy.

To have an experience we must know that we are having it, and to me knowing implies cognition, (or perhaps 'proto-cognition'). This is not really an objection to Rosenberg's argument, more an uncertainty on my part, since pan-experientialism is still coherent (I think) even if one accepts what might be called 'pan-proto-cognition'.

The particular kind of knowing that is about experience ('knowing what it is like') is a deep and problematic issue, and should not be confused with knowing what or knowing how. It's not clear that knowing what it is like must involve a dedicated cognitive mechanism. This point is tied deeply into the roots of the Hard Problem, and perhaps also Searle's Chinese Room, and probably extent philosophical problems of epistemology in general.

I suppose then the question becomes how minimal a cognitive process has to be before it ceases to be a cognitive process and leaves just a non-cognitive experience. This suggests a sorites/boundary type problem as to how, where and whether a line can drawn with non-cognitive experience on one side and cognitive experience on the other. I'm unsure at the moment whether such a line can be drawn without raising some difficult questions.

You're right that this is an ambiguous question, but it seems to attain most of its ambiguity from the rather ambiguous word 'cognition.' It is difficult to draw a sharp line between cognitive and non-cognitive systems, period, even without considering experience.

One that comes to mind is that of self-awareness. Is it necessary to be aware of 'self' in order to be have an experience? I'd say not at an absolute level. But at the level of discrete experiencing objects and entities, where the experience is being had by an individual part of the whole, it seems to me that some awareness of a discrete self is required. Can one have an awareness of self without cognition? Can one have experiences as a discrete and individual entity in the absence of a sense of being a discrete and individual entity? Does pan-experientialism become incoherent if one allows that some minimal cognitive process is associated with the having of experiences?

It really depends on what we mean by self-aware. Some cases of unusual brain function lead people to report that their sense of self-awareness becomes dramatically altered or even outright abolished. But this sense of the term usually picks out a concept of self as a construct of personality, memories, and so on that is taken to be distinct from what is perceived as an external environment. Let's call this 'the cognitive self.' The cognitive self should not be confused with the discrete, bounded unit of experience that Rosenberg draws attention to in chapter 4 and names 'qualitative field' in chapter 5. Let's call this 'the phenomenal self.' Even in cases where people report that their cognitive sense of self has been changed drastically or vanished altogether, it is still (almost certainly) the case that their qualitative field, their phenomenal self, is bounded and unified in the senses of the words used in chapter 4. From this, it would follow that one can have experiences as a discrete and individual entity in the absence of a sense of being a discrete and individual entity. Of course, only a cognitive system could have a cognitive awareness of self.

One more - is pan-experientialism equivalent to microphenominalism?

Define microphenomenalism.

 

[hypnagogue]

I don't think having a quale in the absence of experience is much of a problem.

What would it mean to 'have' a quale in the absence of experience? If there is no experience, then by definition there cannot be any qualia.

Can you have experience without cognition ? Well, animals
infants and sleepwalkers presumably do

Whether or not we presume that (advanced) animals, infants, and sleepwalkers experience, we certainly cannot deny that they are cognitive systems on any reasonable definition of 'cognitive' (albeit perhaps not as cognitively advanced, and in general not as 'aware,' as a healthy, waking adult human). So these would not qualify as examples of experience without cognition.

 

[Tournesol]

What would it mean to 'have' a quale in the absence of experience? If there is no experience, then by definition there cannot be any qualia.

I sketched out an argument in my original comment:-

"I don't think having a quale in the absence of experience is much of a problem. In the old days, people used to think that colours-as-we-see
them were no different from the actual properties of objects; thus
(in terms they would not have used), the surface of an object
was coated with colour-qualia, and if no-one looked at it they
went unexperienced."

IOW:-

Qualia are the ways things seem to us.
Naive realism is logically posible
Naive realism holds that the way things seem to us is just the way they
are.
Thus, NR holds that qualia are just the way things are -- actual properties
of objects.
By most accounts, objects can have properties without anyone experienceing them.
According to NR, objects can have qualia without anyone experienceing them.
Thus, it is logically possible for unexperienced qualia to exist.

 

[Canute]

See the section of chapter 5 titled "Is Panexperientialism Coherent?," particularly the third observation. Basically, what experience without cognition comes down to is the claim that even for some non-cognitive systems, there is something it is like to be such systems. Rosenberg does not claim that we can directly conceive of what an experience in the absence of cognition would be like, but we can understand what such a thing would be in an abstract way, by way of analogy.

I see what GR is suggesting, and see that we can conjecture the existence of non-cognitive experiences.

However, I still wonder if the idea is coherent as put by GR. It seems to me that it may be an oversimplification. If experiences are caused by factors external to an experiencing entity then there has to be some sort of mediating mechanism between the cause and the effect, between the event in the external world and the experience in consciousness. For a human being this is their bodily senses and their cognitive processes. But could an experience be caused by something external to an entity if that entity has no senses and no cognitive processes? To me it seems that non-cognitive experiences cannot be caused by events in the world external to the entity, but must be experiential states that are unaffected by external events. Would you agree with this so far?

The particular kind of knowing that is about experience ('knowing what it is like') is a deep and problematic issue, and should not be confused with knowing what or knowing how. It's not clear that knowing what it is like must involve a dedicated cognitive mechanism. This point is tied deeply into the roots of the Hard Problem, and perhaps also Searle's Chinese Room, and probably extent philosophical problems of epistemology in general.

I suppose the question is one of whether it is possible to have an experience and not know that one is having it. Or, perhaps, of whether an experience can be had in such a way that there is no experience of an experiencer present, just the pure experience. I think that this is possible, but not in the case of experiences that are discrete and bounded.

I agree that it is not clear that knowing 'what it is like' involves a cognitive mechanism, but the reason for this is that it is not clear that it does not involve one.

You're right that this is an ambiguous question, but it seems to attain most of its ambiguity from the rather ambiguous word 'cognition.' It is difficult to draw a sharp line between cognitive and non-cognitive systems, period, even without considering experience.

I very much agree. This is the boundary problem I mentioned. My philosophy dictionary states that 'cognition' means 'knowledge'. No mention of cognitive processes. Can one have an experience but have no knowledge of it?

It really depends on what we mean by self-aware. Some cases of unusual brain function lead people to report that their sense of self-awareness becomes dramatically altered or even outright abolished. But this sense of the term usually picks out a concept of self as a construct of personality, memories, and so on that is taken to be distinct from what is perceived as an external environment. Let's call this 'the cognitive self.' The cognitive self should not be confused with the discrete, bounded unit of experience that Rosenberg draws attention to in chapter 4 and names 'qualitative field' in chapter 5. Let's call this 'the phenomenal self.'

I have no problem with the idea that the self is a personal cognitive construct with no (ontological) basis, and that a state of 'non-self' can be experienced. However I do have a problem with the idea that such a state is a discrete, bounded unit of experience distinct from the discrete, bounded experience of other entities in the same state, since this seems both logically contradictory and inconsistent with people's experience.

