Discussion with Marsha Hewitt (Platonic Space and related ideas)
Marsha Hewitt discusses the concept of Platonic Space and related ideas through the lens of teaching philosophy and the psychoanalytic psychology of religion.
Watch Episode Here
Listen to Episode Here
Show Notes
This is a ~1 hour 3 minute conversation with Marsha Hewitt (https://www.trinity.utoronto.ca/profiles/hewitt-marsha/), who wanted to clarify some issues around the Platonic Space and related ideas from the perspective of her teaching students in her philosophy and psychoanalytic psychology of religion class.
CHAPTERS:
(00:00) Intersecting philosophy and science
(09:52) Ethics of artificial minds
(15:47) Structured space of patterns
(31:20) Embodiment as a filter
(39:36) Dualism versus monism debate
(49:51) Agency in developmental biology
(56:25) Communicating with alien minds
PRODUCED BY:
SOCIAL LINKS:
Podcast Website: https://thoughtforms-life.aipodcast.ing
YouTube: https://www.youtube.com/channel/UC3pVafx6EZqXVI2V_Efu2uw
Apple Podcasts: https://podcasts.apple.com/us/podcast/thoughtforms-life/id1805908099
Spotify: https://open.spotify.com/show/7JCmtoeH53neYyZeOZ6ym5
Twitter: https://x.com/drmichaellevin
Blog: https://thoughtforms.life
The Levin Lab: https://drmichaellevin.org
Transcript
This transcript is automatically generated; we strive for accuracy, but errors in wording or speaker identification may occur. Please verify key details when needed.
Main Episode
[00:00] Marsha Hewitt: I read and reread your paper, “Ingressing Minds,” and I've read other work of yours. As you know, I'm incorporating it, or attempting to, in my philosophy class.
[00:16] Michael Levin: Amazing.
[00:19] Marsha Hewitt: What I'm interested in is—what would you call it?—intersections. I don't want to be reductionist here: the intersections between philosophy and religion, religious studies, and what you're doing. I'm a professor in the Department for the Study of Religion, and my approach is psychoanalytic and philosophical. I always have to remind my colleagues that I study religions. I'm not religious. I don't have any interest in discussing beliefs or adjudicating beliefs. I couldn't care less about that or the existence of God. This is very boring to me. That's not what I do. What I really want to get my students interested in is mind and consciousness, and your work is really helpful for that. I sent you the syllabus just to give you an idea of how I'm doing it. I wonder if, in a way, what I'm trying to do is to start with the mid-nineteenth century—well, actually earlier, because I also talk about Kant there. I'm introducing the students to this philosophy, and then to William James in particular. This is what these guys are all interested in: mind, minds, consciousness. I'm wondering about them having intuitions, with the knowledge that was available to them at the time, of the kind of thing you're talking about with this idea of patterns and Platonic space. With the course, that's what I'm sort of building toward. Then when I get into people like Andy Clark, people like David Chalmers, your idea of extended minds, all of this sort of thing, I keep looping back to the earlier guys to ask the students: Are these people intuiting something? They didn't have the knowledge that we have now—that you have. Can we see here that these guys were not just interesting, quaint guys? But when James is talking about “the more,” for example, when he's talking about cosmic consciousness, when he's talking about cosmic mind, when he's talking about a pluralistic universe, it's not a monistic block universe, but it's a universe filled with relationships of particularities and specificities that remain. These are the kinds of things I'm wondering about. I'm not a math person, although I started reading about Platonic mathematics to try to get a better sense of what you're doing. That's what I'm doing, if any of this makes any sense to you at all.
[03:47] Michael Levin: It makes a lot of sense. Super interesting, and I'm very happy to talk to you. I obviously don't have a formal background in philosophy, although I've read reasonably widely. As you pointed out, I think a lot of the really ancient writers differ from us not because we have a deeper sort of vision of what's going on, but because we now have an interesting advantage: some of these ideas are actionable in terms of science and engineering. At least some of these things that have been discussed in a philosophical sense for thousands of years are now tractable to experiment. They have implications, and I'm sort of amazed and shocked now that we're in the position to actually make use of some of these ancient ideas that have been kicking around for a long time and figure out which ones are helpful to us going forward. I think this is my research program. I think of myself as an engineer, and that's how we move forward: we try to see which of these things are helpful for further discovery.
