Ep 119: Biology as its own metaphor (with Phil Ball)
At what levels does causation happen in biology? Are metaphors useful for understanding biology?
In this episode, we talk with Phil Ball, a science writer who was also an editor for the journal Nature for over 20 years. Phil has written over 25 books, but our conversation focuses on his most recent: “How Life Works: A User’s Guide to the New Biology.” In the book, Phil covers a wide-range of topics from cells to proteins to biological agency, and makes the argument that traditional ideas and simplified metaphors in biology often don’t hold up. We talk with Phil about the concept of the selfish gene and unpack what it actually means and when it’s useful. Then we dive into the paradox that multicellular organisms are composed of multiple levels of agency, yet are complex agents themselves. Phil also discusses the biomedical implications of thinking about cancer as one in a many possible states that cells can inhabit across a landscape.
Cover photo: Keating Shahmehri
-
Art Woods
Hey, Big Biology listeners. It's fundraising time. Yep, sorry, but we gotta do it, cause we're a nonprofit, and we need your generosity to keep the good stuff coming.
Marty Martin
As you might have already heard, if you listen to us via Spotify, we were experimenting with putting ads into the episodes. In some upcoming shows, and then a set of older ones, you'll now hear two or three short ad breaks during the chat. This isn't the windfall, you might think, but it helps.
Art Woods
Second, as you've been hearing over the last few episodes, we're making increasingly aggressive pleas for donations and Patreon signups.
Marty Martin
We are not a big operation. But producer Molly, social media guru Dayna, our artist, Keating, and our webmaster, Steve, they all depend on your support.
Art Woods
So right now hit pause and go to our website, www dot big biology.org and make a donation.
Marty Martin
Or go to patreon.com/big bio, and become a patron for just a few dollars per month.
Art Woods
Patrons get cool insider stuff like access to behind the scenes audio and extras from our guests about their lives, their hobbies, and their careers.
Marty Martin
We're nearing the end of season six, and we very much want to come back for season seven. But to do that, we really need your help. Of course, we're reaching out to funding agencies like the National Science Foundation and a variety of nonprofits.
Art Woods
But right now, it's super tough to get funding from those places. And really, there aren't a ton out there that are supporting podcasts.
Marty Martin
So please think about supporting Big Biology, make a donation for a friend's birthday and buy them a t-shirt from our Redbubble store.
Art Woods
Think of it as passing up on just one stop to the coffee shop or that subscription to Hulu, you really never use, and direct those dollars our way.
Marty Martin
We love making Big Biology. And we especially love hearing from you how much you enjoy the show, how much you learn, how you also find Art's jokes a bit depressing, and how Keating blows you away each episode with his unbelievable visual renderings of our conversations.
Art Woods
So let's keep it up. Become a patron, make a one time donation, or if you can't afford it just spread the word about us on X, Facebook, TikTok, or Instagram. Help us keep Big Biology going strong.
Marty Martin
Now onto the show.
Art Woods
Today's show deals with two general problems that plague biologists.
Marty Martin
Really, biology-specific versions of problems that plague all of us as we interact and think about the world.
Art Woods
The first thing is that we see what we expect to see and not what's actually there.
Art Woods
Maybe
Marty Martin
Maybe you've seen that famous image of the young and old woman both trapped in the optical illusion that first appeared on an 1888 German postcard. Most people see either one or the other at first.
Art Woods
But if you look more intently, what you see flips back and forth. The young and the old women are both there, but we naturally see one or the other more readily. And it takes real mental effort to see the other form.
Marty Martin
Why do some people see the old form first and others the young? There's some evidence that people are more likely to see the form that corresponds to their own age, and the group of people with whom they socialize most often.
Art Woods
And it's also possible that people just filter information about their worlds in different ways. Recall that 2015 viral sensation that basically broke the internet, the famous blue and black striped dress
Marty Martin
You mean white and gold, right?
Art Woods
Okay, let's not go there. Rather on to the second problem, which is related to but distinct from the first. It's what we might call a naming problem which you may have heard of in relation to, of all things, Buddhism.
Marty Martin
Although Buddhism doesn't outright claim that naming things constrains our view of the world, it does acknowledge that words and language more generally shape our perceptions and can lead to attachment, suffering, and misunderstanding.
Marty Martin
For those
Art Woods
For those more scientifically inclined, a similar objection can be raised to using metaphors in science.
Marty Martin
Think the "Tree of Life," "molecular machines," "genes as blueprints," "information superhighway," and others.
Art Woods
These metaphors provide a beautiful shorthand for conveying concepts and processes, but, conversely, they can become mental handcuffs, if we take them too literally.
Marty Martin
These problems loomed large in our conversation today with the science writer Phil Ball. Phil has written over 25 books and contributes regularly to publications such as the New Scientist, The New York Times, The Guardian, and the Financial Times. He was also an editor at the journal Nature for over 20 years, which gave him an unusually broad look at breaking discoveries in chemistry, physics, and biology.
Art Woods
We focus our conversation on Phil's latest book called "How Life Works," in which he describes recent advances across all levels of biological organization.
Marty Martin
As you might imagine, the book covers highly diverse topics. But there's a subtle kind of meta theme that runs through all of the chapters.
Art Woods
And that is we consistently don't really know what we think we do. Both because we're usually not looking at the process or problem from the right angle, and even if we do we tend to construct metaphors that harden rapidly into dogma, this hardening can prevent us from seeing readily beyond the metaphor' s walls.
Marty Martin
That was a metaphor about metaphors.
Art Woods
I know, but onward. Take, for example, Phil's chapter on how proteins work. We all know how they work, right? Well in the prevailing metaphors of the day proteins are tiny "molecular machines." We say that their chemistry and shapes are evolved to carry out some process often by fitting together like a lock and key with a handful of other molecules.