Even in cases where people report that their cognitive sense of self has been changed drastically or vanished altogether, it is still (almost certainly) the case that their qualitative field, their phenomenal self, is bounded and unified in the senses of the words used in chapter 4. From this, it would follow that one can have experiences as a discrete and individual entity in the absence of a sense of being a discrete and individual entity. Of course, only a cognitive system could have a cognitive awareness of self.

I'm not sure you're right about this. People who claim to have experiences of 'non-self' generally also claim that in this state the separation between entities dissolves and that one not only ceases to have a sense of being a discrete entity, but that one actually ceases to be one. In other words, while it seems true that to have an experience one does not have to be a discrete and individual entity it may be that to have discrete and bounded experiences one does have to be one. To me it would seem contradictory to say that one can have discrete and bounded experiences while experiencing not being a discrete and bounded entity. This would not imply that non-cognitive experiences are not possible but, rather, that in the absence of cognitive processes the experience would not be discrete or bounded.

I suppose in this self/non-self scenario conscious entities could be metaphorically likened to entities in an ideal condensate, in which entities are discrete and bounded individuals while disturbed but cease to be individuals when in a state of quantum coherence with the whole. In this metaphorical model a state of non-cognitive experience would be not just a state in which self does not exist experientially, but a state in which the entity cannot have discrete and bounded experiences. In this scenario it would be cognitive processes which creates self and the possibility of discrete experiences.

(Many people argue that it is cognition, defined as knowledge, that hides the unity of the universe from us by making us discrete experiencing entities - hence, for instance, the Garden of Eden/Tree of Knowledge allegory, or the meditative practice of transcending cognition to see 'what it is like' in its absence).

Define microphenomenalism.

That's the trouble. The term is used in the lit. but does not appear in any of my dictionaries. Basically it is the notion that experience/consciousness exists at the level of particles and waves, and perhaps even accounts for aspects of their behaviour. However I'm not clear on the details and I thought someone here might know them. I have an article arguing for it in a back issue of JCS somewhere, so I'll go and check it.

 

[hypnagogue]

Qualia are the ways things seem to us.
Naive realism is logically posible
Naive realism holds that the way things seem to us is just the way they
are.
Thus, NR holds that qualia are just the way things are -- actual properties
of objects.
By most accounts, objects can have properties without anyone experienceing them.
According to NR, objects can have qualia without anyone experienceing them.
Thus, it is logically possible for unexperienced qualia to exist.

The following is a definition of qualia from wikipedia:

Qualia (singular: "quale", pronounced KWAHL-ay) are most simply defined as the properties of sensory experiences by virtue of which there is something it is like to have them.

If there is no experience of qualia, there is no system for which there is something it is like to have them. If there is nothing it is like to have property X, then property X is not a quale. The criterion that qualia must be experienced is built into its own definition.

In any case, I don't see what relevance this issue has for this chapter. Panexperientialism does not claim that qualia can exist in the absence of experience, nor does it need to.

 

[hypnagogue]

If experiences are caused by factors external to an experiencing entity then there has to be some sort of mediating mechanism between the cause and the effect, between the event in the external world and the experience in consciousness. For a human being this is their bodily senses and their cognitive processes. But could an experience be caused by something external to an entity if that entity has no senses and no cognitive processes? To me it seems that non-cognitive experiences cannot be caused by events in the world external to the entity, but must be experiential states that are unaffected by external events. Would you agree with this so far?

I wouldn't agree with this so far. You seem to be roughly equating the roles of 'cognition' and 'causation,' which is the kind of overgeneralization of the meaning of 'cognition' that Rosenberg wants to avoid. Cognitive processes are best regarded as types of high-level causitive processes that tend to be directed towards certain high-level ends, such as seeking food or mating, and use sophisticated causal mechanisms to approach these ends. We can coherently deny that the sophisticated kinds of causation involved in cognition are necessary to support qualitative fields, while still believing that the nature of qualitative fields depends on more general kinds of causitive processes.

The most basic kind of statement we could make about the relationship between the brain and subjective experience is that certain physical/functional organizations of the brain correspond with the existence of a qualitative field, and certain changes in the physical or functional state of the brain correspond with certain changes in the nature of the qualitative field. A panexperientialist view need not abandon this basic observation of the link between physical processes and experience; it merely claims that the relevant kinds of physical/functional processes required to maintain and modulate the existence of qualitative fields are not limited to cognitive processes.

For example, suppose for the sake of argument that individual atoms correspond with the existence of individual qualitative fields, and further that the phenomenal content of a given atom's qualitative field covaries with the orbital configuration of its electrons. Clearly, atoms do not qualify as cognitive entities, so in this case, panexperientialism is true. Just as clearly, we can see that the atom's qualitative field is related to the physical/functional state of the atom in a way broadly analogous to the relationship between human p-consciousness and human brains. Even though the atom is not cognitive, its qualitative field is readily changed by external events; if a photon strikes it and excites one of its electrons into a higher orbital shell, its qualitative field will change accordingly.

I suppose the question is one of whether it is possible to have an experience and not know that one is having it. Or, perhaps, of whether an experience can be had in such a way that there is no experience of an experiencer present, just the pure experience. I think that this is possible, but not in the case of experiences that are discrete and bounded.

Recall that 'discrete' and 'bounded' here just mean that there are boundaries between qualitative fields; for example, mine does not overlap with yours-- I never glimpse a quale that is being experienced by you. Even if I suddenly find that my cognitive sense of self has completely dissolved and I no longer have any distinct notions of 'me, in here' or 'the environment, out there,' it will still be the case that my qualitative field is bounded from yours in the sense mentioned above. It will (almost certainly) not be the case that if you suddenly come to see a stop sign on the street, I will likewise see your phenomenal stop sign.

I very much agree. This is the boundary problem I mentioned. My philosophy dictionary states that 'cognition' means 'knowledge'. No mention of cognitive processes. Can one have an experience but have no knowledge of it?

Again, the knowledge issue is difficult. Rosenberg has a discussion much later on in the book where he addresses the knowledge paradox-- how can we come to talk about our experiences if they do not have a 'direct' causal effect on the brain-- in the context of his fully fleshed out theory. He adds knowledge of the type "knowing what" (eg knowing what it is like to see a sunset) to the already widely recognized "knowing that" (eg knowing that Abraham Lincoln was an American president) and "knowing how" (eg knowing how to ride a bike). [Note: in my previous response, I meant to say "knowing that" where I wrote "knowing what."]

If we are to take the Hard Problem seriously, we must recognize that 'knowing what' is of a nature in some important and non-obvious way distinct from 'knowing that' and 'knowing how.' Understanding the latter two forms of knowledge is largely a matter of working out the easy problems, whereas understanding the former requires us to address the hard problem head on. To me, this dissociation of the natures of these types of knowledge is at least suggestive that there is room for some sort of fundamental separation between them.

I think Rosenberg would say 'knowing what' and 'knowing how' are essential for establishing the particular nature of human consciousness, where experience is heavily embedded in a cognitive context. However, he would claim that these high-level cognitive mechanisms are not necessary for a much more primitive kind of 'knowing what' that would be exemplified by simple, non-cognitive experiencing subjects.

