It shows, as it says, that "color is in the perceiver, not the physical stimulus" and "color is a perceptual construct that arises from neural processing".

You're not going to convince me that it isn't saying what it's literally saying. — Michael

For a start, what it 'shows' is the empirical data, what the authors 'say' is the conclusion they draw from it, which is not that same as the empirical data underlying that conclusion.

The empirical data from that experiment is neural potentials in response to stimuli.

The conclusions bring in a whole.modeling assumption which, whilst perfectly valid, is not universally shared, and certainly not what constitutes 'empirical evidence'.

Notwithstanding that conceptual confusion, I'm not disagreeing that colour is a construction of the brain's processing systems, I'm denying that it is thereby not a property of external nodes.

The switch rivalry paradigm your experiment uses, for example relies entirely on the fact that the colour slide presented is really a consistent magenta. How do the experimenters know what colour the slide 'really' is such as to contrast it with the reported colour?

I've already given one example: — Michael

That's not evidence of an internal 'experience'. It tells us what happens in different regions of the brain in response to stimuli.

Your claim is that this internal state is a) consistent and whole - 'red' and b) the object of an internal sense, the subject of the sentence "I see red", and c) that it is not a property of external objects.

The experiment you've provided shows none of these three claims.

The empirical evidence is that external stimulation triggers brain activity that causes an internal, physiological experience. — Michael

What empirical evidence?

Purpose.

Are you deliberately ignoring this entire line of argument for a reason? All traits which carry significant cost (calorie or otherwise) demonstrate fairly well understood survival purposes (or other forms of selection).

You've not given any account of why you dismiss this meta-theory, or why you think 'feeling red' falls within it.

Outside of such an explanation, you theory explains empirical evidence no better than the standard one, but with this additional gap.

It is an unfortunately pervasive attitude, ignoring the difference between disagreeing and misunderstanding.

That one does not agree with a conclusion, or an interpretation, does not mean one has not 'understood' it. A mistake @Wayfarer seems uncommonly prone to.

or evolutionary benefit, if you're working within the same framework as me) — Isaac


Organisms may have features that have no evolutionary benefit. — frank

I explained that. Thinking is one of the most calorie intensive actions we do. The brain is a very expensive organ. There are no examples in the natural world of traits evolved which are calorie intensive (or otherwise costly) which nonetheless survive in the face of competition.

If you are arguing that features can be costly and still evolve, and that evolved features have no correlation to survival (or sexual selection), then you are literally arguing against the theory of evolution by natural selection. Which is why I asked @Michael for the alternative frame he might be using.

It explains colour blindness and synesthesia and why some people see the dress to be white and gold and others black and blue and why science describes the world as electrons absorbing or reflecting electromagnetic radiation of certain wavelengths rather than “particles of redness” and is exactly what the experiment I referenced days ago concluded. — Michael

So does the standard model, but without these holes. I'm asking why you choose the model with the holes (or why they are not, for you, holes at all). Why choose a model which creates this difficult to explain phenomena contrary to what we already have regarding evolved characteristics, when there appears to be no call for it?

I more that if no one can put in even an example answer, they type of thing expected, then I don't see how anyone can support the claim that no satisfactory answer has been given.

If I don't even know what I'm expecting, I can't claim to not have it.

If you want to criticize something, you first have to demonstrate that you understand it. — Wayfarer

I find this kind if thinking really insulting. Neuroscientists are clever enough people, their intellectual capabilities should not be in question.

As such they are as capable as any other human of thinking about their methodology and how it fits with psychology and any metaphysical questions they think are appropriate.

Philosophers are not magic, nor do they have access to a set of data unavailable to ordinary people. There is not a corpus of facts to be 'understood'. There is a range of viewpoints to be read about, or not, as each person sees fit. And each viewpoint arrived at by any given philosopher (or scientist) on the subject matter of philosophy is equally accessible, merely by thought, to any sufficiently intelligent human on the planet, including neuroscientists.

The idea that somehow all neuroscientists do is examine brains and never think of anything else is grossly offensive caricature.

Scientists think about other stuff, including (quite popularly), science itself. All philosophers do is think. Everyone can think.

If, as a result of this thinking process, a philosopher and a scientist disagree about the meaning or significance of a scientific theory, the philosopher has no claim to any authority, he just used his mind, same as the scientist.

Theories need to fit the facts, not the other way around, and it is a fact that the pain I feel isn't a property of the fire that causes me to feel pain.

So either there is a survival advantage to feeling pain, or feeling pain just happens to be a deterministic effect of something else that gives us a survival advantage (e.g. a complex brain that is able to effectively respond to stimulation). — Michael

Exactly. so what is the equivalent situation with 'seeing red' to which you want to extend this physiological response?

Your argument so far seems to be that because pain is a physiological state, then so can 'red' be. But that's woefully inadequate as a theory. By that same token, so can the number 7, or Donald Duck, or the solar system around alpha centuri. You have to have something more than just "because one thing is X other things can be X too, therefore P=X".

I'm asking what meta-framework (or evolutionary benefit, if you're working within the same framework as me) justifies your belief that 'seeing red' is the same type of thing as 'feeling pain' - it's not sufficient that it merely can be on the grounds that some things are.

The feeling of 'pain' serves a purpose related to the cause of the pain. You 'feel' the fist (in the first sense) so that you know where it is, how fast it's moving etc. You 'feel' pain (in the second sense) so that you avoid such situations in future.

What I'm asking is the equivalent in your model.

I'd say you 'see' red in the only sense I understand of the word 'see', that is you make some prediction of the state of some external data node (that it is 'red'). You do this so that you can respond to it appropriately (say, eat it because it's ripe). for this to work, 'red' has to be a property of that data node, if if's not then it can carry no information from it, and so your response is disconnected.

You want to say that 'red' is a feeling in this second sense, as description of the state of you mind, not the external node. So why? The 'feeling' of pain (second sense) sets us up to avoid the cause. What does the 'feeling' of red do?