Art Woods
But
Marty Martin
But Phil argues, that's all wrong. The past 10 years or so have revealed that many proteins do not have fixed, well-defined structures, but they're rather expansive, floppy, and promiscuous, and they interact in large and small ways with hundreds or even thousands of other molecules.
Art Woods
This fundamentally changes how we should think about the functions of proteins and also about the structures of intracellular networks and the standing variation available to natural selection.
Marty Martin
For sure, not what you get out of most standard textbooks in biology.
Art Woods
And Phil is not nearly done with big questions, including the biggest one that we can ask, which is "What is life?" Here, Phil talks about where causation comes from in living systems.
Marty Martin
Most of our metaphors guide us towards a highly reductionist or bottom up view of causation. Phil argues instead that some causation comes from higher levels of organization with causality flowing downward. The basic gist is that as living systems get more complex, the information inherent to their new structure leads the whole system to be more than the sum of its parts.
Art Woods
In other words, here we get into-
Marty Martin
yet another-
Art Woods
extended but interesting discussion about agency, purpose, and goals and biology.
Marty Martin
And finally, a significant chunk of our talk focuses on what our new understanding of biology means for medicine, how we could manipulate complex systems to keep people healthy, and whether there are better ways to treat cancer.
Art Woods
And, as we did when reading Phil's book, you may feel the vertigo that comes with your point of view, suddenly flipping into an expansive new configuration.
Marty Martin
You are such a nerd, sometimes, I'm not even sure what to say.
Art Woods
I'm Ringo Starr
Marty Martin
And I'm a little embarrassed.
Art Woods
Okay, I'm Art woods.
Marty Martin
And I'm Marty Martin
Art Woods
And you're listening to Big Biology.
Marty Martin
So, Phil Ball, thanks so much for joining us on Big Biology. Art and I read your new book "How Life Works," and we both absolutely loved it. I've since shared it with a lot of friends, including a lot of biologists who would probably be as surprised as I was about how much we've learned about how life works at different levels of organization. Things are very different since we learned them as students. We want to spend most of our time on discoveries in the book, but we do want to hear a little bit more about your method, and then maybe a little bit about why some people have been upset and sort of seen it as controversial in a way. But let's start really slowly and let's talk about that method. So you're a chemist and a physicist, right? What was it I mean, but was it your time at Nature, as an editor? You did that for quite a while was it those experiences or what was it that sort of first planted that seed to get you thinking about these things?
Phil Ball
Yeah, well that's right. So my first degree was in chemistry, and then I kind of switched fields, and went into condensed matter physics, so you could say my qualifications for writing about biology are close to zero. As a chemist, you come across some of these ideas, you know, you come across the ideas about, for example, the way proteins work, you know what these molecules look like, the way they interact, but very much from a chemist's point of view. So yeah, it was the fact that I spent years as an editor at nature, the science journal, which, you know, exposes, in those days, this is back in the late 1980s, early 1990s. Nature wasn't this big publishing empire that it is now.It was just a journal. And it was a journal with a very small staff. So it meant that you spoke to everyone, handling every subject, you came across everything. That was why it was such a fantastic, such an exciting place to be. And it meant that there were a lot of very steep learning curves that I had to go up in order to get some kind of handle on a lot of this stuff. Because as an editor, I was handling papers in biochemistry and biophysics, as well as in straight physics, material science, and so on. So I had to learn a lot there. But also, I was daily hearing about the latest things that were happening in biology, trying to make some sense of them and being able to talk to colleagues to, you know, at least get some handle on that. That was what really, not just gave something of an education in biology, but also made me aware of some of the cutting edge issues in biology.
Marty Martin
Gotcha. I think in the book, you write something about the Wnt genes, were those a particular inspiration? And if the answer is yes, we'll talk about what those genes, in what they're involved.
Phil Ball
They seem to me to be a very good example of the way that we understand genes has changed over the past several decades. So the Wnt genes, they're genes that are involved, really, I think all we can say about them is they are genes that have roles in our development, in our development from an embryo to, you know, to a baby, we can be no more specific about it than that, because they crop up in all sorts of places. And if you look into the name, it's a kind of portmanteau name of various things, including, they began as genes that were identified in fruit flies in drosophila, where a mutation of the wnt gene created flies that had no wings. So the “wn” stands for wingless. And you know, so the idea was that the natural assumption at that point would have been, ah well, maybe these are genes, these are fly genes that have something to do with making wings. And in a sense, they do, clearly. But that's absolutely not their function, because we have genes like this as well. And you know, lo and behold, we don't have wings.
Phil Ball
So they do something else as well. And in fact, it turned out our version of the genes were discovered in a completely different context, in some other developmental process. And they were given another name, and then people realized, hang on this gene seems to actually be the same as, or the equivalent the human equivalent, of what's in fruit flies. So what's really going on here? What are these genes really doing? And then they crop up in, you know, lots of other different developmental processes, because they are example and there are other genes like this Sonic Hedgehog is another one.
Marty Martin
That's my favorite.
Phil Ball
And absolutely, how could it not be with that name? And, you know, there again, is another interesting kind of etymology of how it got that name. But, you know, there are these genes that just seem to have many different roles in developmental processes. And, in a way, it's not that they do anything. They're almost like general purpose genes that development uses, if you like, that our cells use in order to make different sorts of body structures. So it's a reflection of the fact that there are lots of genes that we have that you can't assign specific tasks to, you can't draw a line really from the gene in our genome, to some trait some developmental phenotype, you know, some body shape or whatever that that's not the way it works. So there isn't a sort of plan really, that goes from the genome to the phenotype, something else is going on. And these genes, the way I talk about it in the book, is that it's better to see a lot of them as just resources, of a very general sort, that the cells can draw upon in order to do the things they do to make us.