I'm running short on time right now, but I'll address the rest of your post later.

 

[Canute]

I don't feel that this addresses the issues I raised. The atom example ignores the problem of how the change on electron states becomes a change in conscious state (of the nucleus?). A mediating process is required, and it seems to me that this could be considered a cognitive process.

By 'proto-cognition' I meant just 'knowing what'. This seems to be the minimal condition for consciousness. However what you say suggests that there is a level beyond this - the possibility of experience in the absence of 'knowing what'. Is that what you are saying?

If my dictionary is correct than 'knowing what' is cognition. Perhaps one could say 'cognising what' as well as 'knowing what'. This is why I have difficulty separating cognition from consciousness. To a degree this is probably just a matter of how we define the words, but I'm not sure that it is entirely a problem of defintions.

Recall that 'discrete' and 'bounded' here just mean that there are boundaries between qualitative fields; for example, mine does not overlap with yours-- I never glimpse a quale that is being experienced by you.

Here I think you miss my point. I was suggesting that consciousness in the absence of cognitive processes (i.e. in the absence of self) is one in which those boundaries dissolve. I'm aware that you and GR don't agree, but what you say gives me no reason to change my view.

The condensate metaphor I gave is clumsy but it is useful I think since it illustrates how things that are discrete can cease to be so. As I understand it in a sense the condensate consists of a number of discrete atoms but, in a coherent state, there is a sense in which all the atoms are the same atom. I believe that cosmologists sometimes liken the early universe to an ideal gas, so the metaphor is not entirely plucked from thin air.

 

[hypnagogue]

I don't feel that this addresses the issues I raised. The atom example ignores the problem of how the change on electron states becomes a change in conscious state (of the nucleus?).

By hypothesis, there would be some fundamental law that states that if the electrons of the atom change their configuration in some way, then the experience of the atom changes. Fundamental laws will always have an air of unexplained brutishness to them, since we tend to explain things reductively, and fundamental laws cannot be reduced to further concepts (or, if they can, then they can no longer be called fundmental).

Actually, Rosenberg's framework removes quite a bit of the mystery of the systematic covariance of physical phenomena and subjective experience. We might as well wait for the details, but in broad strokes, the physical is taken to be the relational aspect of the experiential, so it becomes a rather trivial observation that a change in p-consciousness must always be accompanied by a change in physical/functional state.

A mediating process is required, and it seems to me that this could be considered a cognitive process.

Why is a mediating process required? We can add all the mediating processes we want, and we'll still be left with the fundamental question of why all these processes are accompanied by experience. At some point, we'll just have to assert that there are fundamental laws in nature that guarantee that this is the case. If so, then it becomes natural to think that these fundamental laws could apply to many types of causal systems, and (as this chapter argues) it becomes exceedingly difficult to construct the fundamental laws in just such a way that they only grant experience to cognitive systems.

By 'proto-cognition' I meant just 'knowing what'. This seems to be the minimal condition for consciousness. However what you say suggests that there is a level beyond this - the possibility of experience in the absence of 'knowing what'. Is that what you are saying?

If my dictionary is correct than 'knowing what' is cognition. Perhaps one could say 'cognising what' as well as 'knowing what'. This is why I have difficulty separating cognition from consciousness. To a degree this is probably just a matter of how we define the words, but I'm not sure that it is entirely a problem of defintions.

Well, we don't have a sharp definition of what cognition is, but from the treatment Rosenberg gives here, it's clear he considers it to be a sophisticated, high level phenomenon in some sense. If we are to honor this sense of the word, I don't know that a term like 'proto-cognition' makes much sense here. If anything, it would seem that proto-cognition would just be another way to refer to virtually any type of non-cognitive causal process, and of course causation will play a major role in Rosenberg's theory.

As for 'knowing what,' if we could attribute such a thing to a non-cognitive system, it would not be exceedingly similar to the 'knowing what' of humans. Our 'knowing what' is intimately tied into our 'knowing that' and 'knowing how,' which are strictly subserved by high-level cognitive processes. A non-cognitive, experiencing system could only be said to 'know' what it is like to be itself in the minimal sense that it would have some sort of phenomenal experience-- it would, by definition, be like something to be that system. This sense of the word 'know,' however, is a long ways off from the way we speak of ourselves as knowing things.

In fact, it would probably be best to say that a non-cognitive system does not 'know' anything at all, just to avoid ambiguity. For example, I would be much more comfortable saying that a non-cognitive experiencing system feels what it is like, rather than knows what it is like, to be itself. "Feels" is itself somewhat ambiguous and problematic here, but it's not as loaded of a word as "knows."

Here I think you miss my point. I was suggesting that consciousness in the absence of cognitive processes (i.e. in the absence of self) is one in which those boundaries dissolve.

Well, it's certainly possible that the boundaries of experience dissolve in the absence of a cognitive context, or at least we have no way to empirically rule it out. One of the lessons of this chapter is that we have to judge theories of consciousness largely on their simplicity, objectivity, fruitfulness, and so on, as direct empirical falsification of any theory involving subjective experience is going to be problematic.

As we'll see, Rosenberg hangs the boundary conditions of consciousness not on high-level cognitive processes, but rather on more general and fundamental causal processes. The result is that even non-cognitive experiencing systems have experiential boundaries. There is no known way to test this empirically, but Rosenberg's theory does have nice virtues of simplicity, objectivity, and fruitfulness for various applications, so we must take it seriously. Short of coming up with a way to measure subjective experience, your best bet here would seem to be to come up with an even better theory of consciousness which does happen to have the consequence that experience is only bounded for cognitive systems.

 

[hypnagogue]

Here I think you miss my point. I was suggesting that consciousness in the absence of cognitive processes (i.e. in the absence of self) is one in which those boundaries dissolve.

Well, it's certainly possible that the boundaries of experience dissolve in the absence of a cognitive context, or at least we have no way to empirically rule it out.

Just to be perfectly clear, my response here refers to supposed non-cognitive experiencing systems. A human whose cognitive sense of self has dissolved for whatever reason, in such a way that he could arguably be said to no longer have a 'self' for some period of time, does not qualify here. There is much more to cognition than constructing a self, and of course even if a human loses her cognitive sense of self, her brain is certainly still actively cognizing insofar as it is processing sensory inputs and guiding action on that basis, in some convoluted form or another.

 

[Canute]

Hypnagogue

Don't feel obliged to respond to these comments unless you want to. I don't what to bog the thread down in details and I've been longwinded as usual. I just want to suggest that the logic of Rosenbergs argument is not inevitable, rather than attempt to show that he is definitely wrong.

By hypothesis, there would be some fundamental law that states that if the electrons of the atom change their configuration in some way, then the experience of the atom changes. Fundamental laws will always have an air of unexplained brutishness to them, since we tend to explain things reductively, and fundamental laws cannot be reduced to further concepts (or, if they can, then they can no longer be called fundmental).

This seems to be a common approach in physics. We say, when we run out of explanatory reach, that there is a fundamental law at work, as if 'laws' had an objective existence. But what we call laws are patterns of caused behaviour which occur because of the intrinsic nature of the things that are doing the behaving, and not behaviour that arises because these things are obeying some sort of cosmic instruction manual which they have to reference before doing anything.

