I figured as much, that you were intending “I can’t do this and that”. It would be semantically nit-picky, if we were talking about anything except the logical law, which doesn’t allow any ambiguity.

Aristotle would shake his head in great dismay to hear us talk of “two different that’s”.

The question then arises, how is it possible to distinguish whether concepts arise from different mechanisms, or concepts arise by degree from a simpliciter, or, divide into simpliciters from a whole, from one mechanism? — Mww

By comparison. — creativesoul

It’s a fine line between comparison and relation. It suffices to say comparison, when the rational chronology is from knowledge backwards to perception, insofar as knowledge may or may not compare one-to-one apodectically with the object. In the case of that chronology from perception forward to knowledge, which is the major concern of reason anyway, wherein all procedural methodology is strictly a priori, the much more general relational operative is better suited for deducing precisely what the object is, out of the manifold of possible objects it might be. — Mww

So, we're talking about methodological approach here, aren't we? I find some of the language quite unhelpful. Unnecessarily complex.

If the question is how is it possible to distinguish whether concepts arise from different mechanisms, or by degree from a simpliciter, or divide into simpliciters from a whole, aren't we asking two questions about our candidate?

What does it consist of?

What is it existentially dependent upon?

Another broader point, I think, is that in order to determine a relation(or lack thereof) between two concepts, in order to even make a comparison between a plurality of things, we must first know the answers to the aforementioned questions. That is required prior to being able to determine the relation, or perform a comparative analysis.

So, we're talking about methodological approach here, aren't we? — creativesoul

Dunno about anybody else, but I am. No one denies that humans do things, but examination of methodology is required for understanding how it is possible those things are done. If there is an interest in it, of course.
—————

If the question is how is it possible to distinguish whether concepts arise from different mechanisms, or by degree from a simpliciter, or divide into simpliciters from a whole, aren't we asking two questions about our candidate? — creativesoul

Not so much about the candidate, for the candidate, in this case, a concept, is presupposed. In effect, we are asking two questions about possible mechanisms which serve as sources of the candidate.
————-

If the question is how is it possible to distinguish whether concepts arise from different mechanisms, or by degree from a simpliciter, or divide into simpliciters from a whole, aren't we asking two questions about our candidate? 1.) What does it consist of? And 2.) What is it existentially dependent upon? — creativesoul

Because the candidate is presupposed, it’s existence is given but that of which it consists is at this point, irrelevant, because the source that facilitates its existence has yet to be determined.

1.) Without a source, there is no indication of what kind of concept it is, nor what it is used for, which makes its constituency moot. Not to mention, there is no indication concepts have a constituency, even if they absolutely must have a source. After the source is identified, it is the purview of the logical laws of rational thought, as to whether or not a concept has the capacity to do its job as reason asks. Whether or not the concept is drawn from the correct source, is the determining factor of its employment, not that of which it consists.

2.) That which a concept is existentially dependent upon, is the entirety of the argument, is the only potentially informative question to be asked about concepts, for from that, both 1.) and 2.) are answerable.

Back to methodology. Reason the verb, is of course, a procedural methodology. A speculative procedural methodology predicated on logical relations. Logical relations having to do with what we think about the world as it appears, with respect to the world as it actually is.

All that to say this: you’re on record as saying we are too far apart in our thinking, so.....enough is enough, right?

All that to say this: you’re on record as saying we are too far apart in our thinking, so.....enough is enough, right? — Mww

We've had several discussions that widened the gap. That's not to say that every one will. I'm more than willing to continue this one. Are you?

2.) That which a concept is existentially dependent upon, is the entirety of the argument... — Mww

Is it though?

How can we know what a concept is existentially dependent upon if we do not know what a concept consists of?