Art Woods
Yeah no that's beautiful. I want to echo what Marty said about just just loving the book. You know, I had this funny feeling, reading it of, you know, in each chapter, thinking, "Okay, I know what this chapter is about," and then suddenly feeling like, "Wow, there's something like really new here that I didn't, I didn't expect," like the chapter on protein functions and how proteins come about and fold and do the jobs that they do. And I was thinking about this in relation also to just the basic teaching of biology, which I feel unsatisfied about. I've been quite unsatisfied with my own teaching of biology. And I think many, many university courses are structured in a way where it's just kind of a slog through a bunch of facts that, that we know, and that had been known for a long time. And I thought, reading your book, that it would be a really inspiring thing for students to have a book like that, and to have the course follow the progression of that book. Because it you know, it just pushes you so rapidly out onto the edge of where are the interesting questions, and what do we, you know, what might we think about next? So, I guess my question is, did you have that in mind when you were writing this? And do you envision it being used for teaching in any way?
Phil Ball
I would be delighted if it was used for teaching. And you know, one of the things that, obviously, that I wanted to do was to try to bring up to date these stories that we tell about the different levels of what's going on in biology, from DNA to RNA, to proteins, and so on. So, Art, I'm so pleased that you say that that was your experience, because it was very deliberate that for each chapter, you know, it works in a way that could sound very dry, and textbook-ey, but I hope it doesn't turn out that way. It works through the different levels of biology, from starting with genes with DNA, going to RNA, going to proteins going to, you know, networks themselves, and so on. But what I wanted to try to do was just show how each of those levels in each of those levels, something significant has changed over the past 10, 20 years or so, in the way we think about what's going on. So, you know, the the standard thing that, certainly that school kids are taught, and I think undergrads are taught is that, you know, we have these genes, they encode proteins, the proteins are the enzymes that kind of somehow put us together, and this is how it works. And that story isn't wrong, but it's far more complex than that, and it's not so linear as that. And also, I mean, as you mentioned, proteins, this was a real revelation to me. And this was something that was brought home during that time I spent in 2019 at the Harvard Medical School, that I had, as a chemist, I had this idea that just seemed very neat and intuitive that we know the cells are full of molecules, you know, there are thousands of different proteins in every cell, how on earth do they ever, you know, find the right targets and do the right thing without getting distracted by all this other stuff. And the idea is that there's this molecular recognition whereby each protein has a particular shape. That means it fits the target molecule that it's meant to fit and ignores all the others because they're the wrong shape.
Art Woods
That's the iconic vision that we all have, right?
Phil Ball
Absolutely. Yeah. And that too, you know, isn't wrong. There are proteins that work that way. Lots of enzymes do work that way, but lots of them don't. And I had, you know, I really had no conception of this until a few years ago, and it was a revelation, that actually there are plenty of proteins, particularly in our cells, or in the cells of large creatures like us, there are many more proteins of this sort in large creatures than in smaller ones, and certainly than in bacteria, that don't have a fixed structure, don't have a fixed shape. They're floppy, they're what biochemists call intrinsically disordered. And it completely blows apart this picture, because it means because they're floppy, they're slightly sort of indiscriminately sticky to things. They're promiscuous as it's put, so they will bind to lots of different things. So how on earth does that work, that you have these proteins that you know, they're not selective in this way, and not only that, it's not as though this is just something that biology has been forced to do because it can't make proteins, you know, specific enough and structured enough. We know that it can because plenty are like that, and in fact, most of the proteins in bacteria are like that. So during the course of evolution, as organisms got more complex, it seems that the proteins, many of the proteins have got less structured and more promiscuous. Why is that? How does that work when the organism is more complex? And that's one of the central, I think, one of the central themes throughout the book because it points to a different kind of set of operational principles at the molecular level, that involve this promiscuity of molecular interactions. And what tends to happen is rather than the sort of digital logic of this speaking to that we have groups of things that all kind of speak to each other in some way, and they get together in these groups in various ways.
Art Woods
Right, right. It's much more smeared out logic.
Phil Ball
Absolutely. And so I talk about it as kind of committees of molecules. And, you know, there are all sorts of processes, crucial processes in ourselves, particularly how genes get switched on and off, that rely on these committees, where the molecules are all kind of speaking to each other in fairly indiscriminate ways, more or less discriminate, and somehow out of that process, just as out of often to the committee's that we make, a decision arises, a definite decision arises: turn this gene off, turn it on, do this do that. But it's not anything that you can attribute to a nice clean digital logic of one molecule speaking to another.
Marty Martin
Yeah, yeah. Well, I wanted to sort of hear about, you know, the title of your book. We're talking about proteins, and I could sort of infer how that works. But your book is beyond what one might initially assume to come from a book called "How Life Works," as you just represented the novel ways that we've come to understand protein function. You alluded briefly to the novel ways that genes operate, relative to the classic ways or the ways that maybe Art teaches them or taught them in his biology classes. But when you titled the book "How Life Works," it felt like you meant something broader. Maybe let's start with your definition of life, because life also has the evolutionary dimensions. I mean, one of the things that distinguishes that life evolves. So how do you define life? You quoted Francois Jacob, saying that living organisms are the site of a triple flow, matter, energy and information. Those are abstract things, but you've return to them over and over again. I mean, when you when titled the book, "How Life Works," how did it sort of include evolution? And how do you think all of this sort of comes together to challenge the way we think about life? Generally? Yeah. Well,
Phil Ball
This is the bigger and difficult question.
Marty Martin
I know that's a big one.
Phil Ball
And I mean, first of all, I have to say, giving the book this title was an act of incredible hubris. And I hope it's kind of being seen as slightly tongue in cheek, because you know, what a thing to claim, you're going to say in the book, "this is how life works."
Marty Martin
Sit back and hold on.