However, I see your point I think. There is a limit to how much we can understand about the behaviour of things by studying only their behaviour. If Rosenberg is to remain scientific in his his approach then he must respect this limit. It is, after all, the basis of his argument against physicalism that scientific explantions of behaviour do not explain what underlies that behaviour. However this appeal to a fundamental law is, in the end, an appeal to ignorance, even if we consider it inevitable that we make this appeal. If electron states cause the conscious states of an atom then there must be a specific and concrete causal mechanism at work that links cause with effect. (Unless electron states are experiences, in which case brain is mind).

Actually, Rosenberg's framework removes quite a bit of the mystery of the systematic covariance of physical phenomena and subjective experience. We might as well wait for the details, but in broad strokes, the physical is taken to be the relational aspect of the experiential, so it becomes a rather trivial observation that a change in p-consciousness must always be accompanied by a change in physical/functional state.

Ok, and by implication there must be a change in p-consciousness for every change in physical/functional state, hence pan-experientialism. This is what interests me about the book. Rosenberg flirts with the Buddhist/Taoist explanation of consciousness/reality all the way through it and I find the parallels very interesting. Seeing the physical as the relational aspect of the experiential is one close parallel. Unfortunately he does not seem to know this other view, and in my (perhaps non-typical) opinion it is not acceptable in this day and age for someone writing about consciousness rather than brains not to know it. If he is to show that his hypothesis/theory is correct then at some point he has to show that competing explanations are false. He has done this for physicalism but so far everything he has said has been either consistent with the truth of this other explantion or, as I've been trying to suggest, an assumption.

Why is a mediating process required? We can add all the mediating processes we want, and we'll still be left with the fundamental question of why all these processes are accompanied by experience. At some point, we'll just have to assert that there are fundamental laws in nature that guarantee that this is the case. If so, then it becomes natural to think that these fundamental laws could apply to many types of causal systems, and (as this chapter argues) it becomes exceedingly difficult to construct the fundamental laws in just such a way that they only grant experience to cognitive systems.

This makes sense. But 'fundamental law' here is just another way of saying 'don't know', and not equivalent to saying that there is no causal mechanism. Also, saying that we cannot answer the question of what fundamental mechanism is at work between brain and mind (or electron-state and conscious-state) is not the same as saying that there isn't one.

Well, we don't have a sharp definition of what cognition is, but from the treatment Rosenberg gives here, it's clear he considers it to be a sophisticated, high level phenomenon in some sense. If we are to honor this sense of the word, I don't know that a term like 'proto-cognition' makes much sense here. If anything, it would seem that proto-cognition would just be another way to refer to virtually any type of non-cognitive causal process, and of course causation will play a major role in Rosenberg's theory.

That's fine. We should take the term cognition to mean what Rosenberg wants it to in this context, it's his book, so perhaps 'proto-cognition' is an innapropriate term. However, whatever terms one uses in the end the question of where one draws a line between cognition and non-cognition remains. It seems to me another loose end to be tidied up at some point. The same arguments have arisen in consciousness studies in relation to consciousness and proto-consciousness, and they have not been resolved.

As for 'knowing what,' if we could attribute such a thing to a non-cognitive system, it would not be exceedingly similar to the 'knowing what' of humans. Our 'knowing what' is intimately tied into our 'knowing that' and 'knowing how,' which are strictly subserved by high-level cognitive processes. A non-cognitive, experiencing system could only be said to 'know' what it is like to be itself in the minimal sense that it would have some sort of phenomenal experience-- it would, by definition, be like something to be that system. This sense of the word 'know,' however, is a long ways off from the way we speak of ourselves as knowing things.

True, it is very different in a quantative sense. But is it different in a qualitative sense? This is the question that seems to be left unanswered. Certainly non-cognitive experience must be unlike day to day human experience. But the fact that a wavicle is a long way off from what we speak of as a kitchen table doesn't make them ontologically distinct phenomena, and there remains the ambiguity of the term 'cognitive'.

In fact, it would probably be best to say that a non-cognitive system does not 'know' anything at all, just to avoid ambiguity. For example, I would be much more comfortable saying that a non-cognitive experiencing system feels what it is like, rather than knows what it is like, to be itself. "Feels" is itself somewhat ambiguous and problematic here, but it's not as loaded of a word as "knows."

This seems a good compromise, and I'd go along with it for the purposes of discussing the book. However it only postpones the problem, for can one feel if one is not cognisant that one is feeling? I think the answer may be yes, depending on the definition of terms, but what I think is not the point here. I'm not trying to argue that Rosenberg is wrong, just suggesting that this is an issue which, as a reader, I would like to have seen dealt with before we move on as if it were not an issue. It has to be addressed at some point.

Well, it's certainly possible that the boundaries of experience dissolve in the absence of a cognitive context, or at least we have no way to empirically rule it out. One of the lessons of this chapter is that we have to judge theories of consciousness largely on their simplicity, objectivity, fruitfulness, and so on, as direct empirical falsification of any theory involving subjective experience is going to be problematic.

Is this not a misuse of 'empirical'? It is perfectly possible to empirically determine whether the boundaries of our individual selves break down at a fundamental level of consciousness, and there is no problem at all, in principle at least, with the direct empirical falsification of theories of subjective experience. It is just not possible to do it by entirely scientific means.

As we'll see, Rosenberg hangs the boundary conditions of consciousness not on high-level cognitive processes, but rather on more general and fundamental causal processes. The result is that even non-cognitive experiencing systems have experiential boundaries. There is no known way to test this empirically,

I don't know why you say this. Of course this can be tested empirically. Do you think all the people who disagree with this view are just guessing? When Spencer-Brown states that the boundary between oneself and the rest of the universe is illusory it is because he has gone and looked at the evidence, not because it is a part of some conjectured hypothesis or theory. You could test it empirically yourself if you wanted to. You may not believe this, but you can't just state that it not possible to do this without some sort of argument to back up your statement. In a strong sense our conscious states are the only things that we can test empirically.

but Rosenberg's theory does have nice virtues of simplicity, objectivity, and fruitfulness for various applications, so we must take it seriously.

I feel 'theory' is too strong a term, since it leaves too many questions unanswered. Neither do I find it simple or objective as a hypothesis. However I do feel it is a major advance in that it is so much more sophisticted and subtle than the usual scientific attempts, and I am treating it seriously. I just think that as yet it is underdeveloped and, at the risk of sounding arrogant, metaphysically naive.

Short of coming up with a way to measure subjective experience, your best bet here would seem to be to come up with an even better theory of consciousness which does happen to have the consequence that experience is only bounded for cognitive systems.

Something like Buddhism perhaps, or Sufism, Taoism, Advaita Vedanta, Essenism, Theosophy, Brown's 'mathematical Taoism', or Scroedingers Upanishad based version, and various other versions of the same explanation. In this explanations cognition (in the sense you mean it) brings with it boundedness, but this boundedness, or division of the whole into individual experiencers, is ultimately illusory. If people dismiss this explanation as mystical, unscientific, religious, deluded or complete nonsense then this is up to them. But this explanation cannot be simply dismissed, for according to an increasing amount of scientific evidence it might be correct.

