I could be pedantic and ask if the brain is calculating Bayesian stats as opposed to doing something that can be described in Bayesian terms... but that might be the same as asking if a neural network trained to add two numbers is actually doing addition... I'm not sure the question can be made coherent. — Banno

Yeah, I see what you mean. Is the calculator calculating? It would be badly misnamed if it weren't. But yes, a tricky distinction.

Almost the same as the Chinese room. Let's not. :grin:

Almost the same as the Chinese room. Let's not. — Banno

Yes, a peril of cognitive psychology, particularly computational approaches. As soon as you leave the influence of the neocortex you lose intention and then the language we use to talk about the processes starts to sound weird (to those who care about that, many of my colleagues used to use intentional language about neurons without a care - I cannot do so without cringing a little). We would be fine if there were a neat cut-off, but there isn't - so I tend to just use it and cringe away. It made my lectures more visually entertaining anyway - "the posterior superior temporal sulcus suppresses (ugh!) signals from the v4 region which is expects (yuech!) to be in conflict with it's predictions (eurgh!) of social status"

But that language works... suppression, expectation and prediction have Bayesian uses. The danger is a philosopher thinking this explains something about human intentionality.

The danger is a philosopher thinking this explains something about human intentionality. — Banno

Yeah, depends what you mean by 'explains'. There are different standards for what constitutes a satisfactory explanation in different fields. I'll freely admit I've never been entirely clear on what philosophers want of an 'explanation' such that it satisfies their criteria for one.

I'll freely admit I've never been entirely clear on what philosophers want of an 'explanation' such that it satisfies their criteria for one. — Isaac

Neither have philosophers...

I'll freely admit I've never been entirely clear on what philosophers want of an 'explanation' such that it satisfies their criteria for one.
— Isaac

Neither have philosophers... — Banno

Is there something you are trying to explain? If yes, what others are trying to explain is kinda secondary.

freely admit I've never been entirely clear on what philosophers want of an 'explanation' such that it satisfies their criteria for one. — Isaac

Do you think intention is emergent? or an illusion?

As far as I can see atm there are unconscious processes, whatever their structure, that act on sensory input, and we have consciousness of the results of those actions, whatever the structure of consciousness. The unintended implications that e.g. there is some teleological submission process, or some terminus at consciousness, or some implied specific structure to consciousness, aren't really what my argument is about. It is simply that we are conscious of results of unconscious processing. — Kenosha Kid

I see. It seems then that our disagreement (small such as it is) is only over whether dismissal of Qualia in their entirety puts this idea at risk (throws the baby out with the bathwater, as you put it). My feeling is that the idea here is so generalised and applicable to a field much wider than qualia, that dismissing all talk of qualia maintains the conscious awareness of the results of unconscious processing completely intact. — Isaac

It is tricky. The nascent way we split up phenomena and describe them isn't a neutral process of observation and recording with respect to the topic of the thread. Reading off features from our perceptions involves the same process by which perceptual features are formed (to some degree anyway). If the devil is in the details of the formation process of perceptual features, the way we read off features from already formed perceptions effectively has a sampling bias in that regard. We're sampling from an already formed space of features introspectively rather than looking at the process of perceptual feature formation which is constructing the elements of that sample that we later sample from with another (related) process.

We're sampling from an already formed space of features introspectively rather than looking at the process of perceptual feature formation which is constructing the elements of that sample that we later sample from with another (related) process. — fdrake

Yep, this looks like we are getting somewhere towards the hard problem. That to me seems like indeed, a formal version of what is called the Cartesian Theater problem. Related is also the homunculus fallacy.

Another fascinating (to me) aside, but I must stop getting sidetracked. Have a look at this paper, if you fancy, it's really interesting. — Isaac

Reminds me of one of our prior conversations:

‘Theory theorists’ in cognitive development point to an analogy between learning in children and learning in science. Causal Bayesian networks provide a computational account of a kind of inductive inference that should be familiar from everyday scientific thinking: testing hypotheses about the causal structure underlying a set of variables by observing patterns of correlation and partial correlation among these variables, and by examining the consequences of interventions (or experiments) on these variables.

