Sometimes very active discussions about peripheral issues overwhelm a thread, so this is a permanent home for those conversations.
I’ve opened a new “Sandbox” thread as a post as the new “ignore commenter” plug-in only works on threads started as posts.
Bruce,
It’s innate, at least in chickens. See page 46 and onward in this preview.
There’s also the problem that any sentient entity cannot understand something as complex as itself.
Neil,
Sure it does. Where did you get the idea that it didn’t?
Thanks Neil.
There are several possibilities for an OP in that post and I’ll leave further commenting to them.
– nature of concepts
– Bayesian epistemology as a theory of evidence in science
– gods eye view and its relation to both everyday and scientific realism
– measurement, categories, human and animal perception
I’m not looking for anything significant in the OP text; one or two sentences to set rough bounds for on topic would be fine.
I agree with that and that is the point eg de Waal makes in the book with the Guardian review that I linked.
One issue is: how can we describe the concepts we are attributing to animals in a language which is free of human concepts? What is it like to be a bat, conceptually?
I think scientific theories can provide mechanisms for such descriptions. These mechanisms need not be language-like, of course.
Again, I’ll stop there an leave further comments for an OP thread.
Thanks. Here is one for you on Carey’s theory of human concepts and innateness.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3489495/
An important point that I did not make clear about Fodor: He thinks some concepts must be innate because any learning mechanism needs them to bootstrap from.
“Fodor (1975) argued that there are theoretical problems with all models of concept learning in that all such models treat concept learning as hypothesis testing. The problem is that the correct hypothesis invariably employs the very concept to be learned and hence the concept has to be available to a learner prior to the learning taking place”
https://plato.stanford.edu/entries/concepts/#NatAboCon
So I suppose Fodor would want to understand how your learning mechanism works and in particular what initial conditions it depends on.
That Fodor work was 45 years ago; he has an updated 2014 view in this book that I have not looked at:
Thagard’s work seems to be in this direction as well.
The Cognitive Science of Science: Explanation, Discovery, and Conceptual Change
There are also the pragmatic approaches to understanding scinetific explnations and modelling. I agree that they tend to be taken as anti-realistic. Since I am sympathetic both to these cognitive approaches and to scientific realism, I am not in equilibrium, reflectively-speaking. Possibly perspectivalist approaches to scientific realism might help me restore my intellectual balance.
https://plato.stanford.edu/entries/scientific-explanation/#PraTheExp
Section 3 introduces pragmatism regarding scientific models
https://www.academia.edu/26570119/The_Ontology_of_Models
Fair enough. Still, it would be helpful to have a published article serve as a foil for your ideas so that you could contrast your ideas to the terminology in a published paper.
Here is one I just came across while looking at the latest article made available for commentary in BBS. I’ve only looked at the first two sections. They provide a series of definitions leading up to categories. Those two sections alone would be an example of the type of article content I am thinking of:
Above and Beyond the Concrete: The Diverse Representational Substrates of the Predictive Brain
https://www.cambridge.org/core/services/aop-cambridge-core/content/view/67E56681B709473DEF64AFF253C63731/S0140525X19002000a.pdf/above_and_beyond_the_concrete_the_diverse_representational_substrates_of_the_predictive_brain.pdf
https://www.cambridge.org/core/journals/behavioral-and-brain-sciences/information/calls-for-commentary/open-call-for-commentary-gilead
Idealism is how many interpret the destination of Donald Hoffman’s journey to his theory of conscious realism, eg as here
ETA:
Do you have any non-philosophical examples of enactivist cognitive science, in particular the radical kind of enactivism eg Hutto?
As best I know, radical enactivism remains a purely philosophical program, consisting of criticisms of current scientific research programs, but without offering anything scientific as a replacement.
I’m am petruskavist with respect to that kind of philosophy.
As a contrast, consider Chomsky’s successful take down of Skinner’s psychological behaviorism. He offered a rich research program in linguistics as an alternative.
I will be working on preparing one.
I don’t think we actually need to describe the concepts that an animal uses. We really can’t describe concepts that other humans use. It should be sufficient to note that animal behavior strongly suggests that they are conceptualizing their world, even if we cannot be sure what concepts they use.
As for what it is like to be a bat? I think that question is overdone.
I can’t even tell you what it is like to be me. A statement X is like Y is reporting a comparison. We have no basis for comparison.