Phil Ball
And actually, I hope it becomes clear that my aim is to say, we don't understand how life works. You know there's so much we don't understand about it, but I think we can safely say now that it doesn't work the way we thought it did, or at least our life. Because a lot of what I'm talking about, as I've sort of, you know, implied already, a lot of what I'm talking about applies to us as large, complex organisms. So, you know, and one of the key ideas in the book, and this again, was a revelation to me, is that the way we work doesn't seem to often have the same basic principles as the way bacteria work. You know, often we've studied bacteria, understandably, because they're much simpler. They're single celled organisms, you know, you can sort of see into them much more plainly, so we've studied them as a first step, and then often assumed. And you mentioned, Francois Jacob, you know Jacob and Monod, were the ones who first talked about gene regulation in the 1960s. And the assumption was long made that the way they explained it is probably, you know, similar in humans. And it turns out that it's not, that actually the molecular principles of gene regulation are largely different in us. So that was one of the key messages, that actually we can't generalize often from bacteria to us. There are some basic things that are similar, of course, we all have DNA, we all have RNA, we all have proteins, you know, Francis Crick's central dogma, is kind of basically still the same all along. But the real principles by which our cells work aren't the same as those in bacteria.
Phil Ball
So that was one of the key messages: how life works is not a single question, and how our life works is not the same as how bacterial life worked in many key respects. But, you know, the question of, well, what is life anyway, you know, what are we talking about here? That's actually something that I sidestep in the book. I say upfront, that people like, you know, from J.B.S Haldane to Erwin Schrodinger, and there's a lovely recent book by Paul Nurse as well and all are called "What is Life?" and none of them answer it. Haldane's is the wisest, because he begins it the very first sentence in that book or that essay is, "I am not going to answer this question." You know, that's right, because there is no consensus about that. There's a lovely book by Carl Zimmer called "Life's Edge," which explores that question. And Carl rightly points out, you know, we don't know, there is no agreement, basically. So the approach I took is, well, I don't have to answer that question, in a sense, because I'm talking about, we know that we're alive, we're a form of life, how do we work is really what I'm talking about. But, inevitably, in trying to answer that question, you know, I do have to come back to the question of well what is it that is going on it. Which really is about, you know, what is going on in life. So you do have to take a step back, because all of these complexities are so you know, they're so complex, it's so extraordinary, that you have these many different levels, from genes to proteins to networks, to cells to tissues, that somehow, even though they have different operational principles themselves, somehow, they're all operating together in ways that support each other, to build something like us. So what's keeping this show on the road? What is it overall that life is doing? And that led to some questions that really go, you know, they certainly go beyond the molecular specifics. And you get into questions of well what is it that life is, if you like, trying to do. And we get to this question of agency, so I have a chapter towards the end of the book on agency. And really, I think the key point I wanted to make there is it's a real question. It's one that biology has ducked, it's one that biology still seems to be uncomfortable about, because you have to start talking about things like function, and even purpose, which are, you know, taboo words, or have often been seen as taboo words in biology. But you really have to think about those, if we're going to grasp agency, and I think what biology needs is a proper theory of agency. And there is nothing of that sort at the moment.
Art Woods
Yeah super beautiful and we couldn't agree more, I want to just continue on this theme. So you know, if we think about agency and causation at different levels, what I hear you saying is that as you move from bacteria to more complex eukaryotes and multicellular organisms, you get a shift in the sort of loci of causation and the loci of agency. So, for example, at a bacterial level, you might think of the lac operon and E. coli as a sort of a general circuit, right? Tha is, in a very real sense is making decisions about what sorts of sugars are available, and what sorts of ways it should retool its biochemical machinery to use that sort of sugar. That's a very genetic kind of based locus of causation. And I think what you're saying is that as you get to more complex animals like us, those loci occur at other levels. So maybe let's just be more specific about that. What levels do they occur at? And what processes are you talking about?
Phil Ball
Yeah, well, this was another kind of penny dropping when I was, you know, struggling to figure out what the structure, what the general theme of this book was. The issue of causation, I think, is an absolutely crucial one. And there's this idea that has been talked about for a long time, but often in very vague ways, the notion of emergence and emergent systems. And it's often illustrated with reference to things like flocks of starlings or even road traffic, where you get structures and patterns that can't be deduced by looking at the individual components of the system, you can only see them as they emerge at some higher level. And this is something that actually it's familiar in physics, it's been familiar for a long time. In a sense, what we call phase transitions are examples of emergence that there is nothing in a water molecule that will tell you the water has three different phases, gas, liquid and solid, and that there are sharp switches between them. These are cooperative things that happen between many water molecules, phase transitions are emergent phenomena. So it's a real thing that happens, certainly in that part of nature, but we see it also in organic nature, in life. But it's very hard to sort of pin down.
Phil Ball
But one of the aspects of that notion of emergence is what some people talk about as "causal emergence." And here again, it's become possible to quantify this and to get theories to understand this, that there are some complex systems and we seem to be one of them, and aspects of things like the brain are also one of them. Where it turns out that there are ways you can measure the amount of causation that's happening in a complex system at different levels at different levels of magnification, if you like. And it turns out that there are some systems where the causes of the overall behavior of the system are located much more in the higher levels of the system, the bigger scale things than in the smaller scale things. And this is the idea, of course, of emergence. And this seems to be I mean, it's been measured in living things, and it seems that there absolutely is a degree of causal emergence in living things. And once again, there's a difference between simple living things like bacteria, and complex organisms, there is more as you would probably intuitively expect, but you can quantify there is more causal emergence in complex creatures like us, which means the causes of our behavior are located at higher levels of organization than, for example, the genome. So there are many things that happen where you cannot say, you cannot point to a gene, or even two or three genes and say, "These are the causes." And this is what we've known it certainly in genomics for a long time that it's the notion, for example, that if you try to identify which genes or which gene variants are correlated with particular traits, like height, or like intelligence, or like obesity, for example, if we're thinking about, you know, diseases-
Art Woods
Right, you find out that it's most of the genome, right?