I'm more than willing to continue this one. Are you? — creativesoul

Sure. I’m old, retired and lazy. Perfect for philosophical musings. Have to acknowledge, however, we’ll bore the holy bejesus outta the physicalists, anthropologists, empirical psychologists in the audience, as well as the subjective idealists and phenomenalists. Hell...just about every -ist ever invented, except maybe empirical realists, and even those guys are apt to shrug off most speculative metaphysics.

Maybe. :smile:

New thread? "What are concepts?" perhaps as a title?

The SEP begins with this...

Concepts are the building blocks of thoughts.

Of course, I strongly disagree!

How can we know what a concept is existentially dependent upon if we do not know what a concept consists of? — creativesoul

Would it matter, if concepts don’t consist of anything? What does a notion consist of? An idea? Other than to say what each of those does, or from whence they arise, what else can be said about them? If objects are predicated on the concepts that characterize them, how is it possible to characterize the concept, except with another concept, which tells us nothing about the constituency of concepts.

What do you think is the constituency of concepts?

What do you think is the constituency of concepts? — Mww

Thought and belief. Correlations drawn between different things.

OK, so concepts correlate different things. What are those things? And how do concepts correlate them?

Concepts do not correlate different things. Thinking and believing creatures do, and they do so by virtue of drawing correlations and/or associations between different things. Concepts are developed by virtue of drawing correlations between names and referents.

You may be interested in this as well. It has everything to do with our earlier exchange, and it ties directly into a much earlier conversation in the Kripke thread(reading group).

Oh, ok. We draw correlations and concepts are the results of the correlation of names and things named. Does that mean the thing and the named thing are things of thought or things of belief then?

Could you rephrase the question? I cannot make much sense out of it.

Never mind; extracting information in the wrong direction, kinda.

I agree we do correlate things. And such correlation develops out of thoughts. I hesitate to call such correlation the development of concepts.

I agree we do correlate things. And such correlation develops out of thoughts. I hesitate to call such correlation the development of concepts. — Mww

Concepts begin with naming. They gain in complexity by talking about the referent using it's conceptual identity(name), for doing so is to multiply the correlations drawn between the referent and other things.

Simply put, there is no difference between one's concept of a dog and one's thought and belief about dogs.

there is no difference between one's concept of a dog and one's thought and belief about dogs. — creativesoul

This is agreeable, but doesn’t say anything about what concepts consist of, which is the base of the dialectic. And to say concepts begin with naming only kicks the constituency can down the thought/belief road, for now we have not only concepts presupposed as extant with respect to their source, but also names presupposed with respect to their concepts. If concepts begin with naming, the names must already have occurred somehow, in order for them to begin the correlations from which concepts develop.

Try this: concepts do not begin with naming, but end with it. This way, the presupposition of names is eliminated, as well as their constituency, because the concepts are the names. Now, all that remains is the actual correlation, that which a conceptual identity relates to.

And the source of them. Can’t neglect that. Probably the most important aspect of this whole thing.

Done for the night. Been real.

:smile:

Cheers. Til next time.

The SEP begins with this...

Concepts are the building blocks of thoughts.

Of course, I strongly disagree! — creativesoul

Because you hold the reverse, that thoughts (and beliefs) are the building blocks of concepts? Is this what you mean by......

What do you think is the constituency of concepts?
— Mww

Thought and belief — creativesoul

Step 1: reconcile the chicken/egg temporal dichotomy. Do we think, thereby develop concepts to justify the thinking, or do we conceptualize, then think by means of them.

Eneenie meenie miney moe.....

You've proposed this in past. It's not a good characterization. It's not just a matter of which came first. It's also matter of what both consist of. Correlations drawn between different things...

So, it's not so easily characterized as chicken/egg. Eggs came first though, as far as that goes. Chickens evolved from reptiles. Reptiles lay eggs.

The evolution part is important to keep in mind. One's theory of mind(thought and belief) must be amenable to evolutionary progression.

It's not just a matter of which came first. — creativesoul

It certainly is if there’s no way to tell which one of two or more somethings came first. How are we supposed to keep in mind evolution is important if it isn’t just a matter of which came first. How can there be said to even be any evolution if the matter of a first, and thereby a succession in time, isn’t resolved?