It seems then that our disagreement (small such as it is) is only over whether dismissal of Qualia in their entirety puts this idea at risk (throws the baby out with the bathwater, as you put it). My feeling is that the idea here is so generalised and applicable to a field much wider than qualia, that dismissing all talk of qualia maintains the conscious awareness of the results of unconscious processing completely intact. — Isaac

It was more that there is a reason why we have a concept of qualia, and that reason holds even if theories about what qualia are do not, and it seems to me that all of the problems lie with older, less scientific theory and not with the existence or not of properties and objects of consciousness, the things underlying the belief that qualia exist. There's a sense here with how Dennett is being interpreted, not just by Strawson but by yourself and fdrake, that since theories about qualia need improving, the term must be jettisoned. I can't think of any other field where this would be the case. Gravity was modelled as a force field for centuries. When Einstein discovered it was actually geometric feature of spacetime, he didn't jettison the term 'gravity', and that's a pretty fundamental distinction, much more so, I feel, than the difference between ineffable, intrinsic, private, immediate qualia and merely private and immediate qualia.

Renaming gravity because we previously identified it as a force field when it's not would have been extremely confusing and lo there is extreme confusion about whether consciousness is real or not. Throwing the baby out with the bathwater means we get people like Strawson leading people to believe that Dennett has claimed that consciousness is an illusion, because the contents of consciousness are an illusion, when nothing of the sort is claimed.

If I say Pete is 5 ft 11, with brown hair and hazel eyes, and his eyes are in fact grey, does Pete not exist? The existence of the referent is not dependent on the accuracy of my description. When you say 'qualia' do not exist, obviously that is going to be interpreted as 'the referents of qualia do not exist', which is precisely what we have seen, when in fact Dennett is saying that the theoretical description of qualia is wrong (and, furthermore, that qualia themselves, while real enough, are not scientifically useful).

There are plenty of cognitive psychologists and neuroscientist working under the former assumption without ever mentioning qualia or anything like them, so I think it can work. (there are, of course also plenty who do - much to their shame!). — Isaac

And presumably they're talking about the same thing, just using different terminology.

Dennett is saying that the theoretical description of qualia is wrong (and, furthermore, that qualia themselves, while real enough, are not scientifically useful). — Kenosha Kid

Yes, I find this the problem with Dennett. He doesn't seem to approach the hard problem. He keeps trying to hack away at easier problems. The question that people keep throwing back at him, is so what about the hard problem. And in answering this, he keeps going back to people's misconception about qualia which is confusing because that is not the hard problem. I'd rather Dennett just admit, "Fuck it if I know, but here are problems that are easier to possibly get an answer."

If the devil is in the details of the formation process of perceptual features, the way we read off features from already formed perceptions effectively has a sampling bias in that regard. — fdrake

Indeed, I was just reading a paper about how the process can cause information loss and, you're right, we can hardly discuss perceptual features not present due to the processes that lost them. And to this extent qualia may not be useful scientific concepts, as Dennett said. That said, there is, as both yourself and Isaac have pointed out, feedback between what we consciously perceive and the unconscious processes that form those perceptions, so any complete description of perception must surely account for what is perceived.

Indeed, I was just reading a paper about how the process can cause information loss and, you're right, we can hardly discuss perceptual features not present due to the processes that lost them. And to this extent qualia may not be useful scientific concepts, as Dennett said. That said, there is, as both yourself and Isaac have pointed out, feedback between what we consciously perceive and the unconscious processes that form those perceptions, so any complete description of perception must surely account for what is perceived. — Kenosha Kid

Aye. I imagine that the kind of accounts philosophically split along two lines:

(1) How are perceptual features formed?
(2) What is the phenomenal content of a given perceptual feature?