Phil Ball
Absolutely, yeah, so, you know, it's all over the genome, what does that mean? Well, you know, one way of thinking about it is, well, people say that all the genes are somehow causes. But that doesn't really tell you anything, that doesn't sort of mean anything. I think all that's really telling us is that the true causes for those kinds of traits happen at a higher level. And what you can see, and the genes are still affecting them influencing them in some way, but all we're seeing in those genomic correlations are the echoes really, of causation happening at higher levels. And it certainly means that if you want to make interventions, as we want to in medicine, that there are plenty of traits that have some genetic component, in fact, just about every trait we have has some genetic component, but that cannot be, fundamentally cannot be addressed or intervened in at the genetic level, not just because there are too many genes involved. But because that's not where the causation lies.
Phil Ball
You know, we see this a lot, for example, with COVID, that actually, you know, the problematic cases of COVID, what seemed to be happening to cause problems wasn't something genetic, even though some people seem to have more of a genetic propensity to have a bad reaction to COVID. In that case, the level of the immune system. And in fact, you know, it turns out and this is something that has, you know, been recognized for a long time in biomedicine, the immune system is such a locus of causation for so much that happens to us, even things that don't seem obviously connected to inflammation and the immune response. It's a massive locus of causation. And so if we want to intervene in those sorts of conditions, then we shouldn't be looking to find a genetic target or a protein target for a drug to hit, we've got to find a way of intervening that is going to alter the way the immune system does stuff. So we need to intervene at a higher level of causation. So that's what I mean by causal emergent, and also why I think it's an important concept for biology and biomedicine.
Marty Martin
I think we will circle back to the biomedical implications of your book in a bit. Being a bit of an immunologist, I'm especially intrigued to talk about these kinds of things. But before we leave and to touch on agency, as Art brought up, what do you think about the selfish gene concept? Because I want to talk about your book in the context of evolutionary theory or, you know, modern understanding of how life works, what it means for how we think evolutionarily. How does the selfish gene concept fit?
Phil Ball
Right, I'd always had this struggle with the notion of selfish genes, because, you know, it's clear, I think, to everyone in biology that our genes have to operate cooperatively in order to be a part of, I'm not going to say to build because then I don't think that's the right way to look at it, but to be a part of, you know, an organism that enables evolution to happen at all. So to be viable, they have to cooperate, they have to do all their respective jobs at the right place and time, you know, in relation to one another. And in fact, the whole selfish gene idea isn't about what the genes are doing. It's about what different variants of particular genes are doing, what different alleles are doing. So, there is a meaningful sense, if you think about population biology, there is a meaningful sense in which we can think of different alleles of a given gene as being in competition with one another, you know, and if you like being selfish, because you know, it is a bit of a zero sum game that you know, one of the those alleles is trying, if you like, to put it anthropomorphically, it's trying to outdo the others. So I totally understand, you know, why that picture makes sense in terms of evolutionary population biology.
Phil Ball
It doesn't make sense. It is incoherent from the point of view of thinking about genetics from a developmental perspective and developmental biologists have felt this way, forever, really. You know, certainly back to Conrad Waddington, this has been a problem that they have. So, one thing that I talked about in the book is that there is, at least this conceptual difference, between the notion of a gene in evolutionary biology and in developmental biology, and it's always been hard to reconcile those two. And in fact, perhaps they can't be reconciled, because we're talking about genes in different contexts that way. But what I wanted to try to do here, I kind of, it was tempting, and it would have been easy in a way to have simply attacked the notion of a selfish gene from this point of view of you know, if you think about it developmentally, it makes no sense. But I thought, well, it clearly is a useful idea to evolutionary geneticists, and they use it all the time. So what is it really expressing? And it seems to me that it's not a metaphor for what genes are, or for what genes do, it's a metaphor for the models that are used in evolutionary biology to understand that particular aspect. And those models, don't I mean, you know, I have spoken to evolutionary biologists who said there, so two models don't really recognize organisms as such at all. And, you know, Richard Dawkins talks about the organism as simply a vehicle for the genes. That notion, it seems to me, it's almost antithetical to you know, what Darwin was trying to do to understand organisms in their own right, and to sort of respect organisms in their own right. But it is a viable enough picture, if you're thinking about populations of different alleles, and how those, you know, change through evolutionary time. So really, the selfish gene notion, it's a metaphor for the models that evolutionary genetics uses to understand how populations of different alleles change over time. So that's fine. But it's really important, I think, to recognize that that's what it's doing. It's not a metaphor for what genes are, or for what genes do. It's a metaphor for that particular model of evolutionary biology.
Art Woods
Yeah, I'd never thought of it that way. Yeah, that's really interesting. Another word that we've already said multiple times here is agency. And I think it's worth maybe being also a little more explicit about what you mean by this word agency, which we hear increasingly in the scientific literature, it feels like sort of an interesting moment for this as an idea. But to you, what is that? Why is it important?
Phil Ball
Well, I think I'd give a definition of agency, as opposed to life. I'd say something along the lines, that an agent is an entity that is able to act on itself and on its environment in order to achieve self-directed or self determined goals. So an agent is some entity that has goals, for a start. And they are goals that, I mean, there are various ways that they could be set, you know, I think goals for organisms are set by evolution. So you know, evolution is a process that produces goal-oriented things. Evolution itself, the process of evolution, as far as we know, has no goal, there is no obvious reason why we should think of evolution as having a target or a direction or anything. Some people have argued, but you know, there are ways in which that might be the case, and that's an interesting debate, but we don't particularly need to get into that to understand agents. It's simply that evolution creates goal oriented entities. And that, to me, seems uncontroversial. You know, we have no problem with the fact that we are goal oriented beings. And the goals that we have, you know, often we kind of hear that, if we admit that living things have goals, they're survival and reproduction. And you know, often that's the case, that's, you know, that gets you a long way, but there's no way that we can explain everything that we do in those terms, other than with some kind of evolutionary psychological, you know, just-so story, which sometimes happens.