This I think is only important if we think concepts consist of something other than just other concepts. Actually, I guess it could get real muddy, depending on the scope of reductionism being played with.
—————

The evolution part is important to keep in mind. — creativesoul

There is an inductively quantitative evolution, as major name concepts multiply in complexity by a compendium of minor names inhering in the same phenomenal object, which always comes first. There is deductively qualitative evolution, as the procedural method itself reduces from the compendium of possible named identities for a phenomenal object to a particular named identity judged as belonging to it, which always comes last.

There’s your evolution!!! Happy now? Simultaneous bi-directional evolution. Betcha never saw that one coming, didja!?
—————-

Nevertheless. You brought up evolution, so.....lay it on me. What do you consider evolving and how do you consider it evolving?
—————

I would also propose that some of our concepts are capable of describing and/or pointing towards that which existed in it's entirety prior to our reports. — creativesoul

Of course, no argument here. Their names are in the literature, if one knows where to look. Do you have names for them of your own, or from some other literature?

I call our reports cognitions. Will you agree reports are at the end of the cognitive chain? Or for you, where are reports located? Where....when.....does a report manifest?

This article gives an overview. — Isaac

Super interesting article.

This can be seen by expressing the free-energy as surprise
plus a [Kullback Leibler] divergence between the recognition and conditional densities. Because this divergence is always positive, minimising free-energy makes the recognition density an approximation to the true posterior probability. This means the system implicitly infers or represents the causes of its sensory samples in a Bayesoptimal fashion. At the same time, the free-energy becomes a tight bound on surprise, which is minimised through action. — from the article

If I've read the article right, we have that actions and anticipations are both generated to minimise predictive surprise of their motivating model. In turn, actions and internal states are sampled from the internal model and used for this predictive task, then re-incorporated as representations (sufficient statistics) of previous states (specifically their parameters). Predictive surprise and entropy are tightly related, and entropy minimisation is linked to energy expenditure; it makes sense that the relation between our embodied minds and their environment would be an entropy minimising process in a non-incidental manner, as that's an efficient solution to acting in accord with environmental and bodily regularities - utilising both to act.

How this is postulated to work in practice is set out in Box 1.

These is an overall process of external states, sensations, internal states and actions.

External states at time t: are samples from a model of form M1 based on previous external states at t-1 and actions at t-1 with some error.
Sensations at time t: are samples from a model of form M2 based on external states at time t and actions at time t-1 with some error.
Internal states at time t: are samples from a model of form M3 based on internal states at time t-1 and sensations at time t. The specific model of form M3 which is formed at this stage is an approximate minimizer over all models of type M3 with respect to criterion C1.
Actions at time t: are samples from a model of form M3 based on internal states at time t and sensations at time t. The specific model of form M3 which is formed at this stage is an approximate minimizer over all models of type M3 with respect to criterion C2.

(Edit: something interesting here is that precisely what counts as a "time step" is just... one part of the process feeding into another, the paper doesn't write it out like that, it does it in terms of dependency arrows.

Edit 2: another thing presenting it as time steps under emphasises is that a perturbation can intervene at any one of the processes and will propagate its effects through the arrows; if a perturbation happens in, say, the sensations component, it will propagate to the internal state and action components without the intermediary action step, but also with mediation through the action step. Imagining, say, holding an electric fence makes the hand clench up, it might well propagate to the actions component as the clenching reflex but also the internal state as "i'm being electrocuted" or "what the fuck")

Criterions C1 and C2 are functions of model type M3; specifically they minimise the free energy.
In both, there is a representation of the previous internal states and the current external states, a "recognition density", that represents the causal structure of the environment in terms of change propensities given interventions (our actions and environmental activity); this is called Q. Then there is a representation of (the representation of) the previous sensations and the representation of the current external states, which is like a current 'self model' (our anticipations and memories and presence at the time), this is called P. Both criteria add the mismatch of P from Q (the kullback liebler divergence) to the surprise (informational discrepancy) of P given the previous self model.