I don't understand your question. — Michael

I start from a principle that features of human physiology evolved within a system where their cost did not exceed their survival benefit. In such a system, it would be practically impossible for the huge amount of calories mental processes consumes to be justified if all it did was detect internal states of the same system, I can't see the survival advantage.

So I'm asking you what the survival advantage is, or what your alternative meta-biological theory is. Without either I can't see how you can sustain the model with such glaring holes in it.

I just know that it does. — Michael

Here's a model of pain...

1. an external state stimulates a nociceptive nerve ending
2. that signal (among hundreds of others) travels through an hierarchical system of prediction engines which attempt to output a response appropriate to reducing the uncertainty of that external state (either by manipulating the external state by acting on it, or by refining the model by further focussed investigation)
3. one of those outputs is to alter the release rate of certain hormones which in turn influence the output of other prediction engines (shifting their priors slightly in favour of certain types of output)
4. This state of affairs, this hormone affected setup, if you were to report it (either to yourself, or to others) you would use the expression "feeling pain" to describe.

If the above were the case then how would it clash with what you claim here to "know" about your feeling pain. If the last thing your brain does, after going through the process of predicting the state of external nodes, is to render a self-report which you respond to as a 'feeling of pain', then how would you distinguish that from the actual functioning of your brain in response to external stimuli?

The indirect realist argues that what we feel in the former sense is a mental representation of what we feel in the latter sense; that what we feel in the former sense is not a property of what we feel in the latter sense. — Michael

Right. So what's the point of it?

If what we're sensing is not a property of anything external to the system doing the sensing, then why is that system sensing anything at all? Why is it only detecting properties it itself has made up?

What's your evolutionary model for how this situation came about in all biological creatures (or how it evolved out of one in humans), what mechanisms were behind such an evolution, how could it possibly have sustained the enormous amounts of calories such activities involve without causing sufficient biological cost to have evolved out?

I'm not getting anything of the meta-biological framework your theory sits within.

You asked for evidence, not theories. — Michael

No. I asked for answers.

I'm asking what kind of answer would be a satisfactory one. Just an example.

Like if someone asked "why do cars have wheels?" I can say that the kind of answer I'm looking for would be something like a rational reason the engineer designed wheels. It would relate the purpose of the car to the physics of the wheel.

The answer might be completely wrong, or have no evidence at all. It's not about the evidence, it's about the form an answer should take.

"Why did you do that?" - list of motives
"Why is the sky blue?" - physical cause of 'blueness'
"Why did the chicken cross the road?" - surprising answer (or non answer) designed to amuse
"Why do humans have noses?" - evolutionary (or developmental) advantages of the nose...

"Why do we have consciousness?" - ...

... what's the kind of answer that goes there?

I don't understand the issue. Pain isn't a property of external world objects. I feel pain. There's no problem here. Colours aren't a property of external world objects. I see colours. Suddenly there's a problem? — Michael

Yes. Primarily 'feeling' is a term we use for multiple meanings, one of which is a summary of your mental state "how are you feeling today?". So "I feel pain" and "I feel the grass" have two different meanings. The former being used in the sense of describing a state of mind, the latter in the sense of touch-sensation.

You specifically wanted to talk about the problem of epistemology with regard to perception and not want to get caught up in semantics. Given the, it is only this latter sense of 'feel' we're interested in here, the one which is about you sensing the external world with your nervous system.

There are models which make sense of internal 'feelings' (summaries of mental states) in a biological framework, but they'd be of no use to us here since you want to discuss perception, and summarising one's mental state is not an act of perception.

When our nerves are stimulated in certain ways, we feel pain. That pain, even though an "internal" thing, indicates that something is going on outside us, and we react accordingly. — Michael

So this is muddled. When our nerves are stimulated in certain ways we 'feel' the cause of that stimulation (using feel in the sense of sensory perception). If I hit you you would say that you felt my fist contact your arm. You might also make a self report about your mental state "I feel in pain", but that's irrelevant to perception and so outside of the scope of this conversation.

When our temperature is lowered sufficiently, we feel cold. That cold, even though an "internal" thing, indicates that something is going on outside us, and we react accordingly. — Michael

Again, note none of your leading examples are yet straying from the use of the ambiguous word 'feel'...

When our eyes are stimulated in certain ways, we see red. That red, even though an "internal" thing, indicates that something is going on outside us, and we react accordingly. — Michael

... and now substitute 'see' as if it were a given that it had the same dual meaning as 'feel'. No one asks "How are you seeing today", no one says "when you girlfriend left you, how did that make you see?"

'See' does not have this dual meaning where it's also describing a self-report about mental states.

Hence when you say "I see a tree" we're only describing your sensory perception, your estimation of the causes of external stimulation of your sensory perception nodes of your nervous system.

So to say "I see 'red'" but have 'red' as being some internal state, you'd have to posit that those sensory nodes somehow also terminate internally, that they sense other parts of the same system.

n this second phase of the training, the Reinforcement Learning from Human Feedback guides the neural network towards producing responses that prioritize the rational patterns it already had picked up when it was trained on texts, but would otherwise be sidelined by the model in favor of majority (or contextually salient niche) opinions.

Of course, the RLHF phase of the training also is an occasion for the model to assimilate the biases and prejudices of the humans who supervise the training. — Pierre-Normand

Ohh, that's even more interesting! So what mechanisms are in place to determine whether the human trainers are seeing post hoc rationalisation in approaches that give answers they personally find plausible; as opposed to using pre-determined algorithms of their own to push linear approaches?

I suppose what I'm struggling to understand is where the injection of rationality comes from.

We start out with some collection of text and the AI is trying to work out what is likely to come next given some seed text. So broadly speaking, the most likely thing to come next in any corpus of text is going to be whatever human beings think it most profitable to say next (in whatever context the corpus is drawn from).

So absent of any further filtering, the AI would make a successful human (in that context), but not necessarily a rational one. It would, if only statistically, be employing the full suite of human mental algorithms to come up with the overall best thing to say in the context (one of which might be rationality).