Of course there are some things that cannot be explained. These doctrines or teachings do not explain the relationship between brain and mind in a scientifically useful way. However the means of finding out the nature of that relationship is explained, and an understanding of this view, with or without a belief in or knowledge of its truth, would at least provide a valuable starting point for the development of complementary scientific explanation, one which, for a change, would not become inevitably enmired in undecidable metaphysical questions.

Excuse me if I'm a bit tetchy on this issue, but I find it increasingly daft that this explanation is so thoroughly ignored by institutional science, especially since it is so uncannily consistent with current scientific thinking on the nature of the consciousness and the physical universe, and especially since it has been considered correct by a good number of respectable academics and physicists, never mind all the others. Even without personal experiential practice it's no harder to understand the structure and underlying principles of this explanation than it is the mathematics of quantum mechanics, and in fact the epistemilogical systems of Taoism and QM are very similar in a number of important ways. Yet here we are in 2005 discussing a new book on consciousness that does not even mention its existence. GR does not even know Spencer-Brown's work, which is directly relevant to the issues in the earlier chapters. To me it feels a bit like reading a new book on the origin of species that contradicts Darwinian evolutionary theory but does not mention it.

PS. So far in these discussions you've taken the role of 'defender of the book', which is very useful to the discussion and very helpful to my understanding of it. However I'd be genuinely interested to know what your own opinion is on some of these issues. Are you role-playing when always defending GR, or do you actually have no reservations yourself regarding his ideas?

 

[antfm]

I think Rosenberg would say 'knowing that' and 'knowing how' are essential for establishing the particular nature of human consciousness, where experience is heavily embedded in a cognitive context. However, he would claim that these high-level cognitive mechanisms are not necessary for a much more primitive kind of 'knowing what' that would be exemplified by simple, non-cognitive experiencing subjects.

hypnagogue,
When you say this I understand that by 'knowing what' you are pointing to that more basic raw 'qualitative field' introduced by Rosenberg, though still not developed. Correct me if wrong.
However, I think that the boundaries between supossedly cognitive or supposedly non-cognitive systems are becoming a problem in having a better grasp of the idea.

Whether or not we presume that (advanced) animals, infants, and sleepwalkers experience, we certainly cannot deny that they are cognitive systems on any reasonable definition of 'cognitive' (albeit perhaps not as cognitively advanced, and in general not as 'aware,' as a healthy, waking adult human). So these would not qualify as examples of experience without cognition.

If 'cognition' is more related by definition to 'knowledge' I find it difficult to assign it to the cases there mentioned (animals, infants, etc). It seems to me that cognition is more like a 'work in progress', built up or evolved for a system with certain potentiallity for it (animal, infant, etc), than a capacity available from the beginning. That's why I also feel closer to the idea of considering those systems as possible examples of experience without cognition than you seem to admit. Or as it is expressed in other post:

Thinking about the likely nature of first-person experiences had by progressively "lower" animals also helps us conceptually separate experience from cognition.

Well, I think that even if there is not total agreement about the terms employed, it would be helpful to know in which more aproximate sense Rosenberg is using them. But this is my personal view, and perhaps he doesn't consider it necessary in his exposition.

 

[Steve Esser]

Hello to Canute: I’m enjoying your analysis of the book and have found your exchanges with Hypnogogue to be thought-provoking. But one aspect is troubling me a bit.

Now, I know little about Buddhism, Taoism, etc, so I’m just as guilty as Rosenberg of ignoring this alternative approach you repeatedly speak of. So I must ask if you are willing to explain further this competing idea -- which asserts that the boundaries between natural systems are illusory (maybe you’ve done this in another thread?). How does such an approach help us explain nature? I understand that it addresses the hard problem by proposing an underlying monism of consciousness, like idealism (is this right?). But to say our investigations of the natural world are just revealing the details of an elaborate illusion is exactly as satisfying to me as a scientific materialist view which says first-person experience must be an illusion. A naturalistic-panexperientialist approach such as featured in Rosenberg’s book has the advantage of taking both what we call the physical world and experience equally seriously.

 

[Tournesol]

BTW, how does a free-standing illusion work ? Doesn't an illusion require someone to
be fooled ?

 

[hypnagogue]

If electron states cause the conscious states of an atom then there must be a specific and concrete causal mechanism at work that links cause with effect. (Unless electron states are experiences, in which case brain is mind).

A lot of Rosenberg's work is aimed towards overthrowing this notion of the physical brain as causing conscious experience in the way that (say) a cue ball striking another billiard ball causes it to move towards the side pocket. He suggests that we should not be looking for ways to show how consciousness arises from the brain, but rather that we should be looking for ways to get under the physics, somehow. This is the sort of thing his arguments will eventually build up to. It's actually much closer to the sort of identity relation you mention here-- brain 'is' mind-- but requires us to reconceptualize what the brain is (and more generally, what all physical phenomena are).

Unfortunately he does not seem to know this other view, and in my (perhaps non-typical) opinion it is not acceptable in this day and age for someone writing about consciousness rather than brains not to know it. If he is to show that his hypothesis/theory is correct then at some point he has to show that competing explanations are false. He has done this for physicalism but so far everything he has said has been either consistent with the truth of this other explantion or, as I've been trying to suggest, an assumption.

OK, duly noted; between this and previous discussions, I think we've firmly established your viewpoint that Rosenberg does not give enough credence to the Eastern/Buddhist/mystic view, or whatever you might like to call it. At this point, perhaps it is better to relax this criticism for now and re-assess it upon having completed the book (more on this below).

That's fine. We should take the term cognition to mean what Rosenberg wants it to in this context, it's his book, so perhaps 'proto-cognition' is an innapropriate term. However, whatever terms one uses in the end the question of where one draws a line between cognition and non-cognition remains. It seems to me another loose end to be tidied up at some point. The same arguments have arisen in consciousness studies in relation to consciousness and proto-consciousness, and they have not been resolved.

I don't see the issue of cognition being a big problem for Rosenberg's theories going forward. A more precise definition of cognition would be crucial, of course, for any theory that dictates that only cognitive systems can be conscious. But Rosenberg's is not such a theory, and so the term 'cognition' per se will not play a huge role in the proceedings. We will be much more concerned about causation in general.

This seems a good compromise, and I'd go along with it for the purposes of discussing the book. However it only postpones the problem, for can one feel if one is not cognisant that one is feeling? I think the answer may be yes, depending on the definition of terms, but what I think is not the point here. I'm not trying to argue that Rosenberg is wrong, just suggesting that this is an issue which, as a reader, I would like to have seen dealt with before we move on as if it were not an issue. It has to be addressed at some point.