The first admits of functional explanations (it's a "how" question regarding a process), the second evokes an apportioning of phenomenal content to perceptual features, and we might be in a similar situation to the debate we just had (perceptual features are "submitted to" a phenomenal content receptor vs phenomenal content ascription is interweaved with the process of perceptual feature formation).

I strongly suspect that relating to our own perceptions in a manner that doesn't produce these conceptual traps upon reflection is a laborious, ongoing fight. A "relearning how to see".

Edit: I'd suggest that the "phenomenal content" of a given perceptual feature is the perceptual feature itself - or perhaps "the most attentionally prioritised aspects of the nascent perceptual feature components", or when introspecting "the attentionally prioritised aspects of the remembered/introspectively targeted perceptual feature", there's a lot of information loss involved - but Hard Problem enthusiasts wouldn't like that.

I strongly suspect that relating to our own perceptions in a manner that doesn't produce these conceptual traps upon reflection is a laborious, ongoing fight. A "relearning how to see". — fdrake

Yes, I think so too. And relearning how not to see, or hear in particular. Nonetheless, regarding:

perceptual features are "submitted to" a phenomenal content receptor vs phenomenal content ascription is interweaved with the process of perceptual feature formation — fdrake

we can still separate data in a third person way into what we are conscious of in a first person way -- phenomenal content -- and what we are not. As I said before, this does not imply a particular structure; it merely observes that we are conscious of things like 'the car on the left' but not conscious of things like 'transformed orientation of car on what is now the left'.

I'd suggest that the "phenomenal content" of a given perceptual feature is the perceptual feature itself — fdrake

Yes, maybe. The qualia then exists by virtue of us being conscious of it as opposed to not conscious of it rather than packaged and sent into consciousness pret a manger, which isn't really what I was getting at. The important point is that 'the blue car' is not in raw sensory input; it is something the brain adds because it has learned to do so, and that process of recognition is not part of that higher-order set of conscious processes (which may just be because the brain has also learned not to bother amplifying that sort of thing). Or, in other words, whatever triggers those processes are not perceptual features.

Do you think intention is emergent? or an illusion? — frank

Wouldn't it be both? Only tangentially related, but if an effect is emergent, then any reification is an illusion. Flash a bight magenta light on a white background, when it's removed a green shape will appear in its place, yet no green light was shone. This effect simply emerges from the combination of magenta light and antagonistic processing in the retina. It's still what we commonly call an 'illusion'.

The nascent way we split up phenomena and describe them isn't a neutral process of observation and recording with respect to the topic of the thread. Reading off features from our perceptions involves the same process by which perceptual features are formed (to some degree anyway). — fdrake

Yeah, my concern is still with...

we have consciousness of the results of those actions — Kenosha Kid

@Kenosha Kid, I'm still not sure I'm prepared to accept that picking a point in an ongoing feedback process and labelling it the 'result' doesn't set us off on the wrong path as far as perception is concerned.

Gravity was modelled as a force field for centuries. When Einstein discovered it was actually geometric feature of spacetime, he didn't jettison the term 'gravity', — Kenosha Kid

True, but we have jettisoned phlogiston, humours, elan vital, and ether, (haven't we?) so is it not still a case of deciding what category qualia fall into?

I strongly suspect that relating to our own perceptions in a manner that doesn't produce these conceptual traps upon reflection is a laborious, ongoing fight. A "relearning how to see". — fdrake

This is an important point. In my discussion with @Kenosha Kid we've been talking about the way the brain constructs a narrative post hoc to unite it's streams of input. If what we want, for whatever reason, is to grasp what going on beneath the surface, we can't be using our intuitive feeling about it as a guide, we can establish pretty early on that that isn't going to give us an honest answer to that inquiry.

