we can 'hold something to be true'... despite our own propensity to act as though it were true.
— praxis

How would you know? — Isaac

Weren't we just talking about pro-lifers getting abortions? I suspect the inverse also occurs, pro-choicers not getting an abortion because it feels immoral, like murder.

If belief is merely a social construct then we can abandon it should the need arise... at the risk of being abandoned by the social group.

Good policy :smile:

Ah. So all this is just to day that sometimes folk say "I feel certain..." as equivalent to "I believe...", and this is distinct from "I am certain...".

Why didn't you say? — Banno

All this? Why didn't you read? Anyway, not exactly. It seems to me likely that if someone says 'I feel certain', unless they are critically minded, they will mean 'I am certain'. If they are critically minded, they probably won't say 'I feel certain' at all, but rather, 'it seems to me', unless we are talking about what is "common knowledge". I don't buy the equation of belief with 'acting as if', because I lean more towards thinking there are many things people believe, or at least entertain (which you, at least; seem to count as being the same as believing) which have no bearing on action.

It's good exercise but not a lot of meat on those bones.

Others find it filling. — ZzzoneiroCosm

Yes, good exercise, but hardly a satisfying meal. Might as well live on pills and supplements...

What we'd be arguing about is the best rendering into language of exactly the same neural network, where 'best' might be defined by the rules of rational thinking, or ethics, or just social function. — Isaac

Yes.

Interesting article. What I fond problematic is the language that describes the neural net as "representing" how things are, or as "models".

Without an internal model, the brain cannot transform flashes of light into sights, chemicals into smells and variable air pressure into music.

They are embodied, whole brain representations...

My reading of Evers and Lakomski has taught me that neural networks are not rules-based. Here's an excerpt from a Master's theses I wrote on the topic:

Evers and Lakomski give several examples of ANNs in use (Evers and Lakomski 1996, pp. 122 – 127). Their great strength is in pattern recognition, for example in differentiating between a sonar image from a rock and a mine; or in recognising valid and invalid inferences in propositional calculus.
ANNs function in a way that is somewhat different from a typical computer – what is called a “von Neumann Machine”. A Von Neumann machine has a central processor and a memory, and executes the instructions in the memory one at a time in sequence. ANNs do not make use of memory in this way, instead storing the “instructions” in the weighting of the connections between neurons. However, ANNs are subject to the same rules of computation that apply to all computational devices – they are limited Turing machines. “The recurrent perceptron network… is (yet another) form of a Turing machine”. (Hyötyniemi 1996, parenthetic comment in original)
Their advantage, usually described as “pattern recognition”, is that they are able to represent what engineers call “hard problems” (Works 1992). Hard problems are those that are non-linear; or more specifically, that cannot satisfy the general equation
F(aX + bY) = aF(X) + bF(Y)
Complex systems, such as the Mandelbrot set or turbulence in a fluid, are not solvable using this linear equation. Complex systems vary so greatly that there are no general mathematical tools that can be used to analyse them. But non-linear systems can be modelled by the weightings of a neural network
“We can learn a great deal about complex non-linear systems even though we have no comprehensive mathematical tools. One way to describe this knowledge is to state rules describing the behaviour of a system. This is the approach taken by the author of the textbook, or a knowledge engineer writing an AI program. Another way is to map significant states of the system into some more compact internal representation so that later occurrences of these states can be recognized. This is the approach used in teaching by example, or by an ANNs.” (Works 1992, p. 37)
What form does this representation take? While it is not in the sequential form of a Von Neumann machine’s instructions, it can certainly be represented. The weightings of the neurons are given the indexes pi, qi, ri. The output of the network is given by the equation:
{image of matrix}
This example, from Paul Churchland (Churchland 1988), is itself used by Evers and Lakomski (Evers and Lakomski 1996, p. 122). The status of the artificial neural network can be stated precisely by such a matrix. Training an ANN is simply adjusting the values of pi, qi, ri until, for a desired set of inputs x, y, z, the desired output a, b, c, d is achieved. This is done by repeated use of a function called the delta rule (Picton 2000). The delta rule does no more than average the errors for each weighting, in order to find an optimal setting given a particular training set.
It is important to note that it is possible to represent the whole of the workings of an artificial neural network using this mathematics. Therefore it is possible to represent the function of the neural network symbolically (or linguistically, as Evers and Lakomski prefer to call it). The claim that “this would be a linguistic representation that the network itself makes no use of” (Evers and Lakomski 1996, p. 122) is difficult to understand, since pi, qi, ri, which are plainly linguistic representations, plainly are the values given to the weightings that the network indeed makes use of. These are precisely what each neuron uses in order to function, represented linguistically.
Again, in Doing Educational Administration they claim that
“…neural network representations of language raise the possibility that tacit knowledge of language is not rules based at all. For neural nets are not rules based.” (Evers and Lakomski 2000, p. 30)
But we have seen that neural networks are clearly rules-based systems, by explicating how those rules can be represented in a matrix.
Evers and Lakomski wish to use this to show that “all ‘knowing that’ is really ‘knowing how’”, taking their model for ‘knowing how’ as neural networks. We can just as easily claim, since neural networks can be represented by a rule, that ‘knowing how’ is actually ‘knowing that’.
However, it does seem evident that the description of the function of a neural network in a matrix is post hoc. The values that are assigned to the matrix are determined by running the network, and this appears to be the only way of determining these values. So although the values are indeed determined by a set of rules that are in their nature linguistic, those values are a result of the application of the rule. The know-how embedded in a neural network can be turned into theoretical knowledge in the form of a matrix only after the network has been run through a training cycle. In this sense, theoretical knowledge is post-hoc. — Banno

