The following is a rough transcript which has not been revised by The Jim Rutt Show or Alicia Juarrero. Please check with us before using any quotations from this transcript. Thank you.
Jim: Today’s guest is Alicia Juarrero. She’s a professor of philosophy emerita at Prince George’s Community College in Maryland. And this is kind of old home week. I actually grew up in PG County, as we used to call it in the day. They tell me you’re not supposed to say PG anymore.
Alicia: I think not, yeah.
Jim: But that’s what we always called it.
Alicia: We all do. We still do, yeah.
Jim: But I lived there from the time I was two till the time I was 22. And my wife is from PG also. And my brother is a distinguished alum of Prince George’s County Community College. And probably half of my friends went there. So, it’s great to connect with Alicia. She is a complexity person. She is the author of Dynamics in Action. I have not read that book, but I ordered it and look forward to reading. It looks actually very interesting. And she’s co-editor of Reframing Complexity, Perspectives from North and South, and Emergence, Self-Organization and Complexity, Precursors and Prototypes. And she’s written lots of publications and referee philosophy journals.
She’s also written a couple of things that I read while I was doing my research for this episode that I think the audience might find interesting. One’s called Downward Causation, Polanyi and Prigogine, and another one called, Western Science and Philosophy, Can’t Deal With The Relations Between Parts and Wholes. They’re pretty serious, but they’re not quite scholarly papers. And so, I thought our audience could deal with them. And as always, links to those papers and the books will be on our website at jimruttshow.com. So, welcome Alicia.
Alicia: Thank you for having me. It’s pleasure to meet you.
Jim: It’s good. I think very good conversation. I really enjoyed reading the book. It’s quite short, 235 pages, but it is chock-full of stuff. There’s lots of ideas in this book, but we probably aren’t going to get to them all. My topic list is… I try to keep it to seven pages, but it looks like I got more 11. So, we’ll see how far we can get into it. But I also like your writing style’s very clear.
Alicia: Oh no, it is not. That is not [inaudible 00:02:06]. You must have not read much of it, because it is horrendously [inaudible 00:02:09].
Jim: No. For a philosophy book, it’s damn readable.
Alicia: No, it’s awful. I don’t know. [inaudible 00:02:15].
Jim: Well, no, I would disagree. I would disagree. But today, we’re going to talk about her newest book just published called, Context Changes Everything, How Constraints Create Coherence. The book was just published a couple of weeks ago, right?
Alicia: Couple of weeks ago, yes.
Jim: We’ll have a link to the book on the episode page. So, let’s actually start with something we talk about a fair bit. A number of our guests, for some reason, have found this framing useful. And that is Aristotle’s four causes. This is a theme that runs throughout the book. Why don’t you remind the audience what the four causes are and what they are?
Alicia: Well, I think Aristotle probably got it from the potter’s wheel. He started thinking about causes and effects using the example of the potter’s wheel. Four causes are material cause, the clay, the stuff from which the pot will be made. Then, final cause, or purpose, or teleology, which is the goal to which the thing that you’re making will be put, which is pouring water. So, the final cause of the picture would be water. The formal cause, it’s not quite shape. It’s what makes a picture a picture. So, it is the essence, the basic fundamental identity of the thing. And then, finally, efficient cause, which is the actual force exerted by the hands on the clay to turn it into the picture. So, it’s energetic. Efficient causes energetic exchange.
Jim: One of the points you make is that, prior to what, maybe the late 16th century, people tended to consider all those causes when they were thinking about nature and the pre-scientific era.
Alicia: Right, yeah.
Jim: But one of the moves that probably came off accidentally, more or less, from the invention of modern science was a very heavy focus on the efficient cause.
Alicia: Correct. In a sense, material cause, people figured, well, science will take care of that. And formal cause and final cause, were discarded. They went out with a bustle. “This is something we don’t have to worry about anymore. So, we can explain everything in terms of forceful causes.” It seems to me that when you’re talking about complex dynamical systems, somehow, final cause and formal cause gets snuck in indirectly, not the way Aristotle… Aristotle thought people were born with a… or organisms or all natural phenomena had an inherent internal entelechy, like an acorn has the form of an oak built into it and all it has to do is unroll and unfold into the final form. But I think what we have now with complex dynamical systems is that interactions with the environment, and in my view constraints, are the contemporary version of formal and final cause.
Jim: Yeah. And the overfocus on efficient cause matter and motion bumping into each other, I often refer to that as naive Newtonianism. And most nerdy smart kids go through that period, right? And they make the error of thinking the famous Laplacian error.
Alicia: Correct, absolutely.
Jim: Where he says, “Give me the position and velocity of everything in the universe, and I can predict the future in the past with total precision.”
Alicia: Correct.
Jim: But fortunately, once you get exposed to ideas of complexity, you realize-
Alicia: Absolutely.
Jim: … it’s completely crazy. And that has actually caused for those people who’ve gotten the complexity lens to realize there’s much more to our universe than naive Newtonism.
Alicia: But it’s amazing how it’s persisted. And you can’t blame people. It seems to do a hell of a job predicting eclipses. But even Newton knew the three-body problem would mess things up, right? But that was a warning Newton gave us that went heated for many centuries. My first paper was about content Prigogine and exactly about that subject matter.
Jim: Yeah. And then, as you talk about in your book, one of the current manifestations of it, though it depends where you are in the sciences. Now, I’ve had the good fortune to be around complexity people for the last 20 years.
Alicia: Me, too.
Jim: And so, there’s much, much less of that there. But I suppose out in the wilds of solid state physics and places like that, there’s still a lot about what you call nothing-but-ism, right? Explain what nothing-but-ism is.
Alicia: Well, the idea that the whole is nothing but the sum of its parts. And therefore, anything that appears to be an emergent property is really an epiphenomenon. It is froth that’s thrown up, but it really has no causal power. Well, of course not. Because if causal power is only thought of as efficient causality, then clearly, the synchronization of the photon streams in a laser beam don’t align their component laser beams as another efficient cause. So, that’s the problem.
Jim: Yeah. It leads very quickly to absurd conclusions, which makes me wonder, why did it hang in there so long?
Alicia: Well, and you know where it hung in? I came at this because I wrote a dissertation on the difference between explaining and justifying behavior. Well, justification has to do with moral reasons and so on. But explaining behavior, the first sentence of that first book that you quoted is, what is the difference between a wink and a blink? Presumably, it’s the cause, right? Meaning that an intention causes a wink, but my throwing some sand in your face would cause you to blink. But then the next question is, what the hell’s an intention and how can an intention cause be action in an efficient cause way, correct?
Jim: Yep.
Alicia: And my answer was obviously, well, there can’t be just one neuron pushing another neuron pushing. Forget it, that doesn’t work that way. And at the time, I was living in Berkeley and I was hanging around people who were into network theory and systems analysis and so on. I’m going, “You know what? There’s got to be a network property.” We’re talking the ’70s, Jim. There’s got to be a network property that somehow produces emergent properties, which in turn can loop back down and cause the neurons that are responsible for motor control to move the arm in a way that it satisfies the infection that I started out with. All right, so how is this going to work?