Art Woods
Yeah I mean, it feels like, you know, I mean, one way to get around that is to say, you know, there's these prime goals, which are survival and reproduction, and then there's a bunch of subsidiary goals that contribute to that. And you know, life is an elaboration of a lot of those subsidiary goals. I wanted to ask, I mean, like, to me, the idea of agency becomes really interesting when we think about agency occurring at different levels simultaneously. So for example, large multicellular animals, like myself, have goals and purposes, I'm an agent. But each of my cells also you could think of as being an agent, right? So we have sort of nested levels of agency. Right? And I guess that broadens out this idea of agency in an interesting way? And is there a sort of general way to think about agency across all of those levels simultaneously?
Phil Ball
I think the simple answer to that is, if there is an answer to that, I don't know it. But I think that that is a really important and interesting question, you know, and one that informs all sorts of other questions. So how is it that if we recognize that every, and I would argue that every living cell is an agent, in some sense, even bacteria have a kind of agency. How is it that multicellular creatures are viable at all, you know, rather than just being sort of torn apart by the agency of all their parts? I mean, you could say this is Richard Dawkins has talked about the paradox of the organism, but he's talked about that from a gene centric view, which I find harder to understand. But certainly, if we think about it, from the point of view of cells being agents, then there is a sort of paradox of you know how that works. But it is one that I think, you know, we can think about and start to understand. But, you know, we are collective agents, you know, and our agency is collective, and a big part of our agency, of course, comes from our complex cognition.
Phil Ball
And, you know, one thing that I've argued, in fact, I argued it in my previous book, and "The Book of Minds," is that, you know, I would say, I mean, Ary you talked about us having, you know, the sort of overarching goals of reproduction and survival, and then we'd sort of do other things in relation to that. I would argue that we have minds, because we are complex creatures, there is no way that we can predict what we're going to face tomorrow, or in the next moment. Our behavior, and our environments are just too complex to have any sort of hardwired solution to everything that we might face. Bacteria are much simpler and they can have a more automated response, although even they can't have that entirely. But I think brains are evolution's solution to that, that instead of giving us some, you know, hardwired response to every situation, or even instead of giving us some computer, because I don't think the brain meaningfully is a computer that kind of somehow computes what we have to do next, it wouldn't have the time to do that. It's an organ that uses rules of thumb, really, you know, to do something that we have found through trial and error, through experience, and through some degree of innateness to be usually good enough, to be good enough to allow us to get by. And what that ends up doing is giving us more, in a sense, more cognitive capacity than we would seem to really have any need for. You know, what is the evolutionary function of being able to devise the theory of general relativity? Why on earth do we do that? Well, it's because in order to have these brains that can function well enough in our environment, you've got to give it so much capacity that it starts doing these things that seem to have no evolutionary benefit. And in fact, some things that seem to be counter-adaptive, you know, non-adaptive. And that's just what minds are like. But I think, you know, I argue in the book that actually, when we start to think about what agency really is, and what agency requires, it requires things like memory, it requires things like having some sort of internal representation of your environment so that you can anticipate what you might find. When we start to think about it that way, we have to start thinking of every living thing down to single-celled organisms as, in a sense, cognitive agents, not just stimulus response machines that all life is cognitive.
Marty Martin
Yeah, agreed. And, you know, that sort of foreshadows our questions to you about Mike Levin's work, but let's hold off on those for just a second and stick in this space of, you know, why agency? I mean, I think you've articulated some about the utility of agency. But where does it come from? So you know, Art raised the issue of conflicts across levels. But where did it start? How did it get going? Is it heritable? Does it even make sense to ask that question? Is it sort of variable among individuals in a population? Is there a way to merge these concepts or these sort of frameworks really?
Phil Ball
Yeah and these are absolutely the questions that should motivate having a theory of agency so that we can get some handle on them. I mean, where did it come from? I think that actually it correlates completely with living things that actually I don't think anything can be truly alive, if it doesn't have agency. I don't think necessarily the converse is true. I think that we may end up making genuine agential machines. AI, basically, might already have some degree of agency, it might already be setting its own goals to some degree. So you know, I don't think that only living things can be agents, although at the moment, I suspect only living things are genuine agents. but that may change in the future. But I do think that no living thing cannot be an agent, that everything in life has to be an agent. So, you know what that means is that in order, I mean, the origin of life is complex enough already, okay, without introducing agency, but I would hope that actually by introducing agency, we can focus that question a bit more, because, you know, so often discussions about the origin of life, they might start off listing various things that living things have to be able to do, they have to be able to reproduce, they have to be able to metabolize. It's just a kind of shopping list that, you know, no one agrees on, and that might seem more or less arbitrary. But I think that to have a genuine living thing, you have to have created, in some way, agency. And so if we understand what is needed for agency and, you know, as I say, I've mentioned some of them. I think it does need some sort of memory, there are physical theories that suggest that only by having memory, and by using that to create some sort of representation of the environment, only systems like that can be thermodynamically efficient. Otherwise, I mean, you know, crudely speaking, and even literally speaking, otherwise, you're just bumping into stuff all the time, because you're not anticipating what's coming, and that's very inefficient. So we can start by thinking about what agency is and what it requires, breaking it down, we can start to formulate some questions about what would have been required for life to begin. that are outside the normal ones of, you know, how do you make replicating molecules? How do you make DNA and so forth?
Marty Martin
Yep. Do you think that agency is something that we can directly measure? Or is it one of these more abstract kinds of things like fitness, that we know is really important, but takes lots and lots of different forms, and is context dependent? And, I mean, will we ever have an agent-ometer?
Phil Ball
Ha yeah, great question. And I can't answer that with confidence, I do think that it will be possible to anatomize agency to say, well, this is you know, as I've sort of hinted that agents have to have these particular characteristics. I don't expect any convergence on exactly what those need to be. And I think it's quite reasonable to suppose that things will have degrees of agency, in much the same way as we imagine living things have degrees of sentience. But you know, I think there has to be some agency, as I say, for it to be alive, you know, but there are different degrees of agency, but probably also different types of agency. In much the same way as I think we need to stop thinking of consciousness as a single thing, like a single, you know, fluid that we have more or less of as different organisms. I think there are different dimensions to consciousness, and I think there will be to agency, so that some agents will be very good at doing particular things they might have, you know, particularly good memories, for example, some agents might be particularly good at actuating things, being able to move, being able to sense in certain ways. So, you know, I think we can break it down and atomize it, but whether we could ever really sort of quantify that, I'm agnostic about at this point. This is why we need a theory to help us.