I contend that a lot of this really is a debate about whether consciousness is sort of an immediate (instant) "gestalt" of experience or if it is a construction of micro-processes of neurons. I don't know if any modern philosopher would doubt that it is the latter. But to say that hard problemers don't recognize this would be running towards windmills. You cut the corpus collusum, you take away this or that part of the brain and, who would have thunk it, a cognitive capacity disappears! Also, the electro-chemical firings of neurons, and their networks happen in certain parameters like microseconds, etc. This is all recognized. That is not moving closer to the hard question though. It's playing in the same well-trodden sandbox. You can move the Cartesian theater anywhere you like, but its always set up somewhere.

Wouldn't it be both? Only tangentially related, but if an effect is emergent, then any reification is an illusion. Flash a bight magenta light on a white background, when it's removed a green shape will appear in its place, yet no green light was shone. This effect simply emerges from the combination of magenta light and antagonistic processing in the retina. It's still what we commonly call an 'illusion'. — Isaac

If a property is emergent, it has characteristics that are not seen in its building blocks. A tornado is an emergent entity. If I'm reductionist regarding tornadoes, I would claim that the concept of a tornado is misleading. There are no tornadoes and to the extent people believe otherwise, they have bought into an illusion.

If you get a chance, read this article. I'd be interested in your viewpoint.

If a property is emergent, it has characteristics that are not seen in its building blocks. A tornado is an emergent entity. If I'm reductionist regarding tornadoes, I would claim that the concept of a tornado is misleading. There are no tornadoes and to the extent people believe otherwise, they have bought into an illusion. — frank

You're point there seems close to the point I am making here:

We can also add in the odd understanding of how is it something can "emerge" in the first place. Emergence implies some sort of epistemic leap from one stage into another. I'll just leave it at that. — schopenhauer1

Would you agree that is similar to what you are getting at?

I'm still not sure I'm prepared to accept that picking a point in an ongoing feedback process and labelling it the 'result' doesn't set us off on the wrong path as far as perception is concerned. — Isaac

If it's a process, surely it has a result. For instance, what would you call the single firing neurons in response to Halle Berry's face? If it is something we are conscious of, and it is not an input to that process, it is a result of that process, no? I'm not saying that's the entirety of the purported qualia, which must remain the net experience of Halle Berry's face; nevertheless there are processes occurring which feed into that experience, not instantaneously, sure, with feedback, sure. But it's doing something which adds to our experience.

True, but we have jettisoned phlogiston, humours, elan vital, and ether, (haven't we?) so is it not still a case of deciding what category qualia fall into? — Isaac

Yes, because the referents didn't exist at all. The referent of 'qualia' is 'properties of consciousness' and, as Dennett says, these exist.

We can also add in the odd understanding of how is it something can "emerge" in the first place. Emergence implies some sort of epistemic leap from one stage into another. I'll just leave it at that.
— schopenhauer1

Would you agree that is similar to what you are getting — schopenhauer1

Yes. Knowing how tornadoes work requires more than understanding the mechanics of moving dust, and we can understand tornadoes without knowing anything about quantum physics.

Yes. Knowing how tornadoes work requires more than understanding the mechanics of moving dust, and we can understand tornadoes without knowing anything about quantum physics. — frank

I think we are agreeing.. My point was sort of the epistemological paradox of emergence. We know of all other emergence through the process of cognizing it. At what epistemic level do tornados exist? Everything we know about emergence happens within the epistemic framework of a "viewer". Without the viewer, what is it from something to move from one level to another? What does that even look like? There is always a sort of hidden viewer in the equation. I guess I hear key words from types trying to answer this like "top-down causation" but it seems like a modern way of positing Descartes' God that is a necessity for everything else to exist.

If it's a process, surely it has a result. — Kenosha Kid

Yes, we die.

(Sorry, couldn't resist the flippant answer).