You were talking about PG Community College. The nice thing about PG Community College is nobody gives a damn if you publish anything or not. You’re going to be judged by how well you teach. So, nobody cares about whether you publish. On the other hand, nobody’s forcing you to write papers on the fifth decimal point of the existing theory. So, I could play around. And in this area in DC, I could hang around NIH and start listening to people come talk. And so, you start understanding how patterns in neural systems work and so on. And so, it’s got to be something. It’s got to something like that.
So, that’s when I decided, all right, you know what? I’m going to give the term cause to the Newtonians. I’m not going to fight that battle. Because it may be forever to fight that battle. So, instead, what I want to do is look at the notion of constraint, because as soon as the hard scientists get into trouble with that Newtonian silly understanding of efficient causes you mentioned, they retreat and hide behind the notion of constraint. And I thought, you know what? That’s going to work. So, that’s how my trajectory towards reconceptualizing causality, and especially formal and final cause in terms of constraints, developed.
Jim: It’s interesting. And actually, it worked well with the little analogy I use when I’m trying to explain complexity to just random people at a party or something. I will say, “We can think of reductionism, classic science, as the study of the dancer, while complexity is the study of the dance,” right?
Alicia: Dance. Yeah.
Jim: And a dance is not random motion. It has constraints, right? If it’s going to be…
Alicia: Correct.
Jim: If it’s going to be a jitterbug, it has one form of constraints. If it’s going to be a waltz, it adds another. And so, when I read that and saw the movie you were making, I said, “Works perfectly with my good old…”
Alicia: Absolutely.
Jim: “…my good old analogy.” And…
Alicia: Absolutely.
Jim: And it held together quite well. So, now let’s move on to your next topic, which is a term that most non-philosophers will have never heard of. I’ve heard of it a few times, but it’s not a term we use very often, Santa Fe Institute. Mereology. Is that how you pronounce that?
Alicia: Mereology. That’s, again, the whole-parts, parts-whole. Let’s use an example from Brian Arthur’s Noah that I love, that’s a Santa Fe book on complexity of economics, economics and complexity. What an economy is, are a bunch of individual elements that interact in a constrained way to result in an emergent phenomenon that has certain properties that the components don’t have. An economy has certain characteristics that the individual trader and seller don’t.
But once that whole W-H-O-L-E, which is a coherent whole, and I really want to emphasize the fact that a coherent whole is different from a massed clump of stuff because it’s organized and what organizes it are the constraints. But once it’s organized, then all of a sudden, the components also acquire different properties because suddenly, they are now traders and regulators and so on and so forth. Correct?
So, there you have the whole part, the parts of the components, correct, and the constrained interactions among the components. The whole is what I am calling in this new book a constraint regime. Because one of the problems we’ve also had, I don’t know if it’s Newtonianism or what, but we tend to reify things, no pun intended. We tend to reify things.
But an economy is nothing other than all these constraints all held together by an overarching set of constraint regimes once the constraints close into a coherent whole, but they can then loop back down and affect their components and they acquire… now they have a role. In a society, once a society is a socially organized structure, people can be citizens. They can be senators, they can be teachers. Those roles don’t exist except within an organized society.
So, then the next question is, well, on a more general level, what is the relationship between the parts and the wholes and the wholes and the parts? If you really buy the whole Newtonian idea, ain’t no difference between the parts and the holes. The holes are nothing but the sum of the parts. But that means you cannot explain how it is that my behavior is constrained top-down by my living in a particular culture, the fact that I was born and raised in Cuba. All of these affect my behavior, but they don’t do that as an efficient cause.
A culture, an economic system doesn’t cause my behavior to differ in any kind of efficient causeway. But since the Newtonian revolution, pretty much mereology got thrown out of the picture. So, that messes up philosophy of mind because you cannot explain mental events, if you think of mental events as emergent properties, because they should be reducible to a bunch of neurons pushing each other around. But where does the emergent property go when you have that?
Jim: Yep. I’ll tell you a little example again, my little homey examples that I love to use on this is people who doubt, we’ll get this later, the idea of top-down causality. I say, let’s imagine me dead and run through a blender and poured into a bathtub. What’s the chances that those chemicals are going to hop out of the bathtub, walk down the hill, and go to the ice cream store?
Alicia: That’s correct.
Jim: Essentially zero, right?
Alicia: Yeah.
Jim: And on the other hand, if it’s me and I’m operating up in cognitive space and it’s after dinner and we just had a nice dinner and I feel like a walk with my wife to go down and get something nice, then we may decide to haul all these atoms and molecules and they have nothing to say about it because this top-down idea, well, let’s go get some ice cream, causes the… and by the way, Newtonian physics is never… or let’s say physics is never violated.
The atoms are dragged along, and they are dragged along through efficient causes. They’re all stuck together and bound by various forces. But the decision to go get the ice cream came at a higher level in the stack. And again, some of the attempts to deny that that happens is just bizarre.
Alicia: Well, it is to me and you, but it isn’t to an awful lot of people. And even folks, and I hope there hasn’t been a change in the last couple of years, but even somebody like Brian Laughlin, who recognizes the reality of emergent properties, nonetheless gets very weary about ascribing causal properties, causal powers, let’s use that, causal powers to those emergent properties. In a sense, the notion of supervenience is continuous.
So, people like Carl Gillett, people like Robert Bishop, there are very few of us who are willing to go the next step and say, “Not only are emergent properties real, they have causal powers with respect to their own components.” And that’s what homeostasis is, for God’s sake. Homeostasis readjusts the metabolism, the neural system, and so on in order to maintain the integrity of the whole.
But the idea that nonetheless, that should be explicable in a reductionist fashion, that the power aspect of it is still not quite… the physicists haven’t bought it. The philosophers… I hate to say this, I never read philosophy anymore. I grew up, I was trained as an analytic philosophy in the United States. All they’re worried about is the meaning of words. There are very few people.
And you can tell why, how this is when you have people like Chalmers and Christof Koch who Chalmers has thrown in the towel and he’s really saying, “Oh, the only way you’re going to have mental properties is if you build them in from the get-go from the bot. So, therefore electrons have mental quality.” Oh, come on. Rocks? Give me a break.
Jim: And we had Christof cook on the show some time back, and he actually is a pan-psychic as you…
Alicia: They all are. But you understand why because if all of reality comes from innate, internal fundamental properties that only interact with efficient causes, there’s no way you can get a coherent whole, and that’s the keyword, coherently organized, from which emergent causal powers come.
Jim: And it is interesting that they… another extremely bright guy named Ben Goertzel, been on my show many-
Alicia: Oh, yeah. I know Ben.
Jim: … many times, good friend of mine, I think he’s a pan-psychic for exactly that reason because otherwise, he’s wedded to his model of the universe. And where does consciousness come from if it doesn’t-
Alicia: Correct.
Jim: … if it isn’t innate? While a complexitarian would say, “Well, it’s obviously emerged just another level on the stack, right? And that’s not that hard.”
Alicia: Correct. But it’s this notion that by nature, “Oh, it’s in my nature,” is somehow given as an essence in the fundamental tiny, teeny bits that make up the rest of it. That’s still much more widespread than one would think.