Then we put it to human training to try and give it rewards for its reinforcement training, essentially changing the rules of 'success' to being only those next words which follow rationally, not any other algorithm.

But the humans selecting the responses to reward are not somehow more capable of making those selections of the basis of rationality than the original humans were who wrote the corpus on which the AI was trained absent of RLHF.

Something here reminiscent of this modern beatification of fact-checker (who now, it seem, are deemed somehow more capable of unbiasedly checking a fact than the original human who wrote it was).

I'm not sure that, essentially, getting a load of middle class, Western, English-speaking tech industry employees to vet some large corpus of human intellectual output for (statistical) rationality is really going to yield the improvement they might think it ought.

Assuming the AI is working, then it is giving the most statistically likely response from a wide set of human intellectual output in raw farm without the RLHF. I think that's more useful* in its raw form than the form filtered through the narrow culturally-mediated filters of a very small subset of that humanity.

* 'Useful', not necessarily sensical.

This is a great question, and I'm sorry to say I can't even give you an example bare-bones answer. Absolutely no idea. — bert1

Then how do you know that the answers given so far are unsatisfactory. If I went into a room searching for something and you asked what it was I was looking for, if I said "I don't know", you might reasonable ask "then how do you know you haven't found it yet?"

The standard answer has already been given by (albeit with considerably more confidence than I would have given it). There's some evolutionary advantage - where "some" has been expounded at length by various authors, not all agreeing.

It's the standard answer to why we have any feature we have. either random mutation that was simply 'not so bad as to get in the way', or one which conferred an actual advantage (either niche or sexual section).

So perhaps it might trigger something for you to say why that answer isn't an answer for you.

if there is such a thing as first person consciousness, and if first person consciousness is essentially private, then by necessity there can’t be any sort of public, scientific evidence of or explanation for it. — Michael

I'm not seeing how that follows. I can see how, if a thing were inherently and unassailably private we couldn't publicly discuss what is is, but I don't see how we couldn't publicly discuss how it came about or what purpose it might serve.

If there were some completely secret contents of a black box but if every time I added a coin to that box it spat out a can of beer, I don't need to know what's in the box to have a reasonable scientific theory that the box is designed (evolved, if natural) to vend beer, and that it does so in response to money being placed in it. I could experiment with different coinage, different currencies. See if there's a relationship between coin and beer type... I could develop a dozen perfectly valid, sound theories about this box, how and why it works, all without having a clue what's in it.

suppose that this fact is an arcane piece of biological knowledge ignored by most English speakers but known to many French biologists. In this case "Les pommes sont rouges" would figure in its training data. How does that help it predict that "red" is a probable continuation of the English sentence? That might be through the indirect association that stems with "red" being associated with "rouge" or a structural similarity between those two words that is somehow encoded and accessible in the language model. — Pierre-Normand

Yes, that's something like I thought it might work, only in my thinking, the mere act of translation would weigh statistically in the model. It's got to work out some kind of model optimisation algorithm, right? At the end of the day, it's going to have hundreds of possible 'next words' and it needs some kind of policy and then a policy optimisation formula (Bayesian optimisation?) to enable the reinforcement learning.

So what I'm thinking is that, if it'd being trained largely by english speakers of a certain class and background, then it's going to optimise policies which yield results those trainers are going to find satisfactory. They may be simple grammar section policies, but with translation, I suspect there's often a disconnect (phrases translated from French to english never sound quite...English.. if you know what I mean). this may not happen often, but over thousands of iterations it's going to create a statistical bias in acceptance by English-speaking human trainers which could easily bias GPT's policy optimisation algorithm to assume there's something sub-optimal about translation.

One of the fascinating things about reinforcement learning is that you don't always get a human-readable version of the policy the AI is following - it could be anything! We only know the results.

I will! I pretty much report everything I do with GPT4 that generates interesting results, but when I get around to testing this specific issue, I'll make sure to flag you in the post. — Pierre-Normand

Thanks.

When I talk about seeing red I mean it in the same sort of sense as when I talk about feeling pain, and the red I see, like the pain I feel, isn't a property of external world objects. That's the argument I have been trying to make. — Michael

Right. so if it's not a property of external world objects, then what's your theory as to why we sense it? And how do you justify undermining the current paradigm that the brain senses external states in order to predict the results of interaction with them? what's the evolutionary advantage of a system where the brain spends time detecting the state of other parts of itself?

I'll ask you the same as I ask everyone who asks this question...

Why does any of this constitute or necessitate subjective awareness. or consciousness, or the capacity to experience?" — bert1

... What would an answer look like? Give me an example answer. It's doesn't have to be the right answer, just an example of what sort of thing would satisfy you.

Like if I said "no one has yet answered the question of what is 567,098,098 * 45,998,087" I could clearly tell you what sort of thing I would accept as an answer - I'm expecting some big number - even though I don't know what that number is. Without that framework, I don't see how I could possibly claim that no-one's answered the question yet.

So what's the sort of thing you'd be satisfied with? If I went into my lab tomorrow, had a really good look at some brains, and came back to and said "Brain activity requires consciousness because..." What would you accept?

Lol. Really? I'm talking about Ukraine and Europe to avoid talking about other issues???

This is a thread about the war in Ukraine... hence, it is about the war in Ukraine. It's a silly argument to make then that I'm talking about the war in Ukraine in order to avoid something else.

A far better thread to talk about it would be the various US threads, btw. — ssu

Gods, you can't even follow a fairly simple line of argument. I'll make it as simple as possible...

You said - Russia would have been better to have followed Europe's model of 'soft' imperialism.

I said - Europe's model of 'soft' imperialism has demonstrably lead to more harms that Russia's more old-fashioned approach and so no, it wouldn't have been better. We need neither.

You said - But those harms are in Africa, Latin America and other non-European nations, let's just look at Europe and the fact that Ukraine could be European.

I said - But the prosperity of Europe is bought at the expense of exploitation in those other places, so we need to include that in any assessment of which approach causes most harm.

You said - this isn't a thread about those places

... which makes no sense whatsoever in the context.