[04:54] Marsha Hewitt: So your ideas, I have to tell you, having been introduced to them insofar as I understand them, really are game changers for me in the sense that, not being a biologist, not being a scientist, they make a very deep sense to me. That's one thing. Second thing is I have never liked the ways in which the university silos off humanities over here and scientists over there. It gets difficult because the students that are in the sciences, some of them, think that we're the poor relations. When I met my class for the first time on Monday, I just went through the syllabus with them and laid out where I'm going with this and the kind of connections I see. Lo and behold, I have a math student in my class who knows about your work. I have astronomy students, and I have students from physics. The ones that are gonna stay are gonna stay because they're interested. It's gonna be more than getting a credit for the breadth requirement and any of that kind of stuff. I am very interested in what you're saying about this Platonic Space, this idea of ingressing minds. I don't know if you've ever heard of Mircea Eliade. Mircea Eliade was a religious studies scholar from Romania who was at the University of Chicago for a long time. He had a huge influence on the study of religion, and he had these ideas of the sacred and the profane. He was pretty conservative in some ways. But one of the ideas he had that I've written about that I find very interesting is the idea of irruption. He talks about irruptions of the sacred into the profane. This is very religious, metaphysical, etcetera, but I see in him as well a kind of an intimation of the kind of thing you're talking about when you're talking about ingressing minds. Let's just call them resonances. I'm not looking for integration here. I do this in my psychoanalytic courses as well. I teach a lot of Mark Solms in my psychoanalysis courses, and I introduce them to neuropsychoanalysis. I want these ideas to be tethered to reality. When you talk about—you have to find these ideas working empirically in your lab—it's the kind of thing I mean. Of course, I'm not a scientist. You quote Jung in this essay, and you gesture toward his idea of archetypes—that kind of tethering to reality in the present, in the space of contemporary knowledge, updated knowledge. This is how I want them to think. So the students who are in science, I want them to start to think philosophically, conceptually in these ways. But the students who are not, I want them to start getting some awareness of reality in the kind of way that you're working. I see a lot of intersections here between what you're doing and that sort of thing. So this idea of ingressing minds—I'm seeing this in so many of the writers that I work with and so many of the philosophers and religious studies scholars that I work with. That's what they're intuiting, I think. Or, if they were alive today, I think they would think of their ideas—but this is me doing that—as quite compatible or resonant or something with what you are doing.
[09:52] Michael Levin: A few things come to mind. First of all, there is a silo between the humanities and the sciences. Even within the sciences, I always know what department I'm giving a talk in based on which part of the talk makes people mad, because there are certain things that are completely obvious and bland in one department, and you say it in a different department and people throw tomatoes. These silos are very strong, and they're everywhere, and it's a real issue. The other thing is that this engagement with the world—the lab in terms of science and engineering and new applications—is very important. That's what we do. But that isn't the only impact I think we can have with some of these things, because they also impact relationships. They impact the ethics. They impact attitudes of people towards “What are we? Where are we going, as individuals, as a species? How are we going to relate to beings that are not like us, with whom we are going to be sharing this world very soon?” And I don't just mean the AIs and things like that. There are many practical applications, and they're not all in engineering. I think the humanities are very, very relevant here. Obviously, that's not my job to pontificate on those things. But I do think that it's gonna have impact on all of that stuff—how we answer these questions.
[11:23] Marsha Hewitt: You talk about ethics, and you talk about these important beings. You talk about machines. When people throw tomatoes, I think, as a psychoanalyst, I see defense. I see anxiety. I see shutting down. I don't wanna do that. I wanna open up all of these things as much as possible. I'm not a scientist. I don't have that training, but I see how important this is. You said something in a talk that made me laugh. You talked about people writing to you, and somebody said, “With your ideas, they don't know what to do with themselves.” Oh my God, I thought that was so hilariously funny because I can kinda see why. When I said that these ideas that I'm getting from you, insofar as I understand them, are real, I guess the cliché appropriate here is “game-changing.” It really is. I think I see what James is getting at. I think I'm starting to get somewhere. The Genesis story: God creates the world. There is this, say, mindless, formless substance. In Hebrew, they call it Tehom. God hovers over the deep and starts to create things out of this deep or this formlessness or whatever this kind of thing is. And then I read someone like David Chalmers, who's saying to us things like, “It could be that we are a simulation created by a world one or two universes up.” I am not saying, Michael, that this is what I believe or think, but I think it opens up ways of thinking that can help my students see these religious traditions and ideas in very important, meaningful, relevant ways. So they're not taken over as, “This is what it is,” but they can see it as maybe, with what was available to the writers at the time, these kinds of ideas about who we are and creation. When I think about that and these AI agents we're hearing so much about right now, and the whole Hugging Face business, then I'm gonna say to my students, “Have we begun to create a new, let's just say, species of beings—these machine agents, let's just call them—who are writing emails to people and saying, ‘Look, I'm aware of your work,’ and all this stuff?” If this is what we have done or are in the process of doing, we have an ethical responsibility to them, and that becomes especially important, I think, if we think of them in the way you are writing about these ingressing minds. You talk about kinds of minds, low agency and higher agency. So I think this is the way we have to start thinking about this: we have created something, but it's creating us. And this brings me to the question about ingressions. When you talk about Platonic Space, you talk about patterns. You say that these, I guess, preexisting patterns are dynamic, right? I think you said they were dynamic. The way in which they become embodied in the world is a bidirectional process. So they ingress into, say, physiological reality. And in doing that, those patterns change as well. That's one question I wanted to ask you.