Take calculating some iterative algorithm that has no p-type solution. The step you happen to be on isn't the 'result' of the process, it's just the transient stage you're currently at. If we did want a result it might more properly be something like 'you're going to doing this forever', or 'you'll never get a number below 100', or some such limit. That's the way I'm seeing perceptual processing, from day one the perception is not a result, its a prediction to be input into the algorithm generating the next perception...

But we might actually be getting into the weeds here. I think we're not so far from one another. It's interesting to hear how you see things differently, though.

Take calculating some iterative algorithm that has no p-type solution. The step you happen to be on isn't the 'result' of the process, it's just the transient stage you're currently at. If we did want a result it might more properly be something like 'you're going to doing this forever', or 'you'll never get a number below 100', or some such limit. That's the way I'm seeing perceptual processing, from day one the perception is not a result, its a prediction to be input into the algorithm generating the next perception... — Isaac

With sufficient pedantry, what demarcates the steps of the "iterations" of perceptions would also vary too, no? There's no guarantee that update steps correspond 1-1 with "instants" of perception as we'd introspectively, pre-theoretically or even experientially in this case draw the line. The indexical progression of update steps within the updating procedure isn't the same thing as individuation of situated ("subjective") states. It seems there must be indexical progression without reportable changes precisely because selectively apportioning working memory is a component of perceptual updates! The progression of situated states looks to have a slower clock than the indexical progression within algorithm (Libet's delay) and between its components (reflex triggering before conscious awareness that it was triggered).

Regardless, it does seem important to be able to study the "perceptual moment" and to give an account of how that arises from the steps of the updating procedure. Even if that perceptual moment is still a "finite stretching along in time" (as Heidegger puts it) so even "instantaneously" has temporal dependence (priors + expectations).

At what epistemic level do tornados exist? Everything we know about emergence happens within the epistemic framework of a "viewer". Without the viewer, what is it from something to move from one level to another? What does that even look like? — schopenhauer1

The now classic answer is: when the whole is more than the sum of its parts. That is to say, when there is a discernable and somewhat functional structure to the thing. A car for instance is far more than a pile of parts. It's a structure made of parts. Assembling those parts in the right manner for the final structure to work as a car requires skills, tools and work. When you lose a part (eg a wheel), it usely results in the car becoming dysfunctional and needing repair and part replacement.

Likewise, a living being is far more than a pile of atoms: it's an extremely complex structure made of atoms. This structure is able to maintain itself in spite of losing parts (molecules) all the time, by absorbing other molecules from its environment: you perspire, you drink for instance. To a degree, a living organism / structure is self repairable.

A car has been built and repaired, but a living being hasn't been "built" by anyone (in a Darwinian outlook at least - I trust we share such an outlook). So in the case of the living being, the bio-structure emerges somehow, as an extension of one (or two) other living organism, in the form of some seed. Darwinism avoids infinite regress here: evolution is then seen as the slow emergence of life, over at least 4 billion years, which ultimately gave rise to individual x of species y. Note that by and large, ontogenesis follows phylogenesis, so the emergence of an animal from a egg recalls the emergence of its entire biological ancestry.

So in the case of the living being, the bio-structure must have emerged somehow. — Olivier5

But what is emerging into what? How? Similar to the hard problem, you can tell me that these atoms and those atoms come together, but if the whole is greater than the parts, how is it that downward causation or top-down causation works without a viewer? To say it does it just does once things are combined is to beg the question of how the parts that created the whole created a new phenomenon. Certainly a viewer cognizes the results of combinations of parts to whole. What is that on its own without the viewer though? Before you answer this, keep in mind there is no viewer here. What is happening is not like what a human perceives happening just without humans. Rather it is the thing-itself, It is happening outside cognition. How is it this new epistemic event occurred. I remember @apokrisis used to refer to this as the "epistemic cut". And then he would go on a whole tangent on Peircean semiotics. Without invoking that particular philosophy, I think the term is useful so I'm going to use it.