The argument is directly and entirely related to the war in Ukraine - the topic of this thread. It is that promoting Europe's and the US's systems of soft imperialism as a solution to this war - the current war, the one this thread is about - is not ethical because those systems cause more harm to other nations. that's why we're discussing those other nations, they are collateral damage from the type of soft imperialism you are espousing.

Now. Having laid that out. do you actually have a counter-argument? Any evidence at all to refute the notion that soft (economic) imperialism does more harm - with its impoverishment, suppression of development, promotion of terrorism and militaristic foreign policy - than old-fashioned expansion by territorial wars?

I was tempted to think that it was to some degree immune to the Sapir-Whorf hypothesis. But what you here mention suggests that it not only expresses itself in specific languages but might also think (or manifest results of implied cognition) in whatever language it is communicating into at the moment. I wonder what might happen when I ask GPT4 to reply in English to queries that I formulate in French and vice versa. — Pierre-Normand

I'm guessing (not knowing the mechanisms) that it seeks out preferentially information in the language it's asked in? I'm think if it had to look something up in history books or current affairs, would it go to native language sources first?

In that sense, I can see it being very much vulnerable to Sapir-Whorf type effects.

I can also see it being vulnerable to other cultural biases, not restricted to language. I'm thinking perhaps of dominant scientific frameworks. Niche example from my own field, but anyone researching cognitive science in the UK would get a completely different perspective to someone doing so in the US, we have a very different meta-framework here, particularly in the more psychological and social-psychological cross-over points. Would a UK GPT, or a European GPT have a different understanding of scientific theories because of the different paradigms of its primary training material?

I'm tempted to try this while discussing topics where the Sapir-Whorf is expected to come into play and see which one of the two language most impacts its cognition. — Pierre-Normand

Yes, do let us know if you do. It's fascinating. The world moves on so fast with so many interesting new areas of research.

So it is not the case that "feeling" is one thing, that "pain" is another thing, and that the former is "done" to the latter; it is just the case that "feeling pain" is a single (possibly private) thing. The same with feeling cold and seeing red. — Michael

Now you seem to be going back to semantics. Sure, 'feel' can be used to describe more than directly sensing something. But we 'feel' the grass beneath our feet. We 'feel' the rough bark of the tree. You wanted to restrict the conversation to the epistemological problems of perception. The process of perception is one of estimating a surprise-reducing model of external states. So insofar as the epistemological question is concerned, you either 'feel' pain with pain being some external state, or that particular use of 'feel' is not the use we're concerned with.

So when I say you don't 'feel' pain, I'm using the term 'feel' in the same sense as it's used in perception - I 'feel' the grass, the trees, the wind etc.

Me too. I like to periodically remind the quislings here just who it is they're supporting.

An amnesty declared in 2019 retroactively legalised thousands of buildings that did not meet earthquake construction standards, as long as fines were paid. The move came despite warnings from engineers and architects. Up to 75,000 buildings in the earthquake zone were granted such a reprieve, according to the Istanbul Union of Chambers of Engineers and Urban Planners. Across the quake zone, developments slid from their foundations or splintered in on themselves, crushing those inside and entombing thousands. — https://www.theguardian.com/world/2023/feb/19/erdogan-faces-backlash-over-building-standards-in-city-wrecked-by-quake

So? In what way does that make it any better? — Isaac

We are talking about Europe... and especially Ukraine. As I've said, other developments in other continents deserve and have gotten their threads. — ssu

No, You're talking about Europe in order to avoid talking about the misery their policies have caused other nations in the developing world. I'm talking about humanity and the effect certain approaches to foreign policy have on them as a whole.

Sorry, but especially "peace and prosperity" isn't a zero-sum game. It hasn't been stolen from others. In fact, you can see that for Spain and Portugal their colonial posessions didn't create that prosperity compared to other countries without colonies (like Switzerland and Sweden etc.). The prosperity of a country usually comes through trade. — ssu

see - https://www.cambridge.org/core/journals/itinerario/article/abs/economic-dynamics-of-spanish-colonialism-in-the-nineteenth-and-twentieth-centuries/0D79153B10EA4F2422DF2354C406BA22

https://oxfamilibrary.openrepository.com/bitstream/handle/10546/621094/bp-eu-maghreb-trade-migration-policies-111120-en.pdf

https://www.kcl.ac.uk/news/the-paradox-of-eu-trade-policy-and-migration-from-africa

https://www.tandfonline.com/doi/full/10.1080/03056240701449646?casa_token=P8tBqWXo_98AAAAA%3ABhGn22sK9T5wur6XyvXDFAYmRdXIH0AxE_vMUeyoV5DL3bfMv5jksO5hUohksemOwJrWs0ZrzBmo2Hw

...and that's just trade. we haven't even got into environmental exploitation (see foreign oil exploitation in Nigeria, for example, or the impact of investment policies tied to reductions in government social spending, or the illegal fisheries policies, or ....

So you're saying that for the countries which are poor, it's their own fault, just want to get that clear? Not colonialism, not pernicious trade deals, not World Bank and IMF policies, not foreign support for corrupt governments... The people themselves. The starving poor have only themselves to blame... that's your claim?

I think you have to make the argument why Ukraine cannot be what the people want it to be. — ssu

Why? I've no doubt at all Ukraine can be what the people want it to be.

Ukraine has resources, it has an educated people. The problems aren't so great as they are in let's say Afghanistan. That's not bluster. — ssu

Seriously? Have you seen the devastation? You're argument is that it's all worth it because Ukrainians have a higher education standard than Afghanistan?

Political systems have changed, there are good examples of this in history. — ssu

OK, let's have them then... Which political systems have changed for the better as a result of US military aid?

No, you haven't. Things have happened. Not mere ... — jorndoe

List them then.

My house has become a den of red, black, and white. — frank

White Stripes fan?

Nothing you say can convince me that I don't feel pain. — Michael

Where is this 'pain' and what sensory nodes to you use to 'feel' it?

None of which rules out the experience of redness. — frank

No. The experience of 'redness' is ruled out by there being no evidence, nor need, for any such thing.