Jim: We’ll get to supervenience later, and I think we have a somewhat different point of view, but not entirely different. We’ll get to that later. But I think we’re on the same page that hey, this need to smuggle in something like consciousness or cognition as a fundamental property of matter seems to be a big overreach. Do we try to smuggle in the fundamental nature of digestion? I don’t think so. But we get overly confused when it comes to cognitive processes.
Alicia: But digestion’s a good example, and that’s the one John Searle has always used.
Jim: That’s why I always used…
Alicia: Yeah, exactly. And I used to fight him. We were on the NEH board together. But the thing about digestion is digestion really doesn’t do mu… It is an effect. Digestion is an effect of all these other processes. The reason I like homeostasis is because the metastability of homeostasis is causally effective. If I have the fudge brownie, it’s going to switch all my glucose and everything else around to keep the integrity of the whole.
So, it has caused what I would want to call causal powers, but I’ve given up that term because people will say, “Oh, but it is an efficient causation.” No, that’s right. It isn’t. But it is C.S. Lewis’s notion of cause meaning without which not. If that weren’t the case, it wouldn’t be. So, that’s the causal notion I would espouse.
Jim: Okay, let’s now make the next step, which is why don’t we start to define what you mean by constraints, and lay out a taxonomy of different kinds of constraints?
Alicia: Okay. I…
Jim: This is where the book really started to get into new territory for me. I found it very interesting.
Alicia: And that’s what the new one is all about.
Jim: Indeed.
Alicia: That’s what the new one is all about. I think, in the first book, I made a distinction, barring from Lila Gatlin’s book called Information in the Living System which was from 1960, something or other, a distinction between, I used the term, at the time, context-free and context-sensitive or context-dependent. I got a lot of grief from people saying, “Ain’t nothing that’s context independent.” All right, so, in this one, I call it context… I mean, nothing’s context-free.
So, in this book, I call it context-independent and context-dependent. Context-independent, according to Lila Gatlin, are conditions that take a system far from equiprobability. So, whatever takes conditions away from white noise, from random noise. So, if you institute a gradient, you have instituted a context-independent constraint.
If you institute polarity or charge, we’re talking the early universe here probably, those were probably the earliest constraints, and I call them context-independent because in a sense, what they do is they set the context. In a sense, they set the boundaries of possibility space and inhomogeneities within that possibility space. So, it’s no longer white noise. Lila Gatlin calls context-dependent constraints those that take a system away from independence.
So, one of the things that I find interesting also is that we might poo-poo Newtonianism now, but when you look at the second law of thermodynamics, and I’m petrified because I talk about the thickened law of thermodynamics in this new book, but according to the Boltzmann and everybody’s interpretation of the second law of thermodynamics, the events in the particles are independent of one another. And I think you never going to get complexity if you have independent particles.
That’s the beauty of Stuart Kauffman’s button example, button mesh example. You tie one button to another thread, and all of a sudden, the thing turns into a mesh. You have a face transition, and you have a mesh. So, what would I consider context-dependent constraints that take a system far from independence, I would consider catalysts context-dependent constraints. I would consider feedback loops context-dependent constraints.
And I think, since that first book was published 20-something years ago, what’s nice to me is the burgeoning of epigenetics nowadays. Because if epigenetics isn’t an example of context-dependent constraints with a vengeance, I don’t know what is. So, once you have all these context-dependent constraints acting inside a context-independent possibility space, set possibility space, then I think the possibility of complexity appears.
You don’t have a possibility of complexity just with emergent, just with efficient causes, but these forms of constraints do. I also make a distinction between temporal and spatial constraints. I’ll give two examples from playground devices. A playground swing, the child learns very quickly that when they kick is as important as how strong they kick, and the timing of the kick does not impart more energy to the kick.
So, that’s an example of where Newton alone, efficient causes alone won’t work. Does that make sense? You have to, otherwise, it won’t kick. A seesaw is another example of a spatial context-dependent constraint because depending on the length of that plank on the top and the plinth on the bottom, the base on the bottom, then where the child sits will be determined by those context-dependent constraints in order for them to be able to teeter-totter.
Another example that I like a lot, which I know you know Dave Snowden and he likes to use it a lot, I think I used it first but that’s okay, is the roundabout, the traffic circles. The architecture of the roundabout is a context-dependent constraint that affects the behavior of pedestrians and drivers in a system. So, that’s another example of a context. I think sequencing is a beautiful example of temporal constraints.
A has to be done before B, which has to be done before C, which has to be done before E, and it’s done in a different order. The order make a huge difference in the outcome. So, all of these I consider context-sensitive constraints. I lump a lot of those into enabling constraints because I think of enabling constraints as, particularly those context-dependent constraints, that together, achieve closure such that this coherent all emerges.
Jim: Interesting. One of the things that was interesting about your use of the word constraint is you use it very broadly.
Alicia: Yes.
Jim: For instance, one of my mentors in the complexity space was Harold Morowitz.
Alicia: Oh, sure.
Jim: And…
Alicia: He was here nearby at George Mason for a long time.
Jim: Yeah. I used to live out in Loudoun County, and I come in-
Alicia: Oh, yeah, sure. You come back in. Yeah, sure.
Jim: … and see him quite once a week.
Alicia: Yeah, exactly.
Jim: And he has this idea, the 27 emergencies or whatever it is, and that each one is defined by a set of pruning rules which tilt things one way or the other, and that those changes are fundamental inflection points in the evolution of the universe. And some of them probably are contingent and some of them aren’t. And the ones that aren’t, maybe correspond to your non-contextual constraints, for instance. One of his steps is defined by the poly-exclusion principle, for instance, right?
Alicia: Yeah, I mentioned that in the second book. I think notions of symmetry and conservation… It’s funny. Physicists all use the word causality and so on and use it half of the time, they’re talking about Newtonian cause, then when they figure it can’t fit, then all of a sudden, they move to principles to that sort of thing. And I absolutely agree. I think rules, regulations are examples of constraints. They set the possibility space and they determine what is more likely within it than otherwise.
Jim: And I did like the point, and this is Harold makes the same point, that especially higher up in the stack, we’re talking more probabilities than we are blacks and whites, right?
Alicia: Absolutely. And perhaps what I speculate on in this new book is that whenever you have the emergence of a coherent dynamic, you have a phase transition to a continuous function. If you have that, if I’m right on that, then it seems to me that top-down control is analog. It’s a change in the setting of the system to an analog notion.
So, you’re going from your old-fashioned toggle switch to a dimmer switch. And that’s why homeostasis can keep the timeliness because that’s very important for homeostasis in most ecosystems, and it also can keep the sensitivity to local conditions. I think only analog can do that. The Dysons were poo-pooed a lot, but there’s something that tells me that their emphasis on analog control somehow might be onto something.
Jim: And to make it clear for the audience, talk about Freeman Dyson. And it…
Alicia: Freeman and George, both deceased now. Yeah, George died.
Jim: Did George die really?
Alicia: Yeah, recently, about the last six months or a year.
Jim: He used to come by the Santa Fe Institute at the time. Freeman, I don’t believe ever did, at least I never met him there. But very interesting family. I also knew…
Alicia: Very interesting family.
Jim: I also knew Esther Dyson, who is-
Alicia: Oh, that’s right.
Jim: … George’s brother-
Alicia: Of course.