Marty Martin
Yeah, I mean, and that's understandable. But it's also great to hear caution, you know, we don't want to go too far, because these ideas had been anathema for a long time and just starting to come around as we attempt to formalize them. You've used the word a couple of times, which sort of conflicts with something we were talking about a minute ago, or you I guess you've implied that agency may be decomposable? Is that what you mean? I mean, if it is sort of, what is it a substrate of emergent causality, to go back to what we were talking about before ? I mean it raises the bar for measurement, doesn't it? How do we put these two things together?
Phil Ball
Yeah, well, it does, except that I would hope that, if we can decompose it, and I feel we probably can, then perhaps we can quantify it a bit better, because then we can start to, you know, think about the individual components. And again, you know, I'm using memory because it's a familiar one. But, you know, we can kind of measure the amount of memory that something has. And I should say, in fact, that have better discussion than I'm going to be able to give of what, of how to anatomize agency, and you know, how to think about agents.A better discussion is you can find it in a book called Free Agents by the Irish neuroscientist Kevin Mitchell, which I'd really recommend, which I, you know, drew on for quite a bit for my discussions of agency.
Marty Martin
We totally agree. We just had Kevin on about a month ago,
Phil Ball
Fantastic.
Marty Martin
It was a great book, very, very good book.
Phil Ball
Yeah yeah. So I think we probably can anatomize it. One thing, though, I would, I mean, this is perhaps going to sound more controversial, is that allied to the notion of agency and to the recognition that agents have purpose, I think we need to be able to start talking meaning in living systems and that will send a lot of people scuttling for sure. But here's what I mean by that, cause at the start of the book, I say that I think one way to think about living things is that they are creators of meaning. And what I mean by that is that, you know, any living thing is going to inhabit this environment that is full of stuff happening, full of what we might think of as information. We know that that's the case, we have all this sensory input coming towards us. What a living thing has to be able to do is to filter that information to figure out what matters for its survival, you know, first and foremost, and what doesn't. And different organisms have made different decisions about that. So that you know, for us, vision is really important, because that's in the nature of our environment. You know, for dogs, smell is a really important thing. For lots of creatures in the sea, particularly the deep ocean, vision isn't going to get them very far, because it's just too dark down there, so they have other senses. But they have different ways of filtering that information, but not just filtering it, but giving it some kind of valence, you know, how much attention do I have to pay to this information? How worried, if you like, to put it anthropomorphically, how worried should I be that I've just had this signal or that signal? How interested should I be in it? So I think this is something that living systems do. They don't necessarily, you know, there doesn't have to be any awareness, any conscious awareness of that happening, but they have to have some system for creating a sort of valence to their response. And that's what I mean by meaning. And, you know, they have to be able to do that in order to not be overwhelmed, and in order to not collect a load of information and store it, that isn't going to be any use to them, because, again, that's inefficient. So in that very specific sense, I think living things do create meaning from what they experience.
Art Woods
Mhhm. Great. I have a question from the skeptics corner. And this comes from talking with our other co host, Cam Ghalambor. I sort of wish he was here to ask this question himself, but he's asked me and Marty several times some hard questions about what we mean by agency. And his take on it is that it's not clear necessarily what the difference is between agency and just plain old phenotypic plasticity, which he's spent a lot of time thinking about and a lot of time studying. And I think you could argue that, you know, what are the systems that instantiate this idea of phenotypic plasticity? They're complex systems that are sensing the world, and making decisions about how to alter phenotypes across different spatial and temporal scales, from very rapidly to very long- term developmental kinds of changes. And, you know, it's hard for me to come back to Cam and say, "Well, I think agency is something else than that." So would you say something different about that?
Phil Ball
Yeah, that is a very good question. I guess it's not clear to me where, if we're simply thinking in terms of phenotypic plasticity, where in that picture, we're going to get notions of purpose and goals. It may be in there, and maybe that I that I'm not familiar enough with that way of thinking, to know that, but I feel like that's the key thing that we have to recognize with agents, we have to think about them as having some sort of goals that are internally set. And, you know, one reflection of that is even for, you know, certainly some sorts of bacteria, and I suspect for all sorts of bacteria, we can give what looks to be an identical population of cells an identical stimulus, and they won't all respond in the same way. Because, I mean, there'll be some stochasticity in there, some of that will be random, but some of that will be purposeful in the sense that because they've had different histories, they will have different internal settings that will condition the response that they give. And I think it's meaningful to talk about that is those bacteria having different goals, which, you know, rationalizes the different responses that they will have. So it seems to me that that's something that goes beyond simply recognizing phenotypic plasticity, in the sense of recognizing well there can be these differences. Maybe it's really about asking, well, you know, why is that, what are the driving, you know, causes of those different responses. And that has to come down to some notion of the system making a decision, really, making decisions that are related to you know, you only make decisions if you have goals.
Art Woods
So we touched on the consequences of agency and this sort of systems thinking for how we should think about problems and issues in medicine. I think it's a good point just to return to that. And the point you were making at the very beginning of our conversation was that, you know, it's hard often to trace particular medical situations to particular loci. And that's because these loci of causation or loci of agency are spread out across the body. Maybe let's talk about that, in the context of cancer, you have some quite interesting things that you say about cancer. And, you know, if we just sort of step back and say, "Well, what is cancer?" It's effectively cells that are going rogue, in the sense that they're giving up on this common project of running and operating a multicellular body. And they strike off on their own and they become selfish, it becomes about them. And I think one way to conceive of that, is that something about their agential decisions, their purposes, have changed. So what is it about writing this book that informs, you know, are there sort of new horizons for how we should think about the origins and possible treatments of cancer?