That was written almost twenty years ago, so my recollection of it is not all that clear, but the upshot seems to be that while the output of a neural network at some time can be set out symbolically in a suitable matrix, this can be done only post-hoc.

Now this seems to me to tie in nicely with §201 of Philosophical investigations: that there are ways of understanding a rule that do not involve stating the rule, but enacting it. In effect the matrix states the rule that the neural network enacts.

So the language that describes the neural net as "representing" how things are, or as "models", might be misinterpreted as implying that the neural network somehow symbolises the sounds or images it is processing. There is no symbolic representation within the network.

To be sure, I'm not saying that Feldman Barrett does this, only that it is a possible, erroneous interpretation of what she says.

Any chance that could be translated into English?

Putting the anal back into analytic. — Banno

:snicker:

1. All knowledge is belief
2. Not all belief is knowledge i.e. some beliefs are not knowledge (missing justification)

Weren't we just talking about pro-lifers getting abortions? I suspect the inverse also occurs, pro-choicers not getting an abortion because it feels immoral, like murder. — praxis

But how would you know they actually believed in pro-life/pro-choice beforehand...if they committed no action 'as if' they did?

If belief is merely a social construct then we can abandon it should the need arise — praxis

On the contrary, social constructs are some of the most stubborn bits of 'mental wiring' we have. When we say "X is a social construct" we just mean that it is not a biological/physical feature, that society is instrumental in maintaining it by regular feedback. This has little bearing on the ease with which it can be abandoned, some are easy some are hard. What really impacts the ease with which a belief can be abandoned is the degree to which it is embedded with other beliefs rather than its origin.

I suppose the point is that a construct can be seen as a construct and be accepted and entertained provisionally for pragmatic reasons; no believing needed.

Yes, you're absolutely right. One of the main criticisms of the model-based approach is that it is too representational, but I do think that critique misplaced.

The trouble really is translating folk understandings into something cognitive science can work with and unfortunately for that project there's a fundamental problem in that brains don't work like anything we have any folk understanding of such that we can give some comprehensible analogy. When we talk about a 'mental model' of, say, my belief that the pub is at the end of the road, what people most often imagine we're talking about is some part of the brain which is wired such that when "where's the pub?" is entered into one end "at the end of the road" comes out the other, but unfortunately that's not the case. Rather than holding their function in 'wiring', neural networks hold their function in behaviour. They are not 'wired' such as to produce the answer, they 'behave' such as to produce the answer. It's a dynamic system.