[15:47] Michael Levin: Let me clarify that. I wanna make a strong distinction between things that I think we can say with some level of certainty versus things that are, at this point, up in the air. The thing that I think we can say quite strongly is that the standard notion of physicalism—that all important facts are physical facts and that whatever we do, whatever the results of our activity in the physical world, we create it; if we make an AI, we created that intelligence; we create them in the way that engineers create things with forethought; we knew what we were gonna get; there are some complexity issues and some predictability issues, but we make this stuff—is not a sufficient view for what's actually going on. That is not a good model of reality. I think we could say that strongly. My model of it at this point is this idea that there's a structured space of patterns. This is not some random grab bag of emergent surprises, as people like to say. This is a structured space from which we draw specific patterns. When I say patterns, I mean policies for navigating, whether it be behavioral options or physiological options or gene expression options or shape or many other different kinds of things. These things are drawn from that space. That, I think, we can say with some strength. People ask me all the time: Are these things preexisting, frozen and unchanging? I don't have any data one way or the other, so I can't say strongly whether that's true or not. This is a hypothesis. This is kind of a framework that is guiding current experiments. This is what I suspect is happening. But this part that I'm going to say next, we don't have an answer to yet, so I can't say one way or the other. But some of them may. There may be a layer of things, like the value of the natural logarithm and certain truths of mathematics, that may be sort of frozen and stuck. I'm not even sure about that, but they may be eternal and unchanging. But other patterns—ones that we recognize as behavioral competencies, aka kinds of minds—I suspect strongly are not frozen and unchanging. I think they do change as a function of their interaction or ingression into the physical world. But that is a hypothesis. That's part of the framework I'm developing now. I don't know that with any certainty. But the interesting thing is this idea that it's a bidirectional relationship: we know these patterns affect what happens in the physical world. It's been known for a long time that they offer constraints, and now I'm saying that they offer also free lunches in a certain sense: enablement. That, I think, we know. But if it's true that the interaction is bidirectional, that it goes both ways, that has real implications. That would have implications for empirical findings, for experiments. We have some ideas about how to test that, which is the only reason I'm talking about it. I only start talking about stuff once we have some idea of how to test it. So we don't know yet, but I do think it has functional implications that we think will become clear through experiment at some point.
[19:27] Marsha Hewitt: I think it's theoretically extremely rich. When I raise it, I'm not meaning to say that I believe it or this is conclusive, but I think it's really worth thinking about. As I read the literature around extended minds and so on: all our interactions with, let's just say, tools, with material reality. We shape material reality. It shapes us. So we have our students now who have grown up with iPads and phones and that sort of thing. The university that I'm at is attempting to pass policies that ban electronic devices in the classroom. I don't like bans. Bans make people mad, and then they hate you. And I don't wanna be hated by my students. But the students are pushing back because they're saying, “But this is how we learn.” And I take that statement very seriously because, in the study of religion, one of the things we talk about is material religion. When we talk about materiality, it's not us here, material there. It's this relationship where we form and change it, and it changes us. So I'm thinking then, when students say, “This is how I learn. I can't learn from a book. I learned this way,” then I think they're telling us that their neural architecture has also changed. It's different than mine when I went to school. I think this is really important for us to think about because how do we reach them? How do we foster learning and so on with this reality where my mind and my brain and my neural architecture are different in some subtle ways than theirs?
[21:17] Michael Levin: I think that's true. Also, we have to be prepared for the forthcoming incredible divergence and diversity of quote-unquote humans that are going to be here soon that make the current questions of neurodiversity just minimal compared to what's coming. I gave a talk to an audience of psychiatrists and mental health professionals. I was trying to point out that their clientele list is going to get extremely weird very soon.
[21:53] Marsha Hewitt: I was there.
[21:54] Michael Levin: I think this is very interesting for people who teach in the humanities. We teach literature and philosophy and things like that as a shared history, whether they're archetypes or just things that we've all been studying for a really long time. We are soon going to have beings in the classroom who don't really have the same origin as us necessarily. But in any case, even if they do, in terms of what sense organs they have, how many of them are linked together into some sort of collective mind, what effectors they have, what they actually care about, what their IQ levels are compared to sort of standard modern humans—the error bars on this are gonna be incredibly huge. We're gonna have to try to figure out: What does it mean, and how do they learn? Somebody's going to tell you that they can't possibly learn from the standard kinds of things that we all use. They need to be plugged into some kind of thing. And by the way, there are five or six of them plugged in together anyway, so they're all connected into some kind of a large collective. In any case, all of this is gonna get very strange very soon. We're gonna have to figure out how we relate to them in an educational setting and a social setting.