The reason Nagel gives for his bat example, is because the bat uses echo-location. But I don’t see a big deal about that. I expect it is very much like vision, even though it uses sound rather that electromagnetic waves. It is a way of getting information about the environment. Our conscious experience of the environment is just our conscious experience of the information.
In case you ask — no, a computer system is not conscious, even though it has lots of information. That’s because the computer only “sees” syntactic information. We use that information for our semantics, but the semantics are not available to the computer. The computer is just a mindless mechanical moron (or something of a philosophical zombie).
I added the “, conceptually” to Nagel’s bat-quote to switch the quote to refer to the conceptual issue we were discussing , not his phenomenal issue.
I think I agree with the rest of your post, subject to further detailing some other place.
ETA: I think we want to talk about concepts in animals to have an account of human concepts that is in accord with evolution of human cognitive abilities and human linguistic concepts. Similarly, we want an account of concepts in babies.
No doubt, that will involve a scientific definition of concepts, but it is then desirable to have a philosophical explanation which links scientific usage to usage in folk psychology, although that might be nothing more that Churchland-style eliminitivism.
.
I have read Fodor’s arguments.
He is right on one point. The philosophy of learning doesn’t say much about acquiring concepts. It mostly concentrates on acquiring beliefs, and often suggests induction for that. That’s a mistake. Actually, it is two mistakes. The emphasis on belief is a mistake, and the emphasis on induction is a mistake.
As best I can tell, rationalists want to say that concepts (or many of them) are innate. Empiricists want to deny that. But what they have in common, is that neither provides a way of acquiring concepts.
It is far from clear what we mean by “concept”. So maybe we should instead be discussing categories, though there is also confusion in the literature about those.
He probably wouldn’t have been interested. He argued against behaviorism, specifically that of Ryle, early in his career. And he has stayed with that view.
Fodor sees learning as acquiring beliefs. I see learning as acquiring abilities. These are very different ways of looking at knowledge. Fodor’s way is very intellectualist — the kind of intellectualism that Ryle criticized. I am, of course, referring to Ryle’s “The Concept of Mind”.
The trouble with behaviorism, though, is that much of it emphasizes mechanical behavior (physical actions). I’m far more interested in what might be called cognitive behavior or perceptual behavior. So I see perceptual learning as very important. But most of that kind of behavior is not easily visible, because it is not mechanical.
I knew that Fodor had updated his view, but I have not read that book. I did look at the Amazon page (and the “look inside”). Perhaps I will buy it, but I doubt that Fodor’s arguments will persuade me. He is (or was) a good writer.
Neil Rickert,
I decided to take a break and listen to Ezra Klein’s latest podcast. It turned out to be about most of what we are discussing in this thread.
https://www.vox.com/ezra-klein-show-podcast
We don’t just feel emotions. We make them.
No transcript, so it is helpful to have a podcast app which plays at 1.5 speed.
The interviewee explains modern ideas of embodied predictive approaches in cognitive science and how these ideas apply to concepts, categories, and emotions.
She also touches on relations to: Buddhism; encultured child development; concepts and language; intersubjective consensus versus objectivity; and even tangentially on illusionism about phenomenality.
I hope you will listen to this podcast as an introduction to the views in science and philosophy that make sense to me and then direct any criticism in your OP to these ideas. It often seems to be that your concerns are with philosophy and scientific ideas that I also reject, such as
– view of perception which ignore embodied action
– perception as bottom up building of static, perfect representations
– cognition as computation using abstract symbols, eg language of thought views of cognition
If you keep giving me things to read and to listen to, then how do I find time to write an OP?
I’m about halfway through at present.
I’m noting what she says about predictive processing. But I disagree.
I look at my clock, and then I add one hour. I am predicting what the time will be in one hour.
Suppose, however, I set the clock to be one hour fast. And then I read the clock. In that case, I’m not predicting anything. I’m just reading off a time.
Most of her examples of prediction don’t seem to me to involve prediction. I see it as the brain being wired up to measure/categorize and provide the information that we need for our decisions. But we are just reading off that information. We are not reading different values and then making a prediction. Or, at least, that’s my way of looking at it.
If X is correlated to Y, then a measurement of X can also give a measurement of Y. And the brain can use those correlations in wiring itself up to get the information that it needs.