IIRC, Rosenberg addresses this question directly later on in the book. But the answer to your question here is really already given at this point, albeit indirectly. The argument runs like this:

Rosenberg thinks it is at least logically coherent that there could be feeling without cognition of feeling (ch. 5). He furthermore does not think it is likely that cognition is necessary for experiential feeling, based on arguments made in chapters 2 and 6. If cognition is necessary for feeling, then some cognitive theory of experience is true; but if physicalism is false (ch. 2), then cognitive theories of experience will be extremely complex, or interest relative, etc. (ch. 6). So if one believes that the fundamental law of nature are simple, objective, elegant, etc., and one believes that physicalism is false, then one has strong reason to believe that all cognitive theories of experience are false. Accordingly, one would have strong reason to believe that cognition of feeling is not necessary for feeling.

Is this not a misuse of 'empirical'? It is perfectly possible to empirically determine whether the boundaries of our individual selves break down at a fundamental level of consciousness, and there is no problem at all, in principle at least, with the direct empirical falsification of theories of subjective experience. It is just not possible to do it by entirely scientific means.

No, I stand by what I said there. What I said was that we have no way to empirically rule out the possibility that "the boundaries of experience dissolve in the absence of a cognitive context." In fact, we have no way to empirically rule out any claim about experience in the absence of a cognitive context. The reason for this is very simply that even if we allow first person experiences to count as empirical 'data' in some sense (which I am in favor of), all such data will still necessarily take place in a cognitive context. The experiencers will be cognitively organized human brains doing distinctly cognitive things, even in circumstances of unusual brain activity. Meditation, for example, has the effect of producing unusual and interesting types of first person experiences and brain activities, but this is still a cognitive process. The meditator focuses his attention in some manner or another, discriminates different types of experiences, stores these in memory, then accesses his memory to give us a verbal report.

I feel 'theory' is too strong a term, since it leaves too many questions unanswered. Neither do I find it simple or objective as a hypothesis.

Well, we haven't even got up to the real 'theory' part of the book yet. When we've completed that part, we can evaluate the theory's simplicity, objectivity, etc., and what questions it leaves unanswered.

Something like Buddhism perhaps, or Sufism, Taoism, Advaita Vedanta, Essenism, Theosophy, Brown's 'mathematical Taoism', or Scroedingers Upanishad based version, and various other versions of the same explanation. In this explanations cognition (in the sense you mean it) brings with it boundedness, but this boundedness, or division of the whole into individual experiencers, is ultimately illusory.

OK, let's leave this as an exercise for after we have completed the book. We'll take stock of how Rosenberg's theory allows us to answer various questions about p-consciousness and other phenomena, and see where it leaves some questions unanswered; we'll compare this to the types of answers and overall explanitory power given by one or more of the paradigms you mention here. Would that be fair?

PS. So far in these discussions you've taken the role of 'defender of the book', which is very useful to the discussion and very helpful to my understanding of it. However I'd be genuinely interested to know what your own opinion is on some of these issues. Are you role-playing when always defending GR, or do you actually have no reservations yourself regarding his ideas?

I find the reasoning in the book thus far to have been quite compelling, and am in agreement with it (although I may have had an ultimately insignificant quibble with the material here and there). I do have more substantive departures for a couple of issues discussed in the second half of the book (which I will voice), but even these are small matters in the larger scheme of the paradigm Rosenberg constructs there.

 

[Canute]

Now, I know little about Buddhism, Taoism, etc, so I’m just as guilty as Rosenberg of ignoring this alternative approach you repeatedly speak of. So I must ask if you are willing to explain further this competing idea -- which asserts that the boundaries between natural systems are illusory (maybe you’ve done this in another thread?). How does such an approach help us explain nature? I understand that it addresses the hard problem by proposing an underlying monism of consciousness, like idealism (is this right?). But to say our investigations of the natural world are just revealing the details of an elaborate illusion is exactly as satisfying to me as a scientific materialist view which says first-person experience must be an illusion. A naturalistic-panexperientialist approach such as featured in Rosenberg’s book has the advantage of taking both what we call the physical world and experience equally seriously.

I hadn't thought of that last point, and it's a good one. I suppose it must seem that I'm suggesting a solution that's equivalent to physicalism in that it leads to an explanatory dead-end. I'm not sure how to go about explaining that this is not the case but rather the opposite of the case. It would mean doing what you suggest, explaining 'this competing idea'. This doesn't seem the place the do that, but I'd have a go on another thread or wherever. I do not claim to be a skilled Buddhist by the way, far from it. But I've enough grasp of its conceptual scheme and central assertions to wonder why it isn't taken more seriously by science.

A lot of Rosenberg's work is aimed towards overthrowing this notion of the physical brain as causing conscious experience in the way that (say) a cue ball striking another billiard ball causes it to move towards the side pocket. He suggests that we should not be looking for ways to show how consciousness arises from the brain, but rather that we should be looking for ways to get under the physics, somehow. This is the sort of thing his arguments will eventually build up to. It's actually much closer to the sort of identity relation you mention here-- brain 'is' mind-- but requires us to reconceptualize what the brain is (and more generally, what all physical phenomena are).

Fine. Perhaps I've missed some of the subtlety.

OK, duly noted; between this and previous discussions, I think we've firmly established your viewpoint that Rosenberg does not give enough credence to the Eastern/Buddhist/mystic view, or whatever you might like to call it. At this point, perhaps it is better to relax this criticism for now and re-assess it upon having completed the book (more on this below).

Again fine. But note that I'm not pushing this other view like some Jehovah's Witness at the door. I'm suggesting that it's a view that needs to be considered and if possible refuted or otherwise, and indicating that where Rosenberg contradicts this other view he does so without evidence.

I don't see the issue of cognition being a big problem for Rosenberg's theories going forward. A more precise definition of cognition would be crucial, of course, for any theory that dictates that only cognitive systems can be conscious. But Rosenberg's is not such a theory, and so the term 'cognition' per se will not play a huge role in the proceedings. We will be much more concerned about causation in general.

It seems to me that Rosenberg's problem is that pan-experientialism suggests that very simple entities can be conscious, entities that we would expect to be too simple to support computation/cognition. This means decoupling experience from cognition. Not an easy thing to do. But I accept what you say here. (I don't see any of my objections so far as preventing Rosenberg's argument from proceeding, although they may be more telling later).

If cognition is necessary for feeling, then some cognitive theory of experience is true; but if physicalism is false (ch. 2), then cognitive theories of experience will be extremely complex, or interest relative, etc. (ch. 6). So if one believes that the fundamental law of nature are simple, objective, elegant, etc., and one believes that physicalism is false, then one has strong reason to believe that all cognitive theories of experience are false. Accordingly, one would have strong reason to believe that cognition of feeling is not necessary for feeling.

I'm afraid I can't follow that argument. I see your point about objectivity, but why must a non-physicalist theory of experience be inconsistent with the idea that the natural laws are fundamentally simple and elegant?

No, I stand by what I said there. What I said was that we have no way to empirically rule out the possibility that "the boundaries of experience dissolve in the absence of a cognitive context." In fact, we have no way to empirically rule out any claim about experience in the absence of a cognitive context. The reason for this is very simply that even if we allow first person experiences to count as empirical 'data' in some sense (which I am in favor of), all such data will still necessarily take place in a cognitive context. The experiencers will be cognitively organized human brains doing distinctly cognitive things, even in circumstances of unusual brain activity. Meditation, for example, has the effect of producing unusual and interesting types of first person experiences and brain activities, but this is still a cognitive process. The meditator focuses his attention in some manner or another, discriminates different types of experiences, stores these in memory, then accesses his memory to give us a verbal report.