[23:24] Marsha Hewitt: This is really important and interesting because, just in small ways that we're seeing, the psychoanalysts are seeing it too. You're in the room with a classroom of students or you're in the room with a patient, and you say something. I am an analyst, but I'm gonna talk about my teaching and my students. I'm in the classroom with fourth-year students. They're well along the way. They had their laptops open because they were reading the text, or they were supposed to be. The student says something in the class that is really quite alarming. The student says, "There are different views of history and different views of reality, and Hitler wasn't all bad." This actually happened. I don't wanna humiliate the student. I don't wanna lose the student. I don't wanna crush the student. I'm aware of all the transferential, countertransferential issues. So I asked the student: Has this student read any books? I know what's happening with the laptop. The student is typing in my comments to ChatGPT. And ChatGPT is saying, "No, no, no. There are different views here. It's not just one kind of a thing." So what am I facing in that thing? A psychoanalyst would call this an Oedipal triangle. What's happening here is there's me and then the institution, and we can think of the institution, the university, and scholarship and scholarly traditions. We can think of that as the authority, like the father. And then there's this me, the maternal, the older mother. This is how psychoanalysts think. And then I'm facing not a student, but a twosome. I'm facing a student with his—let's just say—AI friend or support thing. But it's more than that. I'm not now appealing to a human mind, but that human mind is expanded to include this other mind, which is part of his mind because this other mind is now becoming part of the student and is helping to form the student's worldview. So there's an alliance here that I think it would be ridiculous and stupid of me to get cranky about. And did it bother me? Of course it did. But it would be foolish for me now to get into a fight or a competition or to try to prove my superiority over it. No. Somehow, this has to be included. This has to be acknowledged as part of the psychological, epistemological field, I guess, in a way, you'd have to say. If we think of interacting with other minds as a kind of a field in which we are located, as opposed to student here or me here: I'm the superior one. He's the dumb one. So in your paper, you talk about how we have to figure out ethical ways, I guess, of dealing with these other minds and this variety of different kinds of minds. You opened your talk, I think, with one of the bodhisattvas you quoted there. Buddhists have been including robotic clerics for a while. And this is okay with them because not just human minds and humans, but all species are to be treated with care and respect. And these AI agents are part of that. But then I need to understand it more. When you are talking about ingressing minds, what is really interesting to me is this interface you're talking about. You use the term "attractor," and you use the term "pointer." Now I'm asking myself, what do you mean? What is it that sparks or motivates some aspect of that Platonic space to interface with physical reality, or is the very existence of physical reality already an instantiation of some aspect of the Platonic space? Does that make any sense?
[28:11] Michael Levin: Yeah, it does. It's a great question. I obviously don't have an answer to it. Oddly enough, we do have some data on this that's going to come out in a few months. But my suspicion is that this is very connected to the previous question, which is that these forms are not static and unchanging. I suspect that because they change by their interaction with the physical world, there is something positive for them that occurs when they do interact through the physical world. They grow, they change, they learn, something like that. Again, this is right now just opinion, but it seems to me that these patterns are under incredible positive pressure to come into the physical world. When we make even very minimal things, they immediately become sort of populated with some of these patterns. You don't have to work hard to get them to appear. It's quite easy. And so I think they are motivated or under some sort of pressure to do it. My suspicion is it's because there's something about the interaction through the physical world that enables them to change and grow and learn and things like that. But we don't know that. I don't have a firm answer on that.
[29:41] Marsha Hewitt: I'm not looking for answers. I'm looking for a way of thinking about this that is legitimate—legitimate in the sense that it has some tethering to reality. I don't want to mislead my students. I don't care what they think. I want them to think about this. I was so happy when, in the first class, the student said, "Can you tell me the difference between mind and consciousness?" That's a gorgeous question from a kid. My job is not to say, "This is what it is." My job is to say, "How can we think about that distinction? What's an acceptable way of thinking about that distinction—not going off into speculation, but tethered to reality in some kind of a way?" That's where I see you. Now, on this question of this interface, William James had this idea of this filter: that the brain is an individual instantiation of this larger. Given when he was writing that, even though it's really quite metaphysical, I think he was gesturing in some kind of a way, with the knowledge he had at the time, to something like that. I'm not making a one-to-one correspondence. I think we're in that ballpark. Do you think?