Neil, to Bruce:
That’s another good reason for you to post one of your rejected papers. It’s already written.
BruceS,
Lisa Barrett has a book on this: How Emotions Are Made. I haven’t read it yet but my friends who work in philosophy of cognitive neuroscience recommend it very highly.
I think that’s probably fair as a criticism of Hutto per se, but there are some interesting ideas in enactivist cognitive science that seem promising. Rodney Brooks’s evolutionary robotics is enactivist without saying so, and the enactivists also inspire a lot of the predictive processing stuff. I like the promise of predictive processing for “ending the representation wars” between cognitivists and enactivists.
I should add that I think PP itself is not quite right: as a theory of cognition it is both too rationalist (because priors and hyperpriors are constrained by learning and evolution, not ‘getting the world right’) and too empiricist (because animals only revise their priors when strongly motivated to do so; otherwise the prediction errors are usually ignored). I think of PP as being too Kantian precisely by being too empiricist and too rationalist. What we need is a version of PP that incorporates the best of the pragmatist criticisms of Kant. Dan Williams has been working on that, and there are a few others.
Thanks KN. The Barrett one is the book on the Klein podcast; she’s explicit about her constructionism.
Do you have any papers that details your concerns with those Kantian aspects of PP? (ETA) I’ve glanced through Zahavi’s “Brain. Mind, World”; is that the type of concern you are referring to?
I agree there are many useful ideas in less radical versions of enactivism.; anything consistent with an active scientific research program would be of interest to me and so robotics work definitely qualities.
I see Williams has summarized his ideas in a series of blog posts so I will look through them. I did enjoy his representation wars paper.
OK, I’ll stop now, except to say I don’t think introspection is a reliable way to understand how our perceptual systems work.
Here is case anyone is interested (not you, Neil!)
https://danwilliamsphilosophy.com/category/thesis-summaries/
I agree. I’m not sure why you brought that up. I have not been using introspection. I have been analyzing, as best I can, the problems that face a biological organism. And then I have been examining what the organism has available to solve those problems.
Much of our disagreement is about the nature of those problems.
I was alluding to this from your post:
“Suppose, however, I set the clock to be one hour fast. And then I read the clock. In that case, I’m not predicting anything. I’m just reading off a time.”
I’ll be interested in how you describe an organism’s problems.
I’m not sure why you think that has to do with introspection. I was just using a made-up example to illustrate why I’m skeptical predictive processing.
An organism’s primary problem is to survive (or, better, to thrive). In order to do better than random, it needs information about the environment. In order to be able to tell whether what it is doing is working, it needs information about self. So that perception and proprioception (which I take to be a kind of perception) are primary problems.
We can perhaps take it that proprioception is, to some extent, innate. But it seems unlikely that perception could be innate. At the very least we should expect it to need some tuning to get it to work well.
Getting perception to work has been a major concern.
You could not possibly do any PP nor use Bayesian methods until you have perception. Somehow, almost everybody wants to take perception for granted. But that way evades the real problems.
PP IS a model of action/perception; it is not something that happens afterwards. Prediction is a neural process, not a personal level inference.
I’ll move to your thread for further comments.
As I already said, you could not possibly do PP before there is perception. So I see you as confirming that PP is completely absurd.
There seems to be a massive failure to understand the problems that perception must solve.
I know that, according to many AI people, when a photon hits a retinal cell, that’s assumed to be data. But it isn’t, and it couldn’t be.
As best I can tell, that view of perception amounts to:
(1) Assume perception.
(2) Using that assumed perception, determine which way the eyes are pointing.
(3) Get data from stimulated retinal cells.
(4) Subtract the direction information from step (2). You now have data about the world. Construct perception with that data.
It is totally circular. It could not possibly work.
Add a gyroscope, as an alternative way of getting that directional information (step 2), and you might have something workable. But there is no gyroscope in our heads. That approach cannot get started.
Podcast where Keith Frankish explains illusionism and consciousness and also many philosophical ideas in the debate. Transcript available:
http://rationallyspeakingpodcast.org/show/rs-242-keith-frankish-on-why-consciousness-is-an-illusion.html
Including, I suspect, a second order failure on where the communication failure originates.
For Neil
Measurements Are Often Full of Lies—and That’s OK
Nothing weighty.
What makes something a lie is intention to deceive.