Jim: … who was a very extremely influential thinker in the early days of the computer industry and then later has become a brilliant venture capitalist, particularly in the Eastern Europe. That was her area.
Alicia: Oh, that I didn’t know.
Jim: A quite amazing woman, really quite.
Alicia: Yeah, she really was.
Jim: Really quite. Okay, I have it way later in my topic list, but let’s talk about it now, which is analog versus digital. I actually have a fair bit of background on this. Two of my companies back in my business career, they weren’t mine, I was a chairman of one, I was a investor and director or another, were both involved in software for designing computer chips and in particular, for designing analog computer chips. And so, I learned quite a bit about analog versus digital. And for instance, I’d learned, I probably should have known it, that digital actually is analog below some level, right?
Alicia: Yeah. It…
Jim: And then, there’s a whole series of clever things they do to go from analog to digital. But then the other point, which you allude to, is that on comparable computational tasks, analog’s literally six orders of magnitude more efficient in terms of-
Alicia: That’s it.
Jim: … the power limit.
Alicia: And that’s got to be the reason why the brain doesn’t overload, correct? That has to be the reason why the brain… So, the brain is a good example of that. Why the mind, or when you do the face transition to a mental event, what you’re doing is you’re transitioning from neuronal electrical exchanges to control on the basis of some kind of typology, a facial recogni… So, now you’re talking on the basis of how close is that pattern to this exemplary face.
Jim: Though, again, to keep in mind, the brain is a good example because the brain is both digital and analog, right?
Alicia: Exactly. And it…
Jim: Neurons…
Alicia: … switches… Correct. And it switches back and forth, correct?
Jim: Yeah. And so, the domain of ideas or concepts-
Alicia: Exactly.
Jim: … or objects-
Alicia: Exactly.
Jim: … the very important object ontology that at least mammals and above develop, probably birds too, is continuously variable and it’s not perfect.
Alicia: Exactly.
Jim: It’s classic analog classification.
Alicia: It’s classic, yeah.
Jim: And yet, it’s actually implemented on digital circuitry-
Alicia: Correct.
Jim: … which is interesting.
Alicia: Correct. So, it’s carried out digitally, but the command top-down is analog. That makes perfect sense to me.
Jim: Though I would also just add that the continuously variable, while true of analog versus digital, is maybe less important than people think because…
Alicia: That’s true. Absolutely. Exactly. George Dyson used to use as an example, even though the internet and all the computers that we use now are digital, when you look at social media and that stuff, it’s what’s connected to what. So, we’re back to the dance, right?
Jim: Yeah. And it’s the people who are analog, right, and it’s the dance of the people.
Alicia: Right. And the people, in a sense, are the condensated nodes of the intersections of all these constraints, right?
Jim: Yeah.
Alicia: So, that’s what I argue in this new book, that everybody’s all bent out of shape about identity because they all think of identity as something internal, essential. Same thing as Koch and Chalmers we’re concerned in. But we need to transition to identity as a set of interdependent constraints. That’s what makes me, me. I am a set of interdependent constraints.
Jim: And high dimensional.
Alicia: That is [inaudible 00:34:33].
Jim: But it also…
Alicia: I am very high dimensional.
Jim: That’s the thing that’s so annoying about this stupid identity politics of both the left and the right is both sides want to condense down to just one or two dimensions and there’s hundreds of dimensions, right?
Alicia: Correct.
Jim: I’m a cat. Someone’s a dog fancier who loves Mountain-laurel, but not Rhododendrons.
Alicia: Correct.
Jim: There’s so many dimensions. And to say that these two are the ultimate ones, just dumb.
Alicia: You know what did it for me that I finally decided, “All right, I can finish that first book,” was work by Geoffrey Hinton, who’s now a big shot, Plaut and Shallice, and they were doing early work in artificial neural networks that read words. It was one of the early text-reading networks. And they were working on simulating… I had never heard it at the time. Have you ever heard the difference between surface and deep dyslexia?
Jim: No. That was very interesting.
Alicia: I thought that that’s what gave a… That all…
Jim: It was in your book. You gave a great example in your book.
Alicia: Oh, God. And I use it in a sense. No, well, it was their example.
Jim: Okay.
Alicia: It was their example. Do you want me to repeat it here?
Jim: Sure, yeah.
Alicia: Okay.
Jim: Why don’t you tell the story? It’s very interesting.
Alicia: I’ll tell the story. Surface dyslexia is what most dyslexic human beings have. They transpose letters. So, they read cut for cat or they read tag for gat. They just transpose letters. Or four gets read as a seven, that kind of thing. The point being there’s no semantics involved. It’s purely the appearance of the input. Apparently, deep dyslexia is very different.
And I’ll give you the example first. What Hinton, Plaut, and Shallice, and the reason I plunked them all together is because Hinton and Plaut would write, and Hinton and Shallice, Plaut and Shallice, though there’s zillions of papers that came out about 25 years ago about this, and they trained these early reading neural networks. And when they trained them with feedback groups, so they were recurrent networks, and they lesioned the networks above the feedback loops, the network would be shown cat and it would say tac.
So, it would produce surface dyslexia kind of errors. But then if they lesioned the artificial neural network below the feedback loops, they might show the network, again, these are silico networks, they’re not organic, they would show them B-A-N-D, band, and the thing output would say it had read orchestra. Or they would show it B-E-D and the thing would say it had read cut.
And so, somebody, not me, asked one of them, Hinton, Plaut, or Shallice, “How do you explain this?” And their answer was, “The only way I can explain this is to postulate that the system, because of the middle layers, that we don’t quite know what’s going on.” And that has got to be what’s going on with ChatGPT by the way. The middle layer has created semantic attractors. Period. Case closed. And the output is issuing from the semantic attractor.
Jim: That makes perfect sense.
Alicia: That makes perfect sense. But there’s your emergent property being causally powerful.
Jim: Interesting. And by the way, for all my computer people who listen and there’s a lot, the G.E. Hinton she’s talking about is the Geoffrey Hinton.
Alicia: It’s Geoffrey Hinton. I shouldn’t say this on the air, but I will. I said to John Searle at one of these NEH meetings, “John, I think Hinton’s work is so very good.” “I don’t understand why everybody’s so interested in Hinton.” I said, “He’s very good. Trust me, he’s very good.” It is the Geoffrey Hinton, right?
Jim: Yeah. He’s the guy that did the breakthroughs that drove-
Alicia: Absolutely.
Jim: … all the stuff that we’re doing today and-
Alicia: Absolutely.
Jim: … keep doing.
Alicia: Nobody was paying attention to these reading networks a long time ago, or at least not as many people should have been.
Jim: I was actually doing neural nets back in 2001, 2002. In fact, that’s how I got invited out to the Santa Fe Institute. It was my work on evolutionary neural nets.
Alicia: Well, my book was published by 1999, and this has been published before, so I think that stuff was published ’95, ’98, thereabouts.
Jim: Yep. It was. Probably thereabout.
Alicia: It did it for me. I thought, “Now I can write this book because I have some kind of evidence that the semantic attractor in the brain, in a culture, whatever you want to call it, has causal effects.”