In fact, your view is more consistent with first person data than opposed to it. — frank

Then you've misunderstood my view. First person data is useless. It needs to climb an entropy gradient and it can't do that by being self-referential.

They are no more representations of apples than pain is a representation of fire or cold a representation of a temperature of 0°C. They are just an effect of stimulation. — Michael

An 'effect' of the stimulation would be something like reaching for the word "apple". You're still inventing 'representations' but now just want to call them something else.

When we say "the post box is red" we don't mean that there's some thing 'redness' which the post box possesses — Isaac


We do according to the (phenomenological) direct realist. They commit themselves to something like colour primitivism. Indirect realism is a response to such claims. — Michael

Well then we'd jointly disagree with such a position, but I very much doubt it's that simple. I'd first want to know what kind of thing the colour primitivist thinks colour is before dismissing their position hastily.

The pain I feel isn't a mind-independent property of fire. The cold I feel isn't a mind-independent property of the air. The sweetness I taste isn't a mind-independent property of sugar. The colour I see isn't a mind-independent property of the apple.

It makes no real difference if we describe this as feeling or tasting or seeing "mental representations" or if we describe this as feeling or tasting or seeing fire and air and sugar and apples. That semantic argument is, really, a non-issue. — Michael

As above, you're still invoking representations. There's no such thing as 'the pain you feel' You don't feel pain, you feel the fist. It's the fist which your systems are trying to predict, it is the external state that is the subject of your speculations, otherwise those speculations are pointless, they have no purpose to the organism. It doesn't need to know how it's feeling, it needs to know what's going on outside (and this means 'outside' of the system - meaning areas of your own body that are not part of the CNS).

The whole model of how and why the CNS does what it does doesn't work if all it's doing is predicting the state of other parts within it's own Markov blanket. Seeing, feeling, hearing, smelling... these are all descriptions of hierarchical prediction processes for which the only plausible model of their evolution is the reduction of entropy from forces external to the system. there's just no need for a CNS to be organised the way it is otherwise. So the idea that is sees, feels, hears, and smells itself is either nonsense, or an idea which stand in need of a complete rebuild of cognitive science from the ground up.

The dismal reckord of the West has taken place especially in the Middle East and also in Africa and earlier in Asia, not actually in Europe. — ssu

So? In what way does that make it any better? Part of the problem here is that the 'peace and prosperity' enjoyed by the West is bought at the expense of exploitation elsewhere. Russia, taking less of a globalist approach to exploitation (though branching out now in Africa), focusses on the direct geographic exploitation of its immediate neighbours. It's not pretty, but the facts are indisputable - it's actually less harmful overall.

There's still the very real question @boethius raise above as to whether Ukraine is sufficiently 'Western' for America to treat it as being 'part of the club'. Here, as a reminder, is how American military intervention has 'helped' African nations...

Throughout all of Africa, the State Department counted a total of just nine terrorist attacks in 2002 and 2003, the first years of U.S. counterterrorism assistance to Niger. Last year, the number of violent events in Burkina Faso, Mali, and western Niger alone, reached 2,737, according to a new report by the Africa Center for Strategic Studies, a Defense Department research institution. This represents a jump of more than 30,000 percent since the U.S. began its counterterrorism efforts. (Wagner has only been active in the region since late 2021.) During 2002 and 2003, terrorists caused 23 casualties in Africa. In 2022, terrorist attacks in just those three Sahelian nations killed almost 7,900 people. “The Sahel now accounts for 40 percent of all violent activity by militant Islamist groups in Africa, more than any other region in Africa,” according to the Pentagon’s Africa Center.

The impact of armed conflict and forced displacement on Nigeriens has been enormous.

Last year, an estimated 4.4 million people experienced dire food insecurity — a record number and a 90 percent increase compared to 2021. Between last January and September, almost 580,000 children under 5 suffered from wasting. This year, the United Nations estimates that about 3.7 million Nigeriens, including 2 million children, will need humanitarian assistance. Many of those in need are also the most difficult to reach due to insecurity.

It’s worth noting that in 2002, when the U.S. began pumping counterterrorism funds into the country, the overall food situation was described as “satisfactory” and undergoing “progressive improvement,” according to a food security monitoring agency set up by the U.S. Agency for International Development. — https://theintercept.com/2023/04/02/us-military-counterterrorism-niger/

I think there is the possibility that Ukraine can transform itself just like the Baltic States or Poland has. — ssu

Bluster. Does it even cross your mind what you're suggesting the world should tolerate on the basis of you're speculation here?

So you didn't read them. Not going to dig them up and list them again, but there's a recent one here (ECFR). — jorndoe

I read them. As I said, opinions speculating on selected evidence. From your article...

... the government has promised to deliver ...

... is expected to ...

... should further strengthen ...

... reforms could translate into a renaissance in Ukraine’s fight against corruption ...

The article then lists a whole load of anti-democratic changes which (unlike the speculations above) have actually happened.

Then we're reminded, regarding the extremely partisan view of this one author, ...

The European Council on Foreign Relations does not take collective positions. ECFR publications only represent the views of their individual authors.

So yes, opinions so far.

Here's another...

https://jacobin.com/2023/02/ukraine-censorship-authoritarianism-illiberalism-crackdown-police-zelensky

One year after Russia’s invasion, Ukraine is backsliding away from democratic freedoms and liberal pluralism.

Fascism is specifically about nationalism and an aggressive military. The ills of corporate rule are not correctly called "fascism ". It's just the dark side of liberalism. — frank

Fascism is a pretty general term these days, but it's exact meaning isn't the point.

As long as pessimism isn’t a preventative for action, I’ve no problem with it. — Mikie

Indeed. Rest assured I'm pissing more people off in the real word than suffer from my "militancy" here.

There's nothing random about his posts. They're very consistent and logical. If you familiarize yourself with fascism, you'll see that his view is exactly the opposite of it. — frank

It's not the foundation that's inconsistent and illogical, it's the manifestation. One can reasonably hold an ideological position that we ought maximise individual freedom, that we ought not impose on others, that we ought let people say what they want and manage the consequences... These are perhaps not positions I agree with, but they are valid foundational principles and (as you say) not fascist.