C1 minimises (the above paragraph) over the panoply of internal states that have previously been sampled; an adjustment of our internal state to the current situation representation.

C2 minimises (the above paragraph) over the panoply of suggested actions that have been previously sampled; an adjustment of our behaviour to minimise the distinction between our model of the causal structure of our environment and our self model.

Our self model updates through a variational inference procedure.

The proposal I gave would have there being more than one functional form in the M3 class; and the functional forms would interact. Less parsimonious, messier. They may be consistent with each other depending on whether the models in M3 factorise over input sources; like partitioning a likelihood into different factors (say, a model for people in Africa and people in America for birth rates, the two being different; analogously a model for visual information and proprioceptory information being independently weighted).

What both emphasise is that our minds realise from a "set of available minds" consistent with its history.

It seems like you're asking about things having properties sans experience, which is fine, but presumably you're not just saying there's no experience period, are you? (In other words, you'd just be saying that there is no consciousness/no conscious beings, etc.) — Terrapin Station

I'm not asking about things having "properties sans experience": I would say that properties are only known via cognition. To speak about properties "outside cognition" would be to posit that noumena have positive attributes. That would be problematic, but in any case was not my concern.

Of course I would not deny that things appear to us, but the idea of experience seems to arise as a reification of the abstract generalization which is thought of as the process of appearance as such.

I also wanted to portray the relevance of embodied cognition to conception; so using concept in both places seemed appropriate. Conception still leverages body stuff that we usually take as conceptually unrelated to it. — fdrake

Yes, it seems reasonable to think that animals, to varying degrees, do something like what we call imagining and thinking, and this seems justified since human imagination and thinking is not strictly dependent on language, but also consists in multi-sensory processes of "visualization" or image-forming.

I guess we could also call this pre-linguistic activity conception, as distinct from language-based conceptualization. So, I think that it's reasonable to say that animals conceive, but that they do not "have concepts", because the "having" relies on the hypostatization that language enables. Something like that anyway

Why is it unhelpful to have an understanding of conceptual architecture which does not distinguish us as well from animals? Or conversely, what is it about distinguishing us from other animals that is so helpful it must be maintained in any understanding of neural systems? — Isaac

Just as I say to fdrake above, I think it is more helpful to maintain distinctions between linguistically and culturally elaborated conceptualizing capacities, and the primordial somatically-based embodied cognition we share with animals.

It's not just a matter of which came first.
— creativesoul

It certainly is if there’s no way to tell which one of two or more somethings came first. How are we supposed to keep in mind evolution is important if it isn’t just a matter of which came first. How can there be said to even be any evolution if the matter of a first, and thereby a succession in time, isn’t resolved?

This I think is only important if we think concepts consist of something other than just other concepts. Actually, I guess it could get real muddy, depending on the scope of reductionism being played with. — Mww

Indeed, it could get muddy, but it need not. It is most certainly quite complex, but there's also a very basic simplicity about it, as must be the case given the evolutionary progression of thought, belief, and concepts. It's not all reduction, though. Not on my view anyway. I mean, language is recursive, so that must also be taken proper account of. The framework we offer must take all sorts of things into account.

Of course I would concur with the need for a timeline. As you've asserted here keeping evolution in mind requires some sort of succession. With all that in mind...

What I meant by "it's not just a matter of which came first", was that that is an gross oversimplification of the methodological approach needed in order to even be able to acquire the knowledge we're seeking to obtain here. I think you'll agree with this? Knowing which came first requires knowing what all thought, all belief, and all concepts consist of. For when we know what each consists of, it offers us solid ground to be able to deduce which came first, by knowing what each is existentially dependent upon.