Jim: Now, when I was reading the book, I write a lot of notes. I always do. What I wanted to ask you about, how does it fit into your concept of constraints? It certainly is one of Harold’s pruning rules, and that’s the idea of the species competitive exclusion principle. Which actually, my core field is evolutionary computing, and so I understand a lot of how speciation works from a mathematical perspective.
Alicia: [inaudible 00:39:55].
Jim: And it’s more or less a thing that is just true. If a competitive dynamic has these attributes-
Alicia: Right.
Jim: … and a fitness landscape has a certain shape, there will be a competitive exclusion principle around species. Does that fit into your idea of constraint?
Alicia: Well, if what I speculate about there existing a constraint regime, correct? Then that is the constraint regime for that possibility space, right?
Jim: Yeah. It basically says if anything is competitive, if you get too far away from the center of what’s sufficient in that part of the fitness landscape, inevitably, you’ll be less… not inevitably, most of the time, you will be less fit than the ones right near the peak and therefore, your numbers will go down. So, it’s very hard to move away from the peak of the species definition-
Alicia: So…
Jim: … the phenotypical collection. Of course, as you point out, they’re not all the same. They’re an ensemble, but they, nonetheless, cluster around a species type.
Alicia: Correct. I don’t know if this is an attempt to answer your question or attempt to evade it. I’m not sure. Tim Allen, T.H. Allen, and Starr, who have that book, and they’ve got two editions of it. I like the first one, the first edition better, called Hierarchy Theory. And they use, as an example, prairie grassland ecosystems in the Midwest where apparently, the prairie grassland competes against flowering plants. It also competes, obviously, against horses that eat the grasses, correct?
Apparently, if you look at that whole ecosystem, and you look at it from the point of view of the grass, competition with the flowers is a lot harder than competition with the horses. Apparently, flowers will really do a number. Flowering plants, I don’t know, dicotyledons or whatever the hell those things are called. All right, so, over history, over a period of time, the prairie grasses have sent out… and of course, it’s all selection.
I’m not, at all, disputing, obviously, Darwinian. They have sent out meristems. Meristems are these chutes right below the surface, but that stick out enough that the horses can eat them. And so, in a sense, Jim, the grass invites the horse to come in to feed and incorporates it into what is now an enlarged ecosystem. So, instead of having two competing species, what you having is the enlargement of the niche or the constraint regime. And Tim Allen says it’s like a shanghai what was a predator, to incorporate it into an enlarged constraint regime now the horse is part of the grassland’s ecosystem. And that’s how they keep the flowers at bay.
I think that way of looking at it as a dynamical, a mutual adjustment system is an awful lot better than two species competing. Yeah, two species are competing. And that I get from the fact that I didn’t realize until fairly recently that the understanding of the term fit during Darwin’s era meant more like you go to a tailor to get a fitting for a new suit of clothes. So, when you think of fit in that sense, it’s what is it? It’s a mutual adjustment, correct? Which means that when I talk about constraint, the form of bottom-up and top-down relationships has to be one of mutual constraint adjustment. That’s what it basically is. You’re adjusting all the constraints to see how you can best satisfy the overarching dynamic.
Jim: Then to your example of the grasses and the horses, etc. it’s always in a co-evolutionary context.
Alicia: Exactly.
Jim: And that’s your idea of context, right?
Alicia: It’s temporal too. You’ve got to include temporality. Absolutely.
Jim: Yep. All right, so let’s move on here. Yeah, that was good actually. In the interest of time, I’m going to skip over the COVID example and let’s talk more about time temporal constraints. And I think you did a really nice job of talking about cardinality, ordinality and indexicality.
Alicia: Indexicality.
Jim: Indexicality. I want you to distinguish those, particularly distinguish cardinality from ordinality.
Alicia: Well, cardinality is just amount, correct? More…
Jim: So, a pile of sand has a cardinality.
Alicia: Pile of sand as bigger or as smaller. Correct. But ordinality is first, second, third. I don’t see where you can get first, second, thirds much out of Newtonian mechanics. Whereas, once you have temporal constraints instituted, this has to occur and this sets the stage for then the next thing to occur.
And that sets the stage for the third thing to occur. Then you have orality, which is first, second, third orders. Indexicality, it’s like perspective or the position you are in, in a complex dynamical structure means there are certain properties that are indexical. This is to the left of this. This is to the right of this. That’s what I mean by indexicality.
It really has wreaked havoc in philosophy of mind because intentional causation, again, is eminently indexical. So, one example I use in this new book is that, that philosophers use all the time, Mary told John’s wife that he was cheating on her. But Mary doesn’t know that John’s wife’s name is Alice.
So, did Mary tell John’s wife that it was Alice or not? Do you see what I mean? It has to be interpreted, I think in terms of the emergent dynamics and emergent property.
So, it has to be treated in terms of indexicals inside and out. So, I think once you have emergent constraints in place, that’s what… How could you make a word out of the word rugged? Ruggedify.
Jim: Yeah, ruggedify. I like that. That’ll work.
Alicia: It ruggedifies the possibility space. And each one of those valleys and tractor basins or a tractor separate tricks is right. They are the ruggedness of a possibility space.
And that explains why the view from inside in a tractor looks real different from the view from the hill overlooking the next basin of a tractor, correct? And when we talk about causality, we have to take that kind of indexicality into account.
Jim: Then with respect to ordinality, many things are inherently ordered. In fact, I just published last night, a very interesting podcast, current’s number 100 with Sarah Walker and Lee Cronin on time.
Alicia: Oh, I’ve got a note to her. Sarah Imari Walker, her stuff’s really interesting.
Jim: Yeah. And on time as an object. Basically, they’re hypothesis is that evolution and other expanding complexity is essentially a series of steps that get taken. They point out that the most complex chemicals created by abiotic processes never have more than 13 or 14 steps.
Alicia: [inaudible 00:48:46].
Jim: Yeah. But biotic processes can go much higher than that. And then manmade processes can go a bit further than that. And so, I thought it was a very interesting juxtaposition with your idea of ordinality. One of the examples you gave was the social evolution of the processing of cassava. I believe it was in South America. Maybe…
Alicia: That’s from Heinrich’s book on the secrets to our success or something. Heinrich is what? Head of sociology or something at Harvard. It’s apparently something that’s poisonous, but yet nutritious if it weren’t poisonous. The indigenous community in South America has figured out a way of leaching out that poison. But the preparation for that root vegetable has to be done in a particular sequence.
Because if you don’t, you’re going to kill out the entire population, correct? Going back to Sarah Walker, somebody told me day before yesterday that apparently there’s something… I’m not on Twitter, which I probably should be.
I’m neither on Twitter nor on Facebook. Somebody told me that somebody wrote a Twitter comment saying that Sarah Walker’s driving forces are mine constraints. Which is flattering, I think for me.
Jim: Yeah, you should chat with her. She’s very good. And Lee too. The two of them were one of the better science, hard science episodes I’ve had in a while.
Alicia: He’s very good. Yeah, I like his stuff.
Jim: Interesting. Now you mentioned, I might not hit on these because they’re classic rich examples, Kaufman’s buttons, Hogan’s pendulums, and Bénard’s cells.
Alicia: Bénard’s cells.
Jim: And paint those in with your ideas around.