But the issue is how to get there from here. It's in that issue that the inconsistency is manifest, as is the fascism. Taking what we currently have, for example, and just ditching government regulation is fascism (it would be a fascism of corporate rule).

It is inconsistent to argue, for example, for current wage and property practices (predicated entirely on the assumption of taxation) and then argue also for the removal of those taxes.

Positions which hold individual freedom as paramount are not, in themselves, the problem. It's position which want to take all the benefits accrued from not have done so for millennia and then keep the wealth whilst ditching the responsibility. That's the inconsistency.

It could have been: "what's that red thing over there?" — frank

Yes, but why?

I've given an account of the need to reduce external surprise from both an evolutionary perspective and from a purely systems theory perspective. any self identifying system has to combat entropy gradients (in terms of information) and those gradients are Gaussian. so we minimise surprise, we treat things consistently, and (to the best of our ability) in ways which give predictable results based on their actual external-world states.

I can't think of a single reason why would just go about asking each other what our private thoughts are called?

"The post box is red" is the answer to some question. Understand the question, and you'll understand the answer. — frank

The question is... "in what way are we treating the post box?"

And the thing with this kind of speculation is that we have a near a perfect historical case study as you're ever likely to get.

In 2014 Russia did steal a part of Ukraine's territory. We did just let them get away with it. We've got 8 years worth of data on how that went. Basically it was as we would predict here. Things were moderately worse for the population there, but not by a substantial margin (compare to Donbas where Ukraine were still in control - similar levels of human rights abuses and restrictions on freedom). The costs to society of a full scale land war against one of the world's largest armies would have been far greater (again, we can measure those costs because we're seeing them today) than the losses we actually measurably saw in ceding Crimea to Russia.

So if you take the best data we have, the costs to humanity of this war far outstrip the costs to humanity of Russian rule (over Ukrainian). We have the data on that, we don't need to speculate about 'trajectories'. Russian rule over Crimea for the last eight years has been awful, but not more awful than the war that would have been required to prevent it.

So far you've just followed the Western propaganda. I've given solid evidence about Ukraine's human rights record, arms dealing, corruption, and oppression and you've come back with nothing but bluster. — Isaac


... just ... propaganda ... bluster? :roll: You've consistently ignored comments regarding the trajectory of Putin's Russia versus Ukraine's trajectory, — jorndoe

Do you understand the difference?

I can disagree with those comments. They are opinions, speculations based on a selection of facts. It is not reasonable to disagree with Amnesty International's record of human rights abuses. they are the experts in this field. It is not reasonable to disagree with OECD's assessment of black market arms sales, you don't have any data to combat that. It is reasonable to disagree with an opinion about the future. There's no data from the future.

And again from you, a complete failure to address the weight. Yes Ukraine are doing better than Russia. We agree. Ukraine are much better than Russia in terms of Humans Rights and democratic freedom (though I disagree with regards to economic freedom). And I agree Russia are getting worse and Ukraine are getting better.

So the question no-on wants to address is - are they enough better to justify the harms of continued war?

It is not sufficient for them to be merely better. It is not sufficient for option A to be better than option B if one has to, say for example, starve a million children, to get option B. Option B has to be enough better to justify the cost of getting there.

No one seem willing to admit that, no one seems willing to measure up the realistic improvements a Ukrainian win could bring vs the costs of getting that win.

I just meant that there's no intermediate object, no 'representation' of an apple. — Isaac


I’m not saying that either. I’m saying that the reality of colour perception is like this: — Michael

That's a representation. Those objects in the thought bubbles are representations of apples.

The essential point is that the apple in between them isn’t coloured. It reflects a certain wavelength of light, but that’s all. Colour primitivism, which naive realists believe, is false. — Michael

I think this simply misunderstands what the words mean. What it means for a thing (in the external world) to have a colour. When we say "the post box is red" we don't mean that there's some thing 'redness' which the post box possesses, we're instead declaring and reconfirming our joint commitment to treating the post box a certain way. But it's the external post box that is the subject of that commitment, not some internal representation.

when the man uses the term “grue” to describe the colour of the apple, he’s referring to what’s present in his experience and not present in the woman’s (in the particular example of that image), not to the fact that the apple reflects light with a wavelength of 450nm. — Michael

Then why have representations at all? Why have the word?

In my preferred model of perception, we attempt to predict the external causes of our sensory inputs so that we might combat the entropy otherwise induced by external forces and maintain our integrity. You could put that in evolutionary terms as being a need to predict the environment so that we can survive what it's going to throw at us.

But this requires that what we're predicting is the external world, the actual thing outside of us which might impact our integrity. And when we live in groups, we do this socially. We co-operate to better predict external causes and make ourselves more predictable to others (in the hope they will return the favour). So the important thing about labelling something 'red' is the co-operation, the surprise reduction, entailed by doing so. It's important that we agree and it's important that what we agree about is an external cause.

If all we're labelling is our own private 'representations', then I really can't see the point. Why would you care? Why would I? What difference does it make to anyone what your private representation is called?

it’s possible. Look at civil rights, women’s rights, gay marriage, even attitudes towards marijuana. — Mikie

Yeah, sure we do make progress. What matter, I think, is how. Did those campaigns succeed just by continually banging the drum, or was it something else? Or... did some succeed just by attrition where others needed something more. Let's not forget, the authorities were, at one point, terrified of the civil rights movement. It was not sufficient for them to merely point out the law needed changing. and yet with gay marriage (not that I'd want to undermine the excellent campaigning done toward that end, but...) it was pretty easily won, by comparison. All that was required really was to keep pointing out how unfair it was and eventually there was enough political will to act.

If you look at the trajectory of progressive success over the last 100 years, I don't think it's random. we've seen an unrelenting concentration of power and wealth in the hands of an increasingly small number of people, but alongside that an increasing amount of freedom regarding the expression of individual identity. Lifestyle choices, sexuality, family etc all seem very much on the cards for progressive change. Economics very much a battle fought tooth and nail for every millimetre of ground.