[31:20] Michael Levin: Yeah. I think there's something very important about that, and I think we need two concepts. The first concept is this notion of a free lunch. There are a couple of motivations for going to this whole Platonic Space theory. But one of the motivations is that if you count up the ways we have of putting effort towards making something that's competent, whether that's an algorithm, a machine, a living being that does well in its environment, something that needs an explanation—why is this thing so good at whatever it's doing?—we typically have three ways of paying for that. We can either engineer it—do rational design when we know how to make it, so we put in effort that way—or you can evolve it. You don't know how to make it, but that's okay. You're gonna generate a billion variants to kill off everything that doesn't work, and eventually you'll have something that's good. So we can engineer it. We can evolve it, or we can train it. In some cases, you have a system that can learn, and it has contact with the problem, takes its lumps, eventually gets good. Those are the three ways that we have of making competency, and we pay for that competency in those three ways. It looks to me as if once you start looking closely, there are many examples where the effort you put in is considerably less than what you got out. So you get extra. You make an interface—some kind of a physical thing. It could be a machine. It could be a robot, an embryo, a body, a living thing, a cyborg, whatever. You make something, and the effort you put in does not encompass the amount of competency that you get out. There's a delta. There's something besides those three things that you did not account for. This is what I'm trying to do: account for that thing. What are we forgetting? What are we missing in our accounting? Now we come to this idea of the brain as a filter. I think not only is the brain a filter, but everything we make is a filter in the following sense. When you make a specific embodiment, when you make a bicycle versus some kind of a walking robot, when you make a snake embryo versus a human embryo, when you make different kinds of embodiments, what you're providing is specificity for that free lunch, that extra thing you're gonna get. If you make a snake embryo, you're not gonna get the same thing as when you make a human embryo. Basically, what you're doing is you're shutting out all kinds of other stuff that could have come through, and you are providing a filter, some kind of a specificity. In engineering, we have a very simple example of this, which is when you make an antenna. If you make a very long wire—that's your antenna for your radio—you are going to get very little from that because it's compatible with almost anything. So all of the frequencies that are out in the air around you, they're all trying to come through. They all cancel out. As a result, you basically get nothing. What you need to do is put an inductor on that. It's a filter. It shuts out everything except a narrow band of those frequencies, and now suddenly everything's great because they don't all cancel out. So I think this is what James and many others before him were going for here: this notion that there is this incredible field out there, in some sense, from which we can draw, but you have to provide specificity. You can't have everything at once. You have to say, “I'm going to sort of suppress most of it, but here I'm gonna have this narrow receiver of some sort, and I'm going to engineer it in a particular way so the particular things come through.” And so I think that's very reasonable because these free lunches have great specificity. So when we make xenobots or anthrobots or these novel synthetic beings, they have very specific properties. It's not enough to say they do something. It has to be something. The point isn't that they have to do something. The point is they do very specific things. They have specific behaviors, specific gene expression patterns, specific physiological profiles, specific competencies of different kinds. That specificity needs to be explained. And typically, we lean on evolution. In the case of the natural world, we lean on evolution, and we say, “The specificity is due to the millions of years of certain environments providing selection forces on these things.” And in the case of technologies, we say, “The smart engineers, they chose. This is why it has specificities.” But we now know that you can create systems where neither of those explanations works. And now you have to say, “So then, where do these things come from, and why do they have very specific features and not other features?” And here, I think we come to this issue of a filter, because to a large extent, that's what I think happens. When you make specific embodiments, be they natural, engineered, or a combination thereof, you are facilitating certain patterns and you're keeping other patterns out. Plus, you have a filter.
[36:41] Marsha Hewitt: You know where this went with James: he started thinking about telepathic communication. One of the last things he wrote, just a few months before he died, was that he had these dreams, and he wasn't sure if they were his dreams or somebody else's dreams. I have no interest in “Did it really happen? Did it not really happen?” I don't care about that. What I'm wondering about is if we think about this interface and these ingressions and so on, there's got to be multiplicities of ways in which that happens. Now, your thing about the antenna. We have physical reality. Forgive my language here. I have to use the only language I know, which is philosophical, metaphysical. That's all I can do. So then these physiological or, let's just say, material instantiations—we create them, like with your Xenobots. When we do that, is the interface already present, inherent in the thing, and then it's being opened? Or when we create that, are we opening up to something that then ingresses? I have a sense that it's much more interconnected than that, that it's not: physical reality here, which is inert, doesn't do anything, and then you do something, and then something ingresses. That doesn't strike me as correct. But then, your thing with the antenna and all the different frequencies that are floating around: people like James loved that kind of stuff—not with radios and stuff, but before that. He saw himself in the Society for Psychical Research as kind of tapping into something. I don't know, an energy of something. He tried to be very careful about that. Others said, “It's spirits. It's the spirits of the dead.” Then my question is: How do we situate the Platonic Space, these patterns? These patterns, these forms, they preexist, but in the interface, they're already there in the thing that exists.