What this means is that the same actual 'bit' of brain can hold several different beliefs because the belief is encoded in how that bit of the brain behaves not how it's wired, and it can behave differently at different times. It creates a much more responsive system.

So the question is how some given cortex knows how to behave, how is it that when I try to imagine where the pub is (or start walking to it) I always picture the end of the road? At the moment, the leading answer is in dynamic causal modelling - basically the output of one neural network can be to shape another's function. there's a good paper on it here, if you're interested https://www.sciencedirect.com/science/article/pii/S1053811919310444?via%3Dihub.

The problem is then how to translate that back into any folk understanding of something like 'believing the pub is at the end of the road' or 'being angry', both of which are dealt with by models in the brain, but not in a one-to-one representation.

My preferred method, that adpoted by Barrett, Friston etc is to put an intermediary step in the translation where we talk about 'models of...' between the world we're trying to talk about (the actual pub) and the means by which the brain gets us to walk to the end of the road to get to it (the dynamic neural networks responsible). I think it's helpful because it allows for actual computations of Bayesian updating which does seem to be how the networks are best represented (see https://www.nature.com/articles/s42003-021-02994-2). Others, such as Varela, Hutto, Clark reject the whole use of representational models at all, which I think just ends up nothing but 'just so' stories...but that's just me.

Having said all that, your piece seems to be about language, and although I'm sure it'll be very similar, my experience with all this extends only to belief (mainly behavioural), and a little on perception (about objects) so it may be that language has not need for the whole 'representational model' intermediary step.

I suppose the point is that a construct can be seen as a construct and be accepted and entertained provisionally for pragmatic reasons; no believing needed. — praxis

What would 'accepting and entertaining' entail that did not constitute a belief?

Having read your post to Banno, you appear to have a wildly different conception of belief than I do.

Having read your post to Banno, you appear to have a wildly different conception of belief than I do. — praxis

It seems that way, but I'm still not sure what your conception is.

Can you make sense of, for example...

"I believe the keys are in the car, but I might be wrong, I can't remember"

"John believed the door was locked, so he escaped through the window"

"We all believe larger objects cannot fit inside smaller ones"

"He was walking the wrong way, he seemed to believe the pub was at the top of the hill"

What expression would you replace the use of 'believe' with in those sentences that would be synonymous?

'Certain', or 'felt certain' doesn't fit in 1, 2 or 4.

A pointless comment eliciting this otherwise pointless response. — Janus

Pointless to you perhaps but that in no way makes my point pointless, just inconvenient for your point.

It’s like the exchange about the Sun rising or setting when in fact it does neither, it’s the Earth that turns!
Believe it!

Right on Janus. We are not dealing here with certitudes. We are dealing with numerous instances of carelas speach and badass, ambigous words and expressions.

Why don't you consult the most accurate and knowledgible of all scientests. The linguists.

Just look it up in the dicionary

You are slicing up the meanings of an inherently confusing batch of words into tinier and tinier shades of meaning. You are slicing them up into such teeny-weeny shades of meaning that you are going to end up with hamburger.

I am not saying I am against all this. On the contrary I love hambuerger.

I think there's a difference between expressing a degree of certainty and expressing an intention to hold something to be the case.

Take these two statements for instance:

"Having studied the subject a bit, I think that Democracy is the best form of government for the people that I know of, at least when it has adequate supporting institutions, checks on power, etc."

"I believe in Democracy."

The latter statement seems more like an expression of faith.

Pointless to you perhaps but that in no way makes my point pointless, just inconvenient for your point. — universeness

Right on Janus. We are not dealing here with certitudes. We are dealing with numerous instances of carelas speach and badass, ambigous words and expressions. — Ken Edwards

Universeness, if your point is that same as Ken's, quoted here, which I agree with, then it is not "inconvenient for my point" at all. I'm still not seeing the relevance, so if you really have a point and want me to get it, you will need to explain.