Colloquialisms are imprecise language, but are not lies unless they are used with the intention to deceive.
Science seems to experience some conflict over the need to communicate, and the need to be precise in language. I’m not sure there is a perfect solution.
But we have the phrase grammar nazi to denote someone who values correctness over communication. I prefer not to worry about metaphors, unless they are used to support an invalid argument.
petrushka,
Except to the materialist, even intention is a metaphor. Intention is just what you think happens, it is not really real. How can physics have intention? It just is.
Why Trust Science?
Book review by Phillip Kitcher. His answer to the books’ titular question:
“In the end, then, we should trust science when it is pursued as a collective enterprise, subject to standards recognized by the practitioners, and when the standards are derived from reliable results. Properly conducted research conscientiously uses techniques of observation and experimentation that have generated recognizably stable successes, and analyzes the results using methods that have been shown to work. Since the seventeenth century, to different extents in different fields, domains of research have acquired a rich corpus of such methods and techniques. That corpus is transmitted to young investigators in their training. It guides their subsequent research, and it supplies the standards against which their activities should be measured. As they pursue their particular projects, their mentors, colleagues, and rivals hold them to those standards”
Not really relevant to anything.
Historically, balances existed long before anyone cared to distinguish between mass and weight.
Ask not what you are measuring. Ask whether the measuring is useful.
Or:
Measure first; work out afterwards what it is that you are measuring.
For perspective, cosmologists have been measure dark matter for some time, but they are still uncertain what it is that they are measuring.
Well, measurement was involved, so your name came to mind.
BruceS,
I should have clarified the point.
We don’t start out knowing what the world is like, and then trying to measure it. (Well, sometimes we do, but that isn’t a good starting point).
The idea is that we start out finding ways to measure or categorize, and test those for usefulness. And only then can we begin to work out something about the world. We have to construct a world based on our measuring. That’s the way of finding out about the world.
On that actual referenced article, my inclination would be to say that a balance measure mass, not weight. That’s because weight is the force of gravity acting on a mass. But a balance compares this for two different masses, and the gravity part cancels out of the equation. It won’t work without gravity, but the strength of the gravitational field doesn’t affect the result.
Thanks Neil. To be honest, I was not expecting a serious reply to the linked post; hence that “not weighty” attempt at a pun, which obviously did not measure up.
In case anybody’s interested, I finally got a publisher for my book on democratic theory. It’s Lexington Books, the academic imprint of Rowman and Littlefield. They’re mainly interested in selling to research libraries, so their books are very expensive. But, as we, know, there are often “other ways” to obtain books these days. Also, I plan to give a bunch away.
Dunno exactly when it will come out, but I’m hoping it will be before the U.S. election.
walto,
Excellent! Do you focus on the United States or is it more general?
I think the principles are general, but when I give particulars I focus on the U.S. because I know more about it.
walto,
Ok. I ask because there’s an election campaign in the UK happening at the moment that affects my future in the European Union.
Congrats, walto!
BTW, which commenter here do you think would be your SoulMate(tm)?
Congratulations, Walt. Will there be an electronic version; this is helpful for “extended” distribution (nudge-nudge, wink-wink)?
I hope you make it big time on the podcast circuit.
Be advised: Corbyn, not Johnson, is Britain’s Trump.
https://www.theatlantic.com/international/archive/2019/11/jeremy-corbyn-like-donald-trump-not-boris-johnson/601957/
Atlantic has paywall which limits free articles.
I think I might be polygamous.
Thanks, Bruce! (I’m afraid I might be a deer in the headlights, though.)
It will come out in an (also expensive) e-book simultaneously, with a paperback to follow if it sells.
BruceS,
Not a convincing case, I have to say. The ‘anti-establishment’ angle would have any revolutionary lumped in with Trump, which would be rather absurd. If I’m looking for a British right-wing narcissist whose appeal relies mainly on being a bit of an oddball, I know where I’d look.
So happy for you, walto. You’re on a roll! 🍻
Thanks, Dazz!
Tim Maudlin on why physics needs quantum theory in addition to QM formalism.
Interesting asides on the role of the manifest image, whether relativity implies time is an illusion, Bohr’s obscurity, the purpose of Schrodinger’s cat thought experiment, the shut-up-and-calculate school, and political/institutional power in physics.
https://iai.tv/video/tim-maudlin-in-depth-interview-quantum-theory