For example, if all A's consist of B, then no A exists prior to B. If all A's consist of B, then each and every A is existentially dependent upon B. That which is existentially dependent upon something else cannot exist prior to that something else. These are the sorts of reasoning that come into play here.

If you could be so kind, I would like for you to confirm whether or not I adequately understand what you've offered here. I've broken it down for easier reference. I don't think that I've broken anything up that needs to stay together, if you know what I mean. It's an interesting take, if I understand you correctly. There's some agreement, I think...

:smile:

There is an inductively quantitative evolution, as major name concepts multiply in complexity by a compendium of minor names inhering in the same phenomenal object, which always comes first. — Mww

The above seems to be referring to the quantitative increase of our concepts via naming practices. I take it mean something like our concepts increase in quantity along with the number of names we use. Specifically speaking, you seem to be also claiming that there is a hierarchy of namesakes involved within our use of concepts. I take this to be referring to all the different names of all the different features of a referent(the same phenomenal object). So, for example, if I have read you correctly, the major name concept could be "tree", and the compendium of minor names would include all of the names for the features, properties, and/or attributes of the tree. "Leaves", "roots", "trunk", etc. would all qualify for being in this compendium of minor names.

Do I have that much right?

What I'm left wondering still, is not only what exactly is it that you're claiming "always comes first", but "first" - as in prior to what else? I want to say that the primary namesake comes first, but I'm hesitant for you may be saying that the phenomenal object comes first. If it's the latter, then I would agree that some conceptions are of phenomenal objects and in those cases the object 'comes first'.

There is deductively qualitative evolution, as the procedural method itself reduces from the compendium of possible named identities for a phenomenal object to a particular named identity judged as belonging to it, which always comes last. — Mww

This bit I cannot understand. Could you set it out with an example?

:brow:

I would also propose that some of our concepts are capable of describing and/or pointing towards that which existed in it's entirety prior to our reports.
— creativesoul

Of course, no argument here. Their names are in the literature, if one knows where to look. Do you have names for them of your own, or from some other literature? — Mww

I thought we would agree there.

:smile:

Do I have names for those concepts of my own? I wouldn't say that. I am very fond of simple ordinary language when adequate. Given that we're discussing the evolution of concepts, thought, and belief, we must also keep in mind that the proposed complexity level that we're claiming exists at some specific time period must belong to a creature capable of having such complexity. I mean, in the beginning, the thought, belief, and/or conceptualization must be at a rudimentary and/or very basic level of complexity. In addition, those rudimentary thoughts, beliefs, and/or conceptualizations must consist of that which is amenable to evolutionary progression.

To directly answer your question, or at least what I think you're asking me for...

Thought, belief, meaning, and truth all exist in their entirety(on the most basic level/degree of complexity) prior to our conceptualizations/names of/for them... that is... prior to common language use.

I call our reports cognitions. Will you agree reports are at the end of the cognitive chain? Or for you, where are reports located? Where....when.....does a report manifest? — Mww

That all depends upon what we're reporting upon. Metacognitive endeavors, such as the one we're involved in, are most certainly "at the end of the cognitive chain". Thinking about our own thought and belief is metacognition. Not all thought and belief is metacognitive.

Not all reports are existentially dependent upon metacognition though. A young child - particularly an honest and talkative one - will offer their own report of all sorts of stuff that they're thinking about. This youngster's report is not at the end of the cognitive chain. Well, strictly speaking, until the child learns to start talking about it's own thought and belief, and it's own language use, it's at the end of what's cognitively possible - at the time - for them. However, assuming that they go on as most do, the end of the cognitive chain(metacognition) comes later.

:smile:

Nice discussion. Kudos!

:smile:

I think it is more helpful to maintain distinctions between linguistically and culturally elaborated conceptualizing capacities, and the primordial somatically-based embodied cognition we share with animals. — Janus

That's precisely the divide that needs bridged...

The SEP begins with this...

Concepts are the building blocks of thoughts.