,

In what is called"folk psychology", or less prejudicially "intentional discourse", belief is used to explain behaviour. Why did Isaac walk up the hill? Because he wanted to go to the pub; and, he believed that the pub was at the top of the hill. That's sufficient for most of our purposes, and is the prime exemplar of what belief is about.

Neuroscience is looking for a different answer. Why did Isaac walk up the hill? Because the neural patterns in his brain caused his limbs to move in various complex ways.

Here's two descriptions:

Painted using a matte house paint with the least possible gloss, on stretched canvas, 3.5 meters tall and 7.8 meters wide, in the Museo Reina Sofia in Madrid.

An anti-war statement displaying the terror and suffering of people and animals.

Both are of Picasso's Guernica. Somehow matte house paint on canvas is the very same thing as a powerful anti-war statement. How those shades of grey do this is a complex business.

...an intermediary step in the translation where we talk about 'models of...' between the world we're trying to talk about (the actual pub) and the means by which the brain gets us to walk to the end of the road to get to it (the dynamic neural networks responsible). — Isaac

I'd urge caution here. There is a notion of "model" descending from Kant and used in philosophy of mind, and by association in cognitive science, insisting that we only see the world as it is modelled by our brains. That view skips significant steps, the core of which Isaac captured here:

Rather than holding their function in 'wiring', neural networks hold their function in behaviour. — Isaac

I might phrase the point as follows. We do not see the model; rather, the model is our seeing the world. It's not that brains construct models and the mind sees the model rather than the world, but that seeing the world can be described as constructing a model.

Hence, in Davidson's terms, the world is always, already interpreted.

All this by way of avoiding homunculi...

And it is the reason of my interest here:

↪praxis That's more interesting. Did you parse these into actions on purpose? — Banno

You are slicing up the meanings of an inherently confusing batch of words into tinier and tinier shades of meaning. You are slicing them up into such teeny-weeny shades of meaning that you are going to end up with hamburger. — Ken Edwards

Sure, I'm mapping out a series of related terms, on the premise that the apparent confusion stems from lack of precision.

Putting it roughly and briefly,

• One believes some statement when one holds it to be true.
• One is certain of some statement when one does not subject it to doubt.
• One has faith in a statement when one believes it regardless of the evidence.

Making this distinction allows us to see the error in thinking that it is wrong to believe anything, while maintaining the moral rejection of faith in the face of objectionable evidence. So faith in the evil of Belgians or the Nazi Party is objectionable, yet we can still believe that the pub is at the top of the hill.

I still don't see the significance.

Btw, we still have mental representations that are internal and in that way we can 'see the model'. For example, I can imagine a ten legged horse, and can play blind chess. Actually, I can't play blind chess, but I could with training.

I still don't see the significance. — praxis

Of what?

we still have mental representations — praxis

Sure, you can imagine stuff. But you are looking at your screen now; you are not looking at a model of your screen constructed by your brain.

Sure, you can imagine stuff. But you are looking at your screen now; you are not looking at a model of your screen constructed by your brain. — Banno

It's constructed in at least the sense that the screen is distinguished from everything else.

Of what? — Banno

Parsing statements into actions on purpose.

One believes some statement when one holds it to be true.
One is certain of some statement when one does not subject it to doubt.
One has faith in a statement when one believes it regardless of the evidence. — Banno

Can't a person believe a religious truth and not doubt it regardless of the evidence?

That's what I said.

I'm sorry, but I still do not have a clear idea of what you are arguing, of what your stance, if any, might be.

One has faith in a statement when one believes it regardless of the evidence. — Banno

Where did you get this definition of faith? It's in conflict with every usage of the term I've ever heard.

Anyway, I don't think that I've suggested it's wrong to believe anything.

I don't know why I would hold something to be true to myself. If I had an idea about something and it turned out to be wrong it would cause me no loss in social standing of any kind. Maybe denial to avoid uncomfortable feelings?

Augustine. Via Kenny.

Kenny?