Hence my pessimism. It's not that I don't see progress, it's that I see that progress being very specific. I think a movement to restrict gun ownership to those with a licence (if easily obtained) might work, and would certainly help - because it's not an economic change, it doesn't really change the fundamental sociao0economic structure of the country. But it also won't help (not much anyway). The country is so suffuse with gun that a psycho is going to have little trouble getting hold of one, licence or no.

What's needed to have an impact, is a change in culture (so that there's actually fewer guns around), and that removes a huge economic tranche from the system (not to mention changing a culture which makes good compliant little consumers out of the now terrified population). That's an economic change, and if history is anything to go by, those in power are going to have to be terrified of those pushing for change before anything will happen... At the moment, they're not even missing a step over it. they have a totally compliant population who treat the word of authority as if it were gospel, why on earth would they change?

What I was referring to was that Russia would have had ways to influence it's neighbors to keep them out of NATO without resorting war and annexations. — ssu

I agree. And if the US, and Europe are any example, those 'ways' will have caused more death and destruction than wars and annexation. The record is in black and white. Deaths, ill-health, famine and ecological destruction wrought by the Us and Europe's 'soft' imperialism outnumber that wrought by Russia's 'hard' imperialism.

I would be happy to talk about Ukraine. And we have had a discussion about the "neonazism" of the current administration, which actually was (and is) one of the main lines of the Kremlin. — ssu

So far you've just followed the Western propaganda. I've given solid evidence about Ukraine's human rights record, arms dealing, corruption, and oppression and you've come back with nothing but bluster. Yes all those things are worse in Russia, but stay focussed on the argument. It's is not "who's worse?" it is "which course of action is least harmful?"

Here, it matters how bad Ukraine is because it matters how much worse it would be if parts were run by Russia, it's not enough to simply point out that it would be worse, because that's ignoring the costs of getting there.

I have simply said that as Russia has attacked independent Ukraine (and not vice versa), Ukraine should get the military hardware it needs to fight on itself to defend itself. — ssu

Exactly. That's literally saying that. You're only prepared to accept one solution no matter the cost. The mere fact that Russia is the invader is not sufficient to support the claim that Ukraine being militarily supported to fight back is the best response. Other responses might bring about less harm overall.

First issue would be for Russia to seek a cease-fire and for what I know, they are still trying to take more of Ukrainian territory. — ssu

Not necessarily. Ukraine could seek a cease-fire. The US could tie future military support to Ukraine's seeking a ceasefire. The US could lift sanctions if Russia seeks a ceasefire... There's loads of next steps that could be taken by agents other than Russia.

Again, this is the problem with your approach, you keep laying the solution at the feet of the one agent in all this that we can be pretty sure doesn't give a shit about human well-being. Why would you take that approach? The one party we can basically ignore in our strategising is Russia because it's run by a psychopathic autocrat, so they're not going to do anything we suggest. We might as well leave them out of the 'first steps' entirely. If anything is going to be done get out of this mess it's going to be Ukraine, the US, or Europe doing it.

if the war would be stopped now, do notice the bleak situation where Ukraine would be left. First of all, if it wouldn't be the Ukrainians themselves being OK for a cease-fire, but the West demanding Ukraine to a ceasefire and cessation of operation, that would be damaging. If Ukrainian leadership comes to the conclusion that they should accept a cease-fire with Russia holding all territory it has now, it's up to them. Not the West. In this situation they would have of their citizens under the rule of Russia, which they do not want. Nobody would invest in Ukraine as the conflict could spark again at any moment. For Putin the war would be a success, and he could finish the job once he has restocked his weapons and munitions. How it would be viewed is that even if the operation didn't go well at the start, it was successful thanks to Russian persistence and the utter weakness of the West. After all, in Russian propaganda the West is faltering on collapse. — ssu

No one is suggesting a ceasefire (with territorial losses) would be a picnic. It would be awful. You're just not even considering how awful the alternative is.

So my understanding of “propaganda” is not based on such broad understanding. And from my definition, I don’t do propaganda. You do. — neomac

It's remarkable that you think you can write this. Do you really read that back and think others would read it as anything other than self-serving delusion. You're literally saying you've chosen your own personal definition of 'propaganda' to make your argument right.

Second, the claim that neither intent nor one-sidedness can be proven is not part of the definition of “propaganda” you offered, and no argument has been offered to support such belief. — neomac

I didn't think one was required. Intentions are private thoughts and cannot be examined or identified by a third party because no-one can mind read. There.

The problem is that if one-sidedness and intentions can not be proven, then how could anyone possibly understand and learn how to apply the notion of “one-sidedness” and “intentions”? — neomac

What? the notion of intentions doesn't require us to always, or even ever, know what those intentions actually are. I don't need to know your memories to know that you probably have some.

These notions must be shareable, reusable, and have contrastive value to be meaningful. — neomac

Of course they're shareable,m but they're not determinable. You cannot determine what my intentions are. You can theorise about them, but then other competing theories will have equal plausibility and you have to choose between them, which is the interesting matter for discussion.

if there are biases you see in my views you must be able to show them in concrete cases by using a notion of bias that is shareable, reusable and contrastive wrt what is not bias. I’m still waiting for you to do that though. — neomac

No. You're not 'waiting' you're ignoring. I've talked extensively about position which are held because of biases in fundamental beliefs that are unexamined. You then use this "Oh, you've never shown any" rhetorical trick any time you're stuck. It's like the other classic where people wait a few pages and then claim I've not provided any sources. Or to quote your good self on the matter...