Alicia: Well, the Bénard cell was the source of all my interest in complexity theory. It was the early 1980s, Jim.
Jim: Could you tell us folks what it is? Not everybody knows.
Alicia: A Bénard cell is you take a pan of water. They’re called Rayleigh-Bénard cells and they were discovered at the beginning of the 20th century by Ray and Bénard. You take a pan of water, any kind of viscous fluid, and you heat it uniformly from below.
All right, you still have conduction. After a certain gradient, after a certain threshold of instability, that’s my context independent constraint, Jim. After you pass a threshold of that gradient, the system cannot handle any kind of fluctuation. And the context will amplify any minor bubble or perturbation.
And all of a sudden, you will get convections cells. Those rolling hexagonal cells made of billions of molecules of water that all align in a self-organized way. And that the cell itself constrains top-down the individual molecules of order. So, they behave as if they knew what was the one next to them was doing.
All right, so it was the 1980s and I had to go to jury duty here in Montgomery County. And so, I took a bag full of stuff that I had to read. And I was reading Kant’s Critique of Practical Reason. Kant’s Critique of Practical Reason says the problem with… And by the way, at the time the Kant wrote teleology was synonymous with self-organization.
Go figure 1804. All right, Kant said, “In order to understand this kind of phenomena.” He said, “We need an understanding of circular causality that is unknown to us.” Because remember, Kant had bought the Newtonian Kool-Aid that was all effective, efficient causality.
So, he said, “But look at how nature works. A tree produces the leaves and then is produced in turn by the leaves. So, the whole tree is produced by the component parts. And in turn, loops back down and produces the components that created in the first place. ”
And so, I’m reading this going, “All right, yeah, but how do we fit this into modern science?” And then the Prigogine and Stenger’s Order Out of Chaos had just come out of print, come in print. And he’s looking at dissipated structures, which all have that process.
Yet individually constrained interactions that after a threshold of instability, cross a phase transition and self-organized to produce a whole, which then loops back down and constrains the component parts. So, that to me was, whoa, I found a scientifically respectable way of explaining teleology and formal cause. That’s what it did for me in the ’80s.
Jim: Yeah, you were at the right place and the right time. Because Prigogine certainly…
Alicia: And again, not having to publish so that you work out…
Jim: Go where you want to go.
Alicia: Decimal point of whatever you already exist allowed me to be a dilettante and play around with ideas just because they were interesting.
Jim: And one of the things that Prigogine predicts is that especially these abiotic complex systems will actually be more efficient at burning energy than their predecessors.
Alicia: Yeah.
Jim: And Bénard cells for sure, they actually move more heat through by convection. And one of my favorites is the whirlpool in the toilet actually allows the water to go down faster, so it actually is dissipating the potential energy of the water in the tank more quickly than if it didn’t form the whirlpool. So, it’s…
Alicia: You know what I used in class when I taught courses about mines reins and machines, I actually taught a seminar in mines reins and machines in PG. But I take the standard two large gallon…
Jim: Oh yeah. Put them together and then watch the tornadoes in them.
Alicia: Exactly. And then the students went, “Ooh.”
Jim: Yeah, we made one of those for my daughter when she was like a middle school student.
Alicia: Yeah, that’s pretty cool. They can really appreciate it intuitively.
Jim: Yeah. So, the idea of Prigogine and the idea of dissipative systems, even though they have more structure and more interesting things going on, are also generally speaking, more efficient at burning energy. And so, the good old second law never actually gets violated.
Alicia: Correct.
Jim: It’s just in a different form. So, this then you’ve set me up perfectly.
Alicia: But thermal equilibrium gets retarded a bit because you have a structure that gets created in the process that persists a bit longer than the component funds, which is true. But that’s the explanation for social systems. That’s the explanation for cities. Cities are more energy efficient than they’re organized.
Jim: Per unit square foot, they burn a lot. But per person burn less. Which actually now sets up to my next topic.
Alicia: You’re not going to fool the second law.
Jim: No, exactly. That’s the one law. If anyone ever comes to you and tells you they beat the second law, tell them to go pound sand. And so, now we get to where it gets more interesting. And this of course is the secret of life is catalyst loops, auto catalytic networks, etc. This is where we go from whirlpools, which have a self-forming part, but they’re not fully closed loops.
Alicia: I really liked a book that came out about seven years ago by two… Well, a lot of them worked out of the University of the Basque Country in Spain. And the good thing about those folks is they published in English. Otherwise, we would go into black hole.
But they published in English. Unfortunately, this book is Springer. And Springer sold expensive. Nobody buys the book. But I think they’re very good. And the book is called Biological Autonomy.
And their argument is that, well, things like the Krebs cycle and so on, these are closures of process. But once you start having auto catalytic and hyper cycles [inaudible 00:58:39] and so on, what you have that… The constraint loop that closes is a loop of constraints themselves.
And so, the constraints create the constraining conditions that make them possible to begin with. And that is what enables their self-enforcing, but they’re self-perpetuating. And so, these Mateo and Moreno and Mocio argue that, that is what makes living things different from say even the BZ reaction.
Where the boundary conditions, the constraints are in a sense self-set from without. You set the conditions of the pan of water or the chemicals. But once you get a situation where the constraints themselves become self-perpetuating, then you have the possibility of reproduction of species and that sort of thing.
Jim: Now, you didn’t really hit on it as hard as I thought you might. But the perfect example of that is that the auto catalytic reactions within a cell are also responsible for building and maintaining the membrane that allows the concentration. And that’s hugely important to my mind.
Alicia: Absolutely. I think that’s what they mean. Because for membrane, we’ve always thought of it as boundary conditions. And that was my beef with Polanyi. Polanyi was ultimately religious. And so, Michael Polanyi, the philosopher at the beginning of the 20th century, I guess it was.
He believed that God sets the original boundary conditions. And then once you’ve got that, then everything else self organizes within it. But the whole point of, I think the closure of constraints is that it creates the boundary conditions within which it self-organizes as well. So, yes, you’re absolutely right.
Jim: Yeah, and I think that’s usually important. And of course, as we’ve learned, the nature of these membranes changed over time and it’s continually changing.
Alicia: Absolutely.
Jim: And the fact that they’re semipermeable with different rules for what goes from the inside out and what comes from the outside in are hugely important to maintain the reactions that are going inside.
Alicia: I think we finally understood that because of the role of interfaces in computers.
Jim: Yeah, I think that helps. And I, being a computer guy…
Alicia: Therefore, what that also means, I think, is once you have a phase transition to a new dynamic, you have a new code. And a new code means simply the settings and the rules that govern that membrane, that boundary conditions, what it allows in and what it produces as waste and as action.
Jim: Yep.
Alicia: Absolutely.
Jim: And then the other interesting example you gave us, I had thought of before, is you talked about the architecture of the circulatory system as another…
Alicia: That’s Moreno and Macio, which is a really nice example. Because the vasculature of the body, the lymph node and the blood circulating system, really prevents the seeping out. But it is not an energetic force. That’s what the heart does.
But the vasculature is more the timing of the playground swing in that it controls the settings. After I submitted this MIT, which was, geez, it was almost a year and a half, two years ago, it took so long. I am fascinated recently by the inflammatory system and the immune system connection. Because they’re now talking about three levels. They’re talking about structure, function, and then regulation.