I can see exactly why you say this but it embodies a misunderstanding of what meditation is about. That probably too big a topic to get into here, and it's probably not necessary to do so, but I stand by what I said about empirically verifying that the boundaries between self and other, subject and object, break down in the final analysis. It is possible to do it, and therefore it cannot be shown by any method that it is not possible. If someone does show that it is not possible then I will accept that I've lost my reason and go back on my medication.

OK, let's leave this as an exercise for after we have completed the book. We'll take stock of how Rosenberg's theory allows us to answer various questions about p-consciousness and other phenomena, and see where it leaves some questions unanswered; we'll compare this to the types of answers and overall explanitory power given by one or more of the paradigms you mention here. Would that be fair?

Absolutely.

I find the reasoning in the book thus far to have been quite compelling, and am in agreement with it (although I may have had an ultimately insignificant quibble with the material here and there). I do have more substantive departures for a couple of issues discussed in the second half of the book (which I will voice), but even these are small matters in the larger scheme of the paradigm Rosenberg constructs there.

Thanks.

 

[hypnagogue]

Again fine. But note that I'm not pushing this other view like some Jehovah's Witness at the door. I'm suggesting that it's a view that needs to be considered and if possible refuted or otherwise, and indicating that where Rosenberg contradicts this other view he does so without evidence.

Refutation and evidence are always going to be dodgey issues when it comes to theories of consciousness, due to privacy of experience. It seems as if our best route is to use the various virtues and features of different theories to play a more prominent role as to what counts as evidence, and accordingly what makes a theory more or less likely to be true. This is why I suggested that we wait to address this issue until after we've completed the book-- at that point, we can perform this task of comparing theoretical virtues and drawbacks and weight our confidence in the theories accordingly.

It seems to me that Rosenberg's problem is that pan-experientialism suggests that very simple entities can be conscious, entities that we would expect to be too simple to support computation/cognition. This means decoupling experience from cognition. Not an easy thing to do.

We should not equate the terms 'computation' and 'cognition,' or at least, not the sense of the word 'cognition' that Rosenberg is operating with is not equivalent to computation-- if anything, cognition should be seen as a certain subset of complex and sophisticated computations.

This point has direct bearing on your statement, "...entities that we would expect to be too simple to support computation/cognition." As cognition is a nebulous term, there will surely be many gray areas where it is not clear if a given system should count as cognitive or not. But for the simplest kinds of things in nature, such as an atom-- or even more fundamentally, a quark or a photon-- this is not a gray area at all. We do not have to say that we don't expect an atom not to be cognitive; we can plainly assert that it is not. That is to say, there is no reasonable construal of cognition as a complex, high level phenomenon that includes simple things such as atoms in the class of cognitive systems. On the other hand, there is no real controversy that atoms could support computation. For instance, in principle we could use the spin of atomic nuclei as individual computational elements, with spin up representing "1" and spin down "0" (or vice versa).

I'm afraid I can't follow that argument. I see your point about objectivity, but why must a non-physicalist theory of experience be inconsistent with the idea that the natural laws are fundamentally simple and elegant?

The claim is that cognitive theories of experience, in the context of a non-physicalist metaphysic, are incompatible with simple, objective, etc. fundamental laws. The motivation behind this claim is precisely what is explained in this chapter. The basic tension is that the various candidates for base properties of experience on cognitive theories of experience seem to require complex and/or interest relative descriptions in varying degrees, and in a non-physicalist metaphysic, these descriptions would have to feature in fundamental laws.

I can see exactly why you say this but it embodies a misunderstanding of what meditation is about.

Whatever meditation is about, it is an experience had by human beings. Us human beings are cognitive creatures, and our subjective experiences are intimately tied into the activity of our brains. That is to say, our experiences take place in a cognitive context. I do not think it is going out on a limb to claim that a meditator's experience could be altered or interrupted by proper disruption of his brain activity, or conversely that sufficiently sophisticated manipulation of brain activity could induce meditative experiences. Insofar as these claims are true, meditation, just like any other kind of subjective experience, is heavily embedded in a cognitive context. That the specific qualitative character of meditative experiences is drastically different than normal, everyday experiences does not contradict this point.

 

[Canute]

Refutation and evidence are always going to be dodgey issues when it comes to theories of consciousness, due to privacy of experience.

They are certainly going to be dodgy issues if one tries to refute the existence of certain first-person experiences with third person evidence. There's no way of doing it.

If we rely wholly on third-person evidence to develop our theories of consciousness then we inevitably find that we cannot provide any evidence that there is a need for such theories.

We should not equate the terms 'computation' and 'cognition,' or at least, not the sense of the word 'cognition' that Rosenberg is operating with is not equivalent to computation-- if anything, cognition should be seen as a certain subset of complex and sophisticated computations.

Is a sub-set of complex and sophisticated computations something quite different to a computation? They seem the same thing to me.

This point has direct bearing on your statement, "...entities that we would expect to be too simple to support computation/cognition." As cognition is a nebulous term, there will surely be many gray areas where it is not clear if a given system should count as cognitive or not.

Yes, this is the problem. Cognition is a nebulous term. So how are we to decide whether the idea of non-cognitive experiences makes sense? How can we say that cognition is not dependent on computation when cognition is a nebulous term? This is why I feel Rosenberg should have pinned it down more carefully.

But for the simplest kinds of things in nature, such as an atom-- or even more fundamentally, a quark or a photon-- this is not a gray area at all. We do not have to say that we don't expect an atom not to be cognitive; we can plainly assert that it is not.

Either cognition is a nebulous term or it is not. If it is then we cannot say that atoms are not cognitive, for we would not know what we mean by saying this. As you say, there is no reason in principle why atoms could not be cognitive, it all depends on how we define cognitive.

The claim is that cognitive theories of experience, in the context of a non-physicalist metaphysic, are incompatible with simple, objective, etc. fundamental laws. The motivation behind this claim is precisely what is explained in this chapter. The basic tension is that the various candidates for base properties of experience on cognitive theories of experience seem to require complex and/or interest relative descriptions in varying degrees, and in a non-physicalist metaphysic, these descriptions would have to feature in fundamental laws.

I'm afraid I still cannot grasp this point. I have no opinion of whether consciousness (or proto-consciousness) has to be cognitive or not. That would depend on what we mean by cognitive. However I still cannot see why a cognitive theory has to be more or less complex than a non-cognitive one. I think I'm missing something somewhere.

Whatever meditation is about, it is an experience had by human beings. Us human beings are cognitive creatures, and our subjective experiences are intimately tied into the activity of our brains. That is to say, our experiences take place in a cognitive context.