I quoted and argued your claims considering what you actually said about them in past comments. And precisely because I did it already, I don’t need to repeat them again every time — neomac

...

if motives are “open topics for debate” why shouldn’t I speculate about them? And if intentions can’t be proven as you believe (but I don’t), what else can I do other than speculating about them? — neomac

Speculating about intentions is[not what I opposed. Read what I've written, it's in the quote you responded to.

quote="Isaac;792753"]Either our motives are open topics for debate, or they aren't. In the latter case, stop speculating on mine. In the former case, you've got to give me more than just your say so as evidence.[/quote]

it’s not enough to say that I’m biased and that I commit cognitive mistakes IN GENERAL, you need to show that to me in concrete cases by using shared, pertinent, reusable rational rules (e.g. fallacies) as much as I do when I rationally examine your claims/arguments. — neomac

Again ...

I quoted and argued your claims considering what you actually said about them in past comments. And precisely because I did it already, I don’t need to repeat them again every time — neomac

...

if you keep saying that we do not share the rules of such rational examination you are going to be unintelligible to me. You would take yourself by your own initiative out of the pool of potential rational interlocutors to me, no matter how many times you keep repeating I’m biased. There is no recovery from this. — neomac

Yes. That's why discussion on actual matters of fact are pointless if you disagree about how matters of fact are to be assessed.

It’s not impossible for the majority to win out— it’s just a matter of effort and time. — Mikie

Your optimism is admirable. I'm afraid I don't share it.

My theory has the additional benefit of actually explaining why it is that we sometimes use different words to describe what we see, despite the shared external stimulus — Michael

We're sometimes wrong.

What exactly is missing in that theory such that a new theory is required which posits phenomena for which there is no other evidence?

We can argue over whether this thing is some non-physical mental phenomena like "qualia" or simply physical brain activity, but we need to at least agree that something different is going on in our heads to explain the different descriptions. — Michael

Yes, we can agree on that. In one case the path goes object>x>x>x>"blue" and in the other case object>x>x>x>"black". There is evidence for those Xs being different neural states and their outputs (different between people and between the same person at different times). There's no evidence that those Xs are some consistent 'experience of blue' that's unique to each person, it's just not there, and people have looked.

It is certainly insufficient to argue that it is just the case that we use different words, and that there's no further explanation as to why this is. — Michael

Indeed. There's all sorts of reasons why we might be mistaken.

And I don’t understand how you think you can gaslight me into rejecting the reality of my first person experience. It is the foundational truth upon which all my other empirical knowledge rests. — Michael

I'm not "gaslighting" you. You're arguing for the elf-evident existence of something I don't feel on the basis that you feel it. That's absurd... unless you do go down the solipsist route...

I can see red things without saying so. I can lie about seeing red things. There are no rational (or empirical) grounds for me to deny this about myself. — Michael

It's not about limiting your range of responses, it's about which response clarifies 'red'. You thinking of post-boxes might be a response to seeing red, but it doesn't clarify you colour response (you might have been reminded of the shape). You saying the word "red" is the one things which clarifies your response is to the object's light reflecting properties.

Given that your brain is inside your head and the apple is on the table in front of you, in what sense does the brain “interact directly” with the apple? — Michael

I just meant that there's no intermediate object, no 'representation' of an apple.

Then perhaps you are, in fact, a p-zombie, which would also explain your inability to make sense of p-zombies. Someone who doesn't have anything like first-person experience/qualia isn't going to understand the proposed distinction between something that has them and something that doesn't.

So at best you can argue that it's unreasonable of me to assume that other people are like me rather than like you. Maybe everyone else is like you, and I'm the only person in the world with first-person experience. — Michael

Yes. And accepting that possibility entails accepting that direct realism is possible. Which them means you'd have to explain why it is that you don't have it. Why has your brain evolved this convoluted system of intermediary representations when it's clearly not necessary to do so?

The example of the dress is one such example that cannot be explained away the way you do here. — Michael

The light source in the image is exactly the definition given for the dress discrepancy. The leading theory is that some people assume it is in daylight and so make a mental adjustment, others assume it is in orange, artificial light and so adjust that way. We don't go around assuming objects actually get darker whenever they pass into the shade, we adjust.

Yes, and different private experiences are the best explanation for the different responses. — Michael

I can see that it's your preferred explanation. I don't see any argument as to why it's the best. You seem to have introduced an element into the hypothesis which is not required. That's traditionally seen as a worse theory, not a better one.

It is reasonable to assume that most people are being honest and that, like me, first the dress appears to have certain colours and then (if they choose) they describe the colours. — Michael

It's not 'reasonable' at all. It don't understand from where you're getting this assumption that assuming the world to be the way you think it is is reasonable, but for others to disagree isn't.

I doesn't seem to me to be that way. I don't recognise this 'appearance' you claim is so obvious to you. So either;

a) I'm lying - in which case people do lie about their world views, in which case you may be too.
b) There's something wrong with my brain - in which case it's perfectly possible for brain to interact directly with the world, and so no reason to think indirect realism is necessary.
c) 'Private experiences' are built post hoc. You're convinced by your story, I'm more doubtful of mine.

You're talking to someone who disagrees with you about these 'private experiences' and yet are wanting to use their apparently self-evident nature as evidence. It's directly contradicted by the fact that I don't feel that way.

in normal lighting conditions objects which reflect a certain wavelength of light always appear to have a certain colour to me, and always appear to have a certain colour to you, and as children when shown such objects we are told that it is blue, and so we come to associate the word "blue" with the colour we see. — Michael

But you were arguing earlier that even language-less creatures see colours. Now you're saying the only reason we're the same is that we were taught the language. You're using our response again (saying 'blue') and then just inserting this other element (a colour experience) in between the actual light and our response to is without any need for it to be there.

The fact that two people, fluent in English, describe the colours of the dress differently is evidence that the colours the dress appears to have to one are not the colours the dress appears to have to the other. — Michael

Again, all you're showing evidence of is responses.

Why is it, do you think, that when shown the actual dress in normal lighting conditions the overwhelming majority of people will see that it's blue and black. What explains that extraordinary convergence?

Direct realism has it like this...

Dress << Response

Indirect realism wants to introduce...

Dress << {representation} << Response

All you keep providing is evidence that people's responses are different. That has no bearing at all on the differences outlined above. People might respond differently to the same representation of a dress too. Different responses doesn't tell us anything about what it is people are responding to.