This thread is a fine example of the way discussions and/or differences in viewpoints about a subject matter ought be performed. Kudos to the active participants here, particularly , , and . It was conversations such as these that initially piqued my interest in philosophy.

Last week, I studied Dennett's paper in the other thread which gave rise to this one, and found it compelling enough to reject the concept of Qualia, based upon it's having been rendered useless as a means to add anything other than unnecessary complications that actually inhibit our understanding of consciousness or conscious experiences. I suspect that's similar to 's take as well.

I agree with Dennett's characterization of what those who argue for qualia/quale are doing, when he says the following...

One dimly imagines taking such cases and stripping them down gradually to the essentials, leaving their common residuum, the way things look, sound, feel, taste, smell to various individuals at various times, independently of how those individuals are stimulated or non- perceptually affected, and independently of how they are subsequently disposed to behave or believe. The mistake is not in supposing that we can in practice ever or always perform this act of purification with certainty, but the more fundamental mistake of supposing that there is such a residual property to take seriously, however uncertain our actual attempts at isolation of instances might be. — Dennett

I think that the above quote is the most important point of the paper, although my reasoning for that may be too far off topic. The quote above has been discussed at length in terms of how the brain works, and in terms of how human perception models work, and although I've found those conversations very helpful, interesting, and relevant to the paper itself, I've also found that there's still much missing in terms of what consciousness consists of and/or is existentially dependent upon, despite the fact that I wholeheartedly agree that there are no such things as the way things look, sound, feel, taste, smell to various individuals at various times, independently of how those individuals are stimulated or non- perceptually affected, and independently of how they are subsequently disposed to behave or believe.

What's missing is the explanation of how those individuals are stimulated or non-perceptually affected, and how they are subsequently disposed to behave or believe that adequately describes thought and belief itself(consciousness). "Consciousness" as described by proponents of "qualia" is based upon a gross misunderstanding of what consciousness consists of, and how it emerges(here I'm quite fond of the discussion regarding whether or not perceptual features/properties/quale can be divorced from the actual individual's history and retain their unity as an entity).

I'd like to see that part of this topic gotten into in quite a bit more detail, but perhaps an aim to adequately explain how consciousness emerges, and what it consists of, is too far off the topic, because Dennett was not concerned about that in this particular paper.

What's missing is the explanation of how those individuals are stimulated or non-perceptually affected, and how they are subsequently disposed to behave or believe that adequately describes thought and belief itself(consciousness). "Consciousness" as described by proponents of "qualia" is based upon a gross misunderstanding of how consciousness emerges(here I'm fond of the discussion regarding whether or not perceptual features/properties/quale can be divorced from conscious experience and retain their unity as an entity). — creativesoul

I somewhat agree with this, if we grant Dennett's arguments for quniing qualia. However, you do seem to be espousing illusionism in this paragraph. Which would be that we're being deluded by some trick of cognition into thinking sensations of color, sound, paint, etc. are something they're not, which is some form of the private, ineffable subjectivity.

If there is such an illusion, the mechanism needs to be explained so that we can see how this illusion comes about. The problem I and many others have with this approach is it implies that sensations themselves are illusions, because that's the only way to avoid espousing qualia. Which would imply that we only think that we see color, hear sound, feel pain.

What could that possibly mean? And what does that do for epistemology if our sensations are themselves illusions? And aren't illusions themselves experiences?

Yep, there's a difference between red and a certain frequency. No problem. What about the qualia? — Banno

The hard problem arises here because we have sensations of a world which is different from our objective explanations of that world. Red isn't a certain wavelength of light, nor is it certain neurons firing. Red isn't part of our scientific explanation of the world. And yet we all have sensations.

Even if we dispense with qualia as incoherent, we're still stuck with the secondary qualities of perception, along with dreams, inner dialog, imagination, hallucinations, etc. We still have a modern form of the mind/body problem. It doesn't go away just because we ditch a problematic term.