What do you mean by "cognitive" here? If Rosenberg is right about the plausibility of pan-experientialism then experiences are not entirely tied to the activity of brains and do not necessarily require a cognitive context in which to occur. If so then the question of whether human beings can have non-cognitive experiences is a practical and observational one, not one that can be answered by theorising or by third-person observation.

I do not think it is going out on a limb to claim that a meditator's experience could be altered or interrupted by proper disruption of his brain activity, or conversely that sufficiently sophisticated manipulation of brain activity could induce meditative experiences. Insofar as these claims are true, meditation, just like any other kind of subjective experience, is heavily embedded in a cognitive context. That the specific qualitative character of meditative experiences is drastically different than normal, everyday experiences does not contradict this point.

Such experiences are not simply non-ordinary. They are experiences that cannot be had cognitively (if we mean via mental activity/computational processes). I cannot prove this of course, but the following diverse quotes illustrate the unanimous opinion of meditators that at the limit experience transcends brain function and cognitive context. Rather, they suggest that once one realizes the nature of the 'ur-experience' that underlies human subjectivity then one realizes the pan-experientialist nature of existence. (By 'realises' I mean discovers that it has this nature as opposed to theorising that it has this nature). To me all these comments are about transcending computation and, by one definition, cognition.

I agree about leaving this issue until later so I'm posting these not so much as argument but rather just as an interesting similarity between the pan-experientialist meta-cognitive experience posited by Rosenberg and what past masters have said about human consciousness. They all say the same thing in different ways, that there are more states of experience than there are states of brains.

(Sorry if it's annoying to have quotes from Jesus and his like in this context. I'm deliberately trying to blur the boundaries between what we normally divide up into science, philosophy, religion and mysticism, for I think that this is the only way forward for consciousness studies, but I'm still experimenting with how to do it without it seeming ridiculous. I've posted a few as just one or two on their own can look like a lucky coincidence).

"Jesus said, "He who has come to understand this world shall discover only a corpse, and whoever has discovered a corpse shall transcend beyond this world."

Gospel of Thomas
V. 56

"Cast off completely your head and skin. Thoroughly withdraw from distinctions of light and shadow. Where the ten thousand changes do not reach is the foundation that even a thousand sages cannot transmit"

Zen Master Hongzhi

Jesus said, "I shall give you what no eye has seen and what no ear has heard and what no hand has touched and what has never occurred to the human mind."

Gospel of Thomas
V. 17

""Pure Being and pure Nothing are therefore the same." This proposition of Hegel’s (Science of Logic, vol. I, Werke III, 74) is correct."

Martin Heidegger
‘What Is Metaphysics?’

"This Kingdom is seated properly in the innermost recesses of the spirit. When the powers of the senses and the powers of the reason are gathered up into the very centre of the man’s being - the unseen depths of his spirit, wherein lies the image of God - and thus he flings himself into the divine abyss… where [everything] is still, full of mystery and empty. There is nothing there but the pure Godhead. Nothing alien, no creature, no image, no form ever penetrated there."

Johannes Tauler (14th century)

"- Try to make your intellect deaf and dumb during prayer; you will then be able to pray."

"- When you are praying. Do not shape within yourself any image of the Deity, and do not let your intellect be stamped with the impress of any form; but approach the Immaterial in an immaterial manner, and then you will understand."

"- Never try to see a form or shape during prayer."

Evagrios the Solitary
On Prayer
The Philokalia

"Blessed is he whose beginning is before he came into being!"

Jesus
Gospel of Thomas
V. 20

"Die before your death."

The Prophet Mohammed

"Being is born of not being."

Lao Tsu
Tao Te Ching

"And he said, "Whoever finds the interpretation of these sayings will not experience death.""

Gospel of Thomas
V. 1

 

[StatusX]

I've been very busy for a few weeks and haven't been able to keep up with the discussion. But I have some free time now, and I just finished this chapter. I just had a couple comments on what's been said in this thread.

First, I think it's important to distinguish between a person and the inner experiencing being. I think most of the confusion concerning cognition and knowledge is due to this subtlety. A person, that is, the physical body and brain, can "know how" and "know that" and is a cognitive system. It is the inner experiencing being that "knows what," and this includes, eg., knowing "what it's like" to "know that" Napolean was a French dictator. I'm suprised this inner being hasn't been explicitly mentioned yet, because it is exactly what Rosenberg is talking about when he discusses the boundary problem.

Also, I think non-cognitive experience is possible, but we should be clearer on what we mean by this. The defintion by analogy is very weak. I could just as easily try to define a person's "torque" or "compression ratio" as being analogous to that of a car. I think all we should mean by such a term is that whatever our experience is at a fundamental level, that property can be present in non-cognitive systems.

Maybe the experiences are isomorphic to the physical properties of the system. For example, physicists define electric charge as the propensity for a particle to be attracted to another charged particle, and nothing more. Maybe an electron experiences charge, (note again that the word "experience" only picks out the fundamental property that takes a certain complex form in human brains; I don't mean experiencing charge is something like seeing a color), and this is sort of why that particle is charged. In other words, it is nature's way of keeping track of the properties of particles or systems. From what hypnagogue has said, I have a feeling this is something like what Rosenberg is going to propose.

However, before we get too carried away guessing what non-cognitive experience might be like, the absolute most important issue to clear up is how we can talk about experience. Every single word any of us have written is apparently a direct result of the structure and function of our brains and nothing more. So we'd better clear up how we can talk about fundamental properties under the physical before we jump to any conclusions. I know this is addressesd in the next chapter, and I'll be reading that later tonight.

 

[hypnagogue]

Also, I think non-cognitive experience is possible, but we should be clearer on what we mean by this. The defintion by analogy is very weak. I could just as easily try to define a person's "torque" or "compression ratio" as being analogous to that of a car. I think all we should mean by such a term is that whatever our experience is at a fundamental level, that property can be present in non-cognitive systems.

That's basically what the definition by analogy does. It says that we can understand the postulated experience X of a noncognitive system Y by saying that X stands in the same relation to Y as human p-consciousness does to human brains. Presumably this includes some sort of property of 'what it is like'-ness for X (albeit a quite primitive one, whose details are probably opaque to human understanding), as well as some sort of systematic correlation between the experiential properties of X and the physical properties of Y.

Maybe the experiences are isomorphic to the physical properties of the system.

I think this is already suggested by the above analogy. We'll have more to say about this once we get into Part II; for now, I'd rather wait until then to get into the details than to give glossing accounts.

However, before we get too carried away guessing what non-cognitive experience might be like, the absolute most important issue to clear up is how we can talk about experience. Every single word any of us have written is apparently a direct result of the structure and function of our brains and nothing more. So we'd better clear up how we can talk about fundamental properties under the physical before we jump to any conclusions. I know this is addressesd in the next chapter, and I'll be reading that later tonight.

Agreed that this is a critical issue. However, we won't be able to clear this up (at least for the purposes of discussion of this book) until much later on, when we've fully reviewed Rosenberg's theory of causation and how he ties it into p-consciousness. Rosenberg's idea is that our current understanding of causation is insufficient, and that understanding causation is critical to understanding p-consciousness; therefore, we'll need to do some work on causation before we can address the knowledge paradox.