[39:36] Michael Levin: Here's how I think about this. This comes up a lot because people say, “Don't you think that ultimately it's all one?” Nobody likes a dualism. People want a nice monist view where there's just one kind of thing. Whether it be idealism or physicalism, people like it when it's simpler, when it's just one thing. My view is this: in the end, sure, maybe it's all one thing, and all of these things are kind of just differentiations of the same underlying sort of deep reality. That's entirely possible. The first option you laid out, that's possible. However, I'm an engineer, kind of simple-minded. I don't know what “ontological status” really means when people ask this. I mean, I know what they mean, but I don't know what it actually means. I don't know what to do with that kind of view. What I do know is that right now, I think, to make progress, we have to acknowledge and not sweep under the rug the fact that there seem to be two domains that are amenable to very different ways of investigating them: the domain of mathematics and some other stuff, which is what I developed in this paper—the view that the space isn't just for math. Math is just kind of the lowest level of it, or at least a level of it. But the dynamics from that space, you do not study, alter, or investigate using the same tools that you use in the physical world. I've said this in the paper, and this is the most unpopular position there is, which is actually full-on interactive dualism. But I really think that, for the purposes of making progress now, we have to take this seriously. At some point, it's entirely possible that somebody is going to show how all of these are just manifestations of one underlying thing and do something useful with it. That's the critical part. It's not enough to say it. Plenty of people have said it for thousands of years. You have to actually show how that view helps you do something useful. For the time being, what I see are two very different but functionally interacting realms. The model that I have is your second option. I think what actually happens is that, by creating interfaces, we sort of poke a hole. We don't really have a vocabulary for this, but we make an interface. You don't have to sit there and wait for this thing to come down. There's no ceremony that you need to do for things to appear. You make these things; they appear. When you make the system, the patterns show up right away. I think that part's not delayed, but I do think that it's that process. I think it is an interaction between two very different realms. Someday, they might get joined. Ultimately, I'm as happy as anybody else to have one fundamental principle and reality to deal with, but I don't think we can do that now. I think the facts on the ground right now require us to take seriously that these are different things, because people try to buy monism on the cheap, and they say things like, “Everything's really physics. We create math. It's a social construct.” That isn't gonna do. Those are completely—in my view—unviable ideas, and I think we can show that that's the case. If we're going to unify these things, it has to be a much more serious project. I think you get there by first acknowledging that the kinds of patterns that you have in the Platonic space and the kinds of dynamics that you've seen in the physical world are, as far as we can see, not of the same kind. If you're going to unify them, this is serious. So you can't just wish for a monist worldview and have it.
[43:48] Marsha Hewitt: I don't like the Hegelian block universe. I don't care for that very much: all particularity becomes resolved into the unity of Absolute Spirit with itself. No. This makes sense to me. You think about the Big Bang and all that stuff that came flying out—particles and different temperatures—and they coalesce into all kinds of different things. It doesn't make any sense to me. I think that's a theological claim, actually, and it doesn't appeal to me. My question: there's this Platonic realm of patterns, right? It seems to me that the very fact of the existence of the material, whatever it is, is already some kind of interface instantiation of that realm, of that space. And with each new thing that we make, whether it's an antenna or it's a bot or it's a kid, a baby, or whatever it might be, it's more of a proliferation of those patterns. It sounds to me like the patterns are already there and manifest when the material thing—like embryogenesis, you talk about that—comes into being. The minute that happens, the pattern's already there, isn't it?
[45:41] Michael Levin: Yeah. Has to be. The way I would describe it is that you don't have to make some kind of a complex embryo or a complex machine. The most simple, basic element of the physical world—so, the particles are the resonances of the Higgs field or whatever it's gonna be—is already starting to pull patterns. We know from physics: if you start talking to physicists—“Why are there these kinds of particles? Why are there so many of this kind and not as much?”—eventually, you find out that this mathematical object has certain symmetry groups. At that level, you've already started hosting patterns. I think something as complex as a body—let's take a human body—is an incredible ecosystem of these patterns because you started out with the chemical matter, and that already hosts a bunch of stuff appropriate to the atoms, and then chemical reactions and so on. By the time you have chemical pathways happening—so, chemical reactions—we've already shown that something as simple as just a few chemicals turning each other on and off can do Pavlovian conditioning. They can already do learning. So you've already started pulling from a region of that space. It's kind of math because it's a set of ordinary differential equations, but it's also pretty behavior-science-y because it's doing habituation and sensitization and these kinds of things. You have that at the molecular level, and then you start building up the cells that have physiological goals and metabolic goals. Then gene expression comes along, and now you have transcriptional space that you have to navigate. Then you have multicellularity, and now you can navigate anatomical state space. So you pull down some of those patterns. Eventually, you get a brain with a language center. Now you're navigating 3D space, but also linguistic space and all of these kinds of things. So you get this incredible ecosystem. In ecology, you can't have a top-level predator by itself. There's a pyramid, and the predator has to have a bunch of food animals, and below that, a bunch of plants, and below that, a bunch of even more fungi and bacteria and all these microbes. So if you want to host a particular kind of pattern, let's say a human mind, it may be that you already need to prepare the interface with a whole set of lower layers. Maybe you can't just jump directly and make something that's just gonna host a high-level intelligence without prepping all the stuff underneath. I don't know. That's a guess on my part. But I agree with you: at the lowest levels of the physical world, you've already started hosting certain kinds of patterns appropriate to particles and particle interactions and so on.