And that if my arm gets cut off, then the inflammation hits and immediately to try to repair the wound. If all of a sudden the interactions in the body are out of kilter, then the function of homeostasis may not work as well. That you get diabetes and so on and so forth.
But the idea recently is that perhaps there is a third level, and that is the setting of the functional system. I’ll call it the set point or the settings. And that perhaps, things like PTSD, chronic inflammatory disease syndrome, that kind of thing, is that the on and off switch for example…
The dimmer switch, the analog is screwy. So, it’s the set point, it’s the regulatory control of that function that’s often that maybe that’s the way to attack PTSD for example. There’s nothing wrong with the function of PTSD. We’re supposed to freak out if we think someone’s attacking us at night.
What’s wrong is the fact that we are now reacting in a different context the way it should have been before. Then that means there’s something wrong with the toggle switch. There’s something wrong with the switch. Which I find really interesting because I think that has a lot to say for social systems.
It has a lot to say, not just for the inflammatory system. Perhaps all of complex dynamical systems have those three layers, structure, function, and then regulation. And then, of course the question is how do these levels interact?
And they do not interact by efficient causes, they interact by constraints. That everybody who’s done hierarchy theory in biology is comfortable with that, I think.
Jim: Let’s move on to another topic here which is, and you described them as constraints, though I would not normally think of them that way. But I think the ones they work. And that’s the idea of scaffolds and scaffolding of affordances, etc. Talk about your thoughts on scaffolding and how you can use the language of constraints around that.
Alicia: I think you have the old-fashioned architectural scaffolds that are external artifacts that are temporary and that guide the construction of new buildings, correct? But if you think of context-sensitive constraints as conditions and factors that take a system away from independence, they link things together.
What scaffolds do, and I love the work of Bishop and especially the word of what’s his name? Wimsatt. They’ve done work on scaffolding long before they got fashionable recently. They provide a temporary equilibrium point from which to take the next step. It’s like a ratchet. I think of scaffolds almost like ratchets.
They provide a temporary mental stable position from which the next step whose direction the scaffold itself also suggests can be more easily taken. But that chapter you’re thinking about from the new book… There’s so many other different types of constraints, of that kind of constraints.
There’s entrenchment. That’s a hell of a constraint that we use a lot especially in social systems, correct? To retard any innovation or buffers. I think probably the difference between a buffer and an entrenchment might be how long it lasts.
But it’s a way to control the relationships between the inside and the outside and the next step. That’s why I think of it as a form of constraint. Because the scaffold again, is not… My [inaudible 01:07:48] is always efficient cause.
I’m always thinking, well this is something that has effects, but it not as an efficient cause. So, I have buffers. I have entrenchment. I have scaffolding. I have that kind of process that we use a lot.
Jim: And now you could throw those in with catalysts and what a…
Alicia: Absolutely.
Jim: And certainly let’s say scaffolds and catalysts both have the effect. While they don’t necessarily provide energy themselves…
Alicia: Correct.
Jim: … they lower the activation energy for something to occur.
Alicia: Correct. And even those scaffolds that are not temporary, like the flying buttresses of gothic cathedrals that end up being part of the structure. And that is also true of these lattices they implant that are embedded with nutrients that promote bone growth. The point being that the location and the direction of the holes in that lattice are what pattern the bone growth.
But then they end up getting absorbed and becoming part of the bone itself. So, these are all forms of affecting consequences that are not, or that are in addition to. I don’t want to discount efficient causes, obviously. I just don’t believe they’re the full story.
Jim: And this, in some sense, the basic laws of physics continue to be true. But there’s much more interesting structure being built that’s in addition. And that’s what the naive reductionism misses. Is that these…
Alicia: Correct. Go ahead. I’m sorry.
Jim: And I was just saying…
Alicia: I interrupt. I’m Cuban. I’m sorry.
Jim: You talk with your hands, right?
Alicia: That’s right.
Jim: And again, I think that it’s kind of this myopia of reduction over reductionism. Nothing wrong with reductionism.
Alicia: No, there’s nothing wrong with that.
Jim: You need to know both the dance and the dancer. But there’s the sense that somehow the relational dynamical are not real. They’re every bit is real as the primary properties.
Alicia: And that’s my problem with meteorology and nothing but [inaudible 01:10:14] and so on. The problem that everybody complained about top-down causation is not possible is because it would violate physical closure and it would violate the conservation of energy.
Sure, if you think of it as efficient causes, of course it’s going to do those two things. But if it operates as constraining dynamics, you’re not violating physical closure or constraint or conservation. So, that’s why it works very nicely. It’s not violating any basic physics tenets.
Jim: Yeah. Why don’t you do a little riff on that? Because that is one of the questions in complexity. And it befuddles the layman in particular.
Alicia: Well, what?
Jim: Top-down causality and how you can have top-down causality and no magic needs occur.
Alicia: Well, in the very same way that homeostasis changes my glucose production. Or how does the culture affect me? That is top-down causality. That is causality from the whole in which I am embedded. If I were not part of that culture, it would not affect me correct?
Jim: Right.
Alicia: So, that means that the constraint dynamic of the culture in which I’m embedded, be it the college at which I taught, the society in which I live, the family to which I belong. The constrained structure of each of those organizations changes the likelihood of different behaviors that might otherwise have been open to me that are not.
In the very same way, that once entrained into a Bénard cell, the molecule of water has different probabilities of where it’s going to go. Because the constrained structure of the Bénard cell affects… And that’s what I mean by top-down causation.
So, there’s nothing magical about it. But it’s not efficient causality. If you think of it as efficient causality, then of course it’s magical.
Jim: Got you. And yeah, there’s been many pointless conversation on this as you no doubt have experienced.
Alicia: Absolutely. Well, my centuries of this going on.
Jim: Indeed. Let’s move on. We’re getting kind of late on time here. Got about another 13 minutes. And this was something new to me, very interesting to run across something new. And that’s the idea of many to one transitions. Maybe you can dig into this in some depth.
Alicia: Once behaviorism got put to bed, then functionalism came into play or the identity theory. So, the idea was that mind is to brain as a computer software is to its hardware. So, then people said, all right, just like a lot of different…
Microsoft Office can be run on Apple and the different hardware devices, then perhaps that’s the explanation that gives some legitimacy to the notion of a mind. Vis-a-vis the notion of the brain. The brain is the hardware. The mind is the software. Fine.
But there was always the idea that the notion of convenience and Donald Davidson’s term is there will be no change in the supervening properties without any corresponding changes in the subvening, in the hardware. So, there was always implicit in the notion of supervenience, a one-to-one relationship.
And the reason I think that was true was because I don’t think they ever got away from efficient causality. So, there was always that one-to-one relationship. So, the idea was, all right, so a mental event, my thinking of my grandmother, my pain in my leg, will always be correlated one-to-one with this particular neuro pattern in the brain, which indicates pain in my leg.
This one about indicates that’s my grandmother. But very quickly, multiple realizability. Many to one relationships all of a sudden came about. And that was that, well, lo and behold, if this part of the brain is excised, the hearing function might be taken over by the other part of the brain.