Of course, I strongly disagree!
— creativesoul

Because you hold the reverse, that thoughts (and beliefs) are the building blocks of concepts? — Mww

Yes and no...

Thought and belief are the building block of concepts(on my view), but, that's not exhaustive enough. Correlations drawn between different things are the building blocks of everything ever thought, believed, spoken, written, and/or otherwise uttered.

Of course, I'm setting that aside in order to understand the position you're presenting, for the time being anyway. I hope to compare the two later.

Try this: concepts do not begin with naming, but end with it. This way, the presupposition of names is eliminated, as well as their constituency, because the concepts are the names. — Mww

This approach puts all concepts on equal footing as being the names. It would only follow that there are no concepts prior to naming. I could agree actually, but something tells me that you may not? My agreement to that would lead to a denial that that which exists prior to it's namesake is a concept.

There are cases in which we would find ourselves in dire need of drawing a distinction between our conception and it's referent. There's no difference between our conception of games and games. There is most certainly a difference between that which exists in it's entirety prior to our naming it and our name. There's an distinction regarding existential dependency to be drawn and maintained here...

How else do we discriminate between such pre-existent things and our names for them?

Here is where elemental constituents come to bear alongside the considerations of existential dependency...

it makes sense that the relation between our embodied minds and their environment would be an entropy minimising process in a non-incidental manner, as that's an efficient solution to acting in accord with environmental and bodily regularities - utilising both to act. — fdrake

Yes. The great thing about the free-energy approach is that it gives both a mechanism and an evolutionary story to the Bayesian modelling system which we already have a good idea the brain uses.

The specific model of form M3 which is formed at this stage is an approximate minimizer over all models of type M3 with respect to criterion C1. — fdrake

I can't find a non-paywalled version, but the paper which started all this is Ernst and Banks's 2002 paper on visuo-haptic relations. Basically, they presented a ridge-measuring task where visual stimuli were deliberately made uncertain and haptic stimuli more certain (we traditionally trust our vision more readily that we trust our touch). They modelled an estimation method using Bayes Theorem and a normal Gaussian distribution for widths of the ridge as the prior. They then tested people's actual estimates and found a really strong correlation indicating that people were somehow actually using Bayesian inference to estimate the width of the ridge in the face of visual uncertainty.

Similar experiments have been since to confirm this (Ernst and Banks didn't really set out to find this). A classic one with poor contrast moving dots where participants had to estimate the direction of movement despite not being able to see the dots clearly. Unbeknownst to them, there was a slight probability bias in favour of one particular direction. It took, I think, abut two minutes for their Bayesian model to pick up on the fact that the dots more often moved in one direction and adjust their priors for movement accordingly, even when there were really no dots at all, in fact. We do Bayesian modelling in our heads, it seems.

something interesting here is that precisely what counts as a "time step" is just... one part of the process feeding into another, the paper doesn't write it out like that, it does it in terms of dependency arrows. — fdrake

There was an experiment done on this, but I can't remember it well enough to even look it up. If I recall correctly, there are time steps involved but they're specific to the model being used (visual, auditory etc). I wish I could track down the paper, but I don't even know where to start.

The proposal I gave would have there being more than one functional form in the M3 class; and the functional forms would interact. Less parsimonious, messier. — fdrake

I'm inclined to agree. Most work on this has been done on perception and although the data correlates well, there is some discrepancies. My guess is that the discrepancy is caused by other models in other areas of the brain interfering at a sort of meta level.

Back to the topic...the sense is that our feeling of 'an experience' is exactly this meta-model trying to put some sharp edges to the whole fuzzy procedure. We can't actually work well with fuzzy data, we can't possibly 'look behind the curtain' to actually have an awareness of the variational inference procedure going on. Why not? - Well on an ecological level, we simply wouldn't behave as efficiently as a creature which had a clear answer, on an epistemic level, how would we experience that with some means of modelling it...

Really nice summary by the way.