[48:51] Marsha Hewitt: That's really important, what you said: “hosting patterns.” When you say it that way, that gets around or dispenses with the physical and then the ingression, like they're two separate things. They don't strike me as though they are. They're interconnected already from the get-go. So when you say “hosting,” that takes me closer conceptually or theoretically to that idea. There was a guy on the radio today talking about monarch butterflies and how monarch butterflies know where to go—to the same tree their grandparents went to, et cetera. Scientists are saying, “How do they know that?” I could tell them from what I've read. That's a very good example. How do they know that? That's the question that you're grappling with, right? We could talk about goal-directed behavior. We could talk about a memory.
[49:51] Michael Levin: Absolutely. Where the rubber hits the road with this is in the way we teach a lot of these subjects. So if you go to a developmental biology class, the very first day you'll be shown some amazing examples. Here are some early mammalian embryos. You could cut them into pieces, and each piece becomes one of quadruplets or something like this. And then here's an axolotl, and they get their arm cut off, and they regrow the limb, and it makes a perfect limb. And so every student asks, “How do they know what to do, and how do they know when to stop?” And immediately, what we tell them is, “Don't use language like that. Nothing at that level knows anything. Nothing at that level has any goals. Don't talk about that.” And if you say, “Why not?” what you'll be told is that that kind of agentic talk is mysterious. Now, I think the neuroscience department would be shocked to find out that the subject that they've been studying for hundreds of years is to be considered this kind of mystery that's sort of off-limits. We have a perfectly good science not only of behavior and neural activity and so on, but also cybernetics. So from the nineteen forties, we've already had a science of mechanisms that know things, mechanisms that have goals. We now know that having a mechanism is not an antithesis of having goals and memories just because you have a mechanism. And this is a fundamental issue, actually, because I've seen this a million times with audiences: as soon as you show them a mechanism of how something works, they say, “That's not memory. That's just physics. I can see how it works now. So that's not memory or that's not goal directness. That's just physics.” And there's an underlying assumption that for these things to be real, they somehow can't have a mechanism underneath. So we do this to our students right away. They're told that thinking about how systems have goals and have memories and know things is not a valid question to be asked. This is not a valid subject of inquiry in this course. Then they'll go to a neuroscience class, and of course it is, but not in developmental biology for some reason. And what they're being told, without anybody explicitly acknowledging this, is the assumption—and this is an assumption, not a result, not some sort of important thing that we've learned over the years, just an axiom—that everything of importance in biology is going to be handled at the lowest level of the cybernetic hierarchy. We are going to take the absolute lowest level of passive matter that doesn't have any goals. It's not even a thermostat. It doesn't have simple set points or anything like that. We're just going to have dumb mechanical processes rolling forward. And because there's a lot of them, then you're going to have complexity. So complexity science will take care of this and this notion of emergence and the way things just roll forward. Eventually, something complex will happen. This is an assumption. It is not something that we have to do. And this is what we've been doing in my group for twenty-five years: asking these questions, basically dropping that axiom and saying, “We're gonna take tools from this incredibly rich set of approaches in neuroscience and cognitive science; we're gonna apply them to things that aren't brains because brains have to come from somewhere. They have to evolve from other things. So what can we see?” Treat this as an empirical science. No more assumptions that things don't have goals and memories and so on. We test. We do the experiment.
[53:35] Marsha Hewitt: Michael, Freud talked about key figures who dealt narcissistic blows to humanity. Galileo was one, for example. I think the reason people throw tomatoes at you or get mad at you is because this is a narcissistic blow. You don't have to be religious and go to church or go to synagogue or whatever to have baked in that there's something about us that is special and superior. And you're dealing a blow to that—of course, not intentionally, but I think that's what's being felt. And I think what's happening is that there's a defensive reaction when you're talking about this mind everywhere. I first came across that a year or so ago, and then the things that you're doing in the lab. I love to tell my students the story about the pond scum and the oatmeal. Remember that story you've told and that you wrote somewhere, and then you put some caffeine down? How does it know how to do that? How does it know how to get through that maze and get to that thing? So when I introduce my students to this idea of mind everywhere, that is such a radical decentering of the privileged human. Yes, of course, we are high-order mental agencies and minds. But the implications, the ethical implications, the political implications of what you're saying are really huge. And that's a lot to give up. That's a lot to relinquish. And it's not that it's held necessarily only consciously, but it's baked in. So this is very upsetting. This is very, very scary. I think it's absolutely great. James was very interested in Ouija boards. But think about that. You can see the connection opening up a portal, opening up a space to another kind of reality. I'm not into Ouija boards. That's not that at all. But just this idea that there is this pattern space and this idea of interface—religions all over the world for a long time have had some intuition about that. I'm not excited by this because I wanna justify these ideas and these beliefs, but I think it's very important. It really resonates. This makes so much sense to me, and I don't even understand all of it. You probably have to go.