Another part of the brain that was supposedly not at all dedicated to hearing, but all of a sudden it was. And I almost wanted to name the title of this new book In Praise of Degeneracy. But I thought, no, don’t get cutesy here. Because biologists have always been very comfortable with degeneracy.
That is, there are many ways for different amino acids to produce the same protein. That’s what I mean by many to one. Lower level paths to realize the same emergent property. That’s what I mean by many to one.
And that seems to be true in general for the higher level properties of complex dynamical systems. An economy can stay itself despite many different varieties or many different configurations, as long as that overarching constraint structure remains within a certain range. Have I made sense here?
Jim: Yes-ish. Let me drill it a little further. But you also talk about many to one. Because we do have multiple realizable domains. And the idea of degeneracy is very important. And for the audience, degeneracy basically means that things with different forms can have the same function.
Alicia: Correct.
Jim: Two catalyst…
Alicia: Same lower levels, same higher level.
Jim: Same higher level. And there’s a bunch of famous examples in biochemistry where it’s more famous. But you also talked about many to one as something analogous to dimensional reduction in systems. And think about the fact that with the example I gave before. Jim gets up out of his chair and goes down the hill to the ice cream store.
There’s lots and lots and lots of predicate signals that probably led to that. And they eventually got concentrated down to a single decision. Should I get up, go down the hill, get an ice cream cone? Or should I go to the freezer and pull out a frozen yogurt bar? And the fact that there’s many, many, many inputs, but there’s a single decision around the affordance was also the way you described many to one.
Alicia: The reverse of it is pluripotentiality. Which is one lower level that has the potential for becoming much more different functionality. And of course, stem cells are the example of that. They are pluripotent potent. They are not totipotent, which people thought they might be for a while.
That they could become any other. Not quite, they’re totipotentiality. But that was the problem with supervenience. Because then the question was how can the same neural processes in the brain produce such very different, differently propertied functions, higher level properties?
And Davidson said, “If you can’t identify a causal relationship between the two, then you might as well go back to behaviorism.” And at that point, Jaegwon Kim at Brown, who had been a big advocate of supervenience at the time, decided supervenience won’t work.
It either is one-to-one or it’s not. And that’s why he titled that. Was it a paper or was it a book? Called Descartes’ Revenge. Because then again, the problem was how do you get top-down causality? In that once you decide to go to the fridge, Jim, you could decide to walk directly to the fridge.
You could decide to go out outside because somebody is going to watch you take something from the fridge and you go outside, you go all the way around. So, there are a lot of different ways to implement that top-down decision, correct?
And the problem with, again, efficient causality is that it only worked for instantaneous relationships. Efficient causality cannot handle what used to be called in philosophy, standing causes. In other words, I decide I’m going to write a book, and it took me two years to do so.
So, how on earth does that intention continue enforce and continue exerting an influence all throughout those two years? That’s what I meant in that sense by multiple realizability top-down.
Jim: And then of course, that goes back to the very classic Greek philosophical question of the ship of Theseus. Which for our audience goes, okay, the supposed ship that Theseus took to go to the Creta, I guess it was, then it was preserved in the Athenian Harbor.
And over the years, the boards rotted and they were placed one by one. Eventually every plank on the ship had been replaced. There was nothing original. But it was the same ship. Or was it?
Alicia: Correct. And we can say the same thing about organizations and cultures and societies.
Jim: And humans. Basically, every one of our atoms gets replaced what? Every year or something like that.
Alicia: Whatever. Seven years every cells or whatever. Exactly. So, I think I stick my neck out more than maybe I have a right to. But that’s why I wanted to show in this new book that people say, oh, you are reducing everything to physics and chemistry. I go, “No, I’m showing that there are, I never know the difference, homologous, analogous constraint dynamics that operate all along the spiral.”
It’s not a reduction. It is to show that once you do the phase transition from physics to chemistry, then you’ve got new emergent properties, new codes, new everything. Once you go there from chemistry to biology and so on down the line.
Jim: Got it. Well, let’s wrap up with the future of these fields. And you talk a fair bit about the relatively new 4E approach to cognitive science. Which I think is maybe getting closer to your way of thinking.
Alicia: Yes. I mean, I really like the work of Depaolo, probably Andy Clark started the whole thing with the Embodied Mind idea. And people like Merlin Donald, whom I love his book. All of these folks started realizing the mind just ain’t in the brain. It takes me forever to write, because I literally realize that I work out the ideas as I write.
Now I type. But it’s not that I sit down, think it all through, and then I just write it out because it’s all been worked out in my mind. So, the notion that our minds extend beyond the boundaries of our body to artifacts, to tools for example.
I’m the designated teach your grandchild to parallel park person because their parents are petrified. And nobody else wants to do it. And I’m a good parallel parker. I say, “Look, after a while, you realize that you can almost feel the car as the extension.” You know exactly where it’s going to fit in a tight space.
So, the idea that the mind ain’t just in the brain, it’s embodied. But then it’s also enacted. Because it’s not just that it’s embodied in the agent’s body, but it’s also the mind is enacted in their behavior within a particular context and so on.
What I try to do in this new book is then the question is, well, how does this holistic ecosystem within which this embodiment and alignment, how does that coherent dynamic come about? And that’s why I want to show that before you have an enact…
I don’t think that if you are not Japanese, a tatami affords sleeping. I don’t think a Victorian throne enacts seating to somebody who’s never seen it before. So, to somebody who’s used to sleeping on the ground, I don’t think it does it automatically.
It is part of a whole ecosystem, a whole coherent dynamic. And my question is, how do those coherent dynamics come about? And again, my answer seems to be, I don’t know what else to call them, constraints. All of these processes…
Jim: Yet accumulation of constraints over time. And they’re channelized…
Alicia: But it’s not just accumulation in that one damn thing after another. It’s how they interweave with one another. That’s what creates this overarching interlocking set of constraints.
Jim: Yeah, and those constraints are real, in some sense.
Alicia: Are real. See, I say from the beginning in the book, “Ever since Kant, every time something gets bored, we all hide behind epistemology.” Oh, well, it’s the way we make sense of things. No, damn it. I think that’s the way reality works. I have tenure. I don’t need to worry about pleasing somebody else. I can say this. If it’s wrong, it’s wrong, but that’s okay.
Jim: All righty. Well, I really want to thank Alicia Juarrero and her new book Context Changes Everything. As you can tell by our conversation, I learned some things. I had some new ideas I’d never been exposed to. I thought it was really interesting. And despite what she said, I thought it was damn well written. So, thank you, Alicia.
Alicia: Thank you so much. [foreign language 01:25:22] next time you’re in D.C.
Jim: All right. I do get up to DC from time to time. Though I do refer to it as Babylon, so I try to get up there as little as possible.
Alicia: No, it has gravitas.
Jim: Gravitas.
Alicia: You used to have gravitas
Jim: Actually, I was up there for a scientific conference in May. It was downtown. I was amazed how little traffic there was, and how few people were around.
Alicia: It’s horrible. It’s really sad. It’s really bad.
Jim: They never come back. Anyway, I’m going to wrap her up right there.
Alicia: All right, thank you so much. Bye-bye.
Jim: All right, bye-bye.