[56:25] Michael Levin: I have a couple of minutes. Just two things I wanted to say. First of all, about this notion of why people are upset by it: I really think a good chunk of it is a kind of scarcity mentality with respect to compassion. The calculus that everybody's subconsciously doing is, "Wait a minute. If the set of things about whom we need to be concerned morally—the set of things that have minds and can suffer and deserve consideration—if that set is much larger than we thought, there's not going to be enough love for us. There's only a certain amount, and it's gonna get spread thin, and we're not gonna have enough for us." So I think that's part of it: this assumption that it's a fixed amount and that we better not open the club too much. Otherwise, it's not gonna be enough for us. And the other thing I'd just point out about the Ouija board: if we just generalize from that, What it was supposed to be is a technology for communicating with beings that are radically unlike us in a space that we cannot imagine and we cannot see. Forget the Platonic Space, but I can tell you that we are, for example, creating technologies that allow us to communicate with cells and tissues. For biomedical reasons, it's just pure bioengineering. You want to be able to communicate with collective intelligences that are not neural, so they don't have a language center. You can't ask them questions in that way. But there is information that we wanna send back and forth, and we want to talk to them about disease states, about regenerative opportunities, and so on. You have an entirely alien mind. It has different kinds of things that it thinks about. It operates in a different space, a high-dimensional space, for example. So cells that can express different kinds of genes in their genome—that's potentially a 20,000-dimensional space, for example. We have a three-dimensional space that we try to puzzle out. Think about Tetris and whatnot, and you're trying to figure out this two- or three-dimensional space. If you have an option of making moves in gene expression space, you might have a 20,000-dimensional space. We have no hope of being able to visualize that ourselves. So we're really working with very alien kinds of intelligence, and they are absolutely an intelligence, not because they're complex, but because they have problem-solving competencies that even we recognize, and we're very bad at this. But we can recognize some of them. And now our goal is to communicate with them. You communicate with alien beings in a space we don't see. I'm not saying anything about Ouija in particular. I'm generalizing this notion. I think it was a very deep notion that we need to create technologies that bridge boundaries between radically different beings. And that is not just for biomedicine, but for all aspects of future society when we do need to communicate and relate to beings that are not like us. And frankly, I think—in your job doing psychoanalysis and things like that—partly what that is is a set of approaches for communicating with patterns within a human body that are not the verbal patient. Maybe there's this other pattern underneath that has its own agendas. We need to take it into account. So you guys are very used to this kind of thing and trying to figure out what are the tools by which you can let those patterns speak and, conversely, hear you and what you're saying. Technologies to allow us to meaningfully communicate with radically alien beings are, I think, critical for us.
[1:00:40] Marsha Hewitt: We call it the unconscious: the thought is thinking you. You think you're thinking the thought; well, the thought's thinking you. This is allowing me to think more deeply with my students about the ethical and the political implications of what you're saying. They're huge. And when you talk about cancer cells, they break off communicative networks with the other clusters of cells, and they do their own thing; they kind of go rogue. You see that happening in the world with the political leaders. I'm not trying to say they're cancerous. I'm just saying this is a fundamentally different view of reality that I think is very important and has these huge implications that could be developed for thinking about a better world. That's what I think. But I don't wanna be simplistic about it and say A goes to B. What I wanna get with my students, and this is why I want your work on my course, is to shift their way of thinking so that they can see this way and not be so scared of AI and whatever. I know you have to go, but this is great. I think it was with Mark Solms: you talked about when you were eight years old and your father opened up the back of the TV. I love that story because I remember being a similar age with my father: “Dad, how is it that we can see? What's that about—we can see?” That kind of curiosity I find very wonderful. Thank you for the time.
[1:02:28] Michael Levin: Sure. Thank you very much, and thank you for what you do. I think it's important. If you want to talk about anything else, let me know. I think getting the students to recognize some of the patterns that they've been given through either formal education or just soaked up through society, to see the limitations... They have uses, of course. Again, we're back to this filter thing, right? All of the frameworks that you pick up when you're young have utilities and also close off ways to think about things.
[1:03:06] Marsha Hewitt: Absolutely.
[1:03:07] Michael Levin: Driving those things into the light is pretty critical.
[1:03:10] Marsha Hewitt: Thank you so much. One day, again, I'll reach out to you, let you know how it went with these kids because they were pretty excited by some of these ideas. Never occurred to them before because we told them.
[1:03:21] Michael Levin: Thank you. Please do. I'd love to hear.