Phil Ball
Yeah, I think there absolutely are. And I think that one of the things that seems to me to have happened again, over the past maybe 10, 20 years, is that that picture that you just presented about the way to think about cancer, that that I think, too, has changed. So this notion that the cancer cell is a rogue cell that has become selfish, and it's doing its own thing. Clearly, cancer cells are, in a sense, if you like, they're not cooperating with the rest of the body. But they aren't totally individualists. It's really, really striking that now that we have the technologies to be able to, on a single cell basis, look at what is really going on inside a tumor. We find that actually, rather than just being this undifferentiated mass of proliferating cells, each of them an individualist, it's more like, and I use this phrase, I think, I heard this phrase from Brad Bernstein, originally, that it's more like a deranged recapitulation of development, that actually tumors have some characteristics of organs, that there are different types of cell within them. And they seem to be operating more like a kind of crazy organ than just a, you know, a mass of individuals. And even to the extent that we see tumors are co-opting healthy cells and including healthy cells within the tumor, to do things for that tumor. And doing things that are, they sort of do normally, you know, I mean developing, you know, blood supplies, for example, you know, vascularization of the tumor is really important. And that's one of the things that is often now attacked to try to attack the tumor. So I think the view of cancer, you know, has changed in that regard, that it's perhaps better to see it as a an aberrant form of development, rather than, as you know, on a single cell basis.
Phil Ball
But I also kind of, I think it's more useful to see it not as so much as cells that have totally gone rogue, but that cells sort of acquired one of the many different possible states that our cells can form, that it's I think about it more in terms of Washington's famous, you know, landscape of possibilities. That is absolutely the way that cells are now being thought of in terms of how they find their fate, it's on a landscape. Which, you know, at face value seems to be a landscape of enormous complexity, of multi-dimensional complexity. If you think of how many, you know, genes we have, and how many different types of regulatory system we have, the possibilities for cells seem endless. And yet, there seem to be only a few different states that they form, relatively few. And they seem to find them quite, sort of, reliably. And some of those states will be ones corresponding to cancer-like states, and cells can flip over into those. And you know, this is being thought of and analyzed now in a sort of dynamical systems framework, where we think of, you know, what are the different collective states that the gene network can form, and one of them or some of them, correspond to these dates that we think of as cancer cells, and thinking about it that way, I mean for a start, I think it creates more realism about cancer because I think cancer is just a consequence of being multicellular organisms. You know, every organism has, some organisms like whales have better defenses against it, so they have less cancer, but it's something that all multicellular organisms are going to be vulnerable to. Just as misfolded protein states that give rise to things like you know, neurodegenerative diseases are a consequence of having proteins, this is just something that proteins will do. And this means that we need to think about cancer treatment less as something to be eradicated, which is just not possible. But instead, if we think about it as wanting to guide our cells to stay on track, to stay in one of these healthy valleys, and not to go over into one of the others, and even to the extent that there are now these approaches to treating cancer, called differentiation therapy, where what you're trying to do is instead of just blasting them and trying to kill off as many as you can, without killing all the healthy stuff as well, it's more along the lines of thinking about it in terms of cell reprogramming, which we're doing for other purposes, but reprogramming cancer cells to bring them back on track to bring them back into a healthy state. Or perhaps programming them in a way where they get stuck in a dead end where they can't proliferate, and where they're kind of sitting ducks for chemotherapy. So, you know, hopefully, this way of thinking about cancer as a sort of dynamical systems sort of picture, that this might lead to other approaches to attacking it.
Marty Martin
Wow. And, you know, one thing that this conversation has now made me think two things. First, do you think that tumors may be used as sort of these quasi agents to understand the compositionality of agency? Because if there's sort of organs of a sort, they're not quite, you know they're some composite of things that haven't existed before, but they do have agency, in a way. I mean, maybe tumors could be a useful model of understanding agency?
Phil Ball
Well, I think that's a really interesting way of thinking about it. I certainly think that thinking about tumors from an agential point of view makes a lot of sense. Because, you know, then we need to ask what manner of agents are they? What are their goals, and as I say, you know, their goals now seem no longer to be, well, I'm just going to go and proliferate as much as possible. It's much more that they have collective agential goals, you know, to develop in ways that are mutually supporting, within a tumor, and that make use of some of the agency that healthy cells have.
Marty Martin
Well, and I'm also sort of motivated to think about that, because I know you wrote a lot about Mike Levin's work with frog embryos. Mike's done amazing work we've had on the show many times, but one of the things that you emphasize are his xenobots. So I wonder, have you talked with Mike about sort of combinations of frog embryo cells to produce a variety of seeming tumors, and sort of understanding their behavior in a search to understand cancer? Is that something that he's progressing?
Phil Ball
It absolutely is, I've talked a lot to Mike over the years, you know, for this book, and for other things. In fact, when I was in Boston, when I was at Harvard, I visited Mike in the lab, and that was when I first woke up to just I mean, you know, every time I speak to Mike, it seems like he has 10 new ideas that he's thinking about. And amongst those, he's absolutely now, and you know, he's gone public about this already, so I can say. He's in the process of forming a startup company to make use of the work that he's done on these xenobots systems to find applications for them. And you know, biomedicine is absolutely going to be one that they're going to be looking at. It's very early days, whether it's going to pan out or not, I don't think Mike knows, let alone anyone else. But, you know, it absolutely seems to be a sensible way to proceed. Yeah, so his work on xenobots, was another sort of, I guess, you know, trigger for getting me to really think about what the broader concepts of this book are, because one of the things that it seemed to jump out from the idea of thinking about cells as these complex systems that have, you know, a landscape of different possibilities. Once you start thinking about that, it seems, I think, pretty inevitable, that it's unlikely that evolution has explored all the possible states that our cells have, I think it's very unlikely, you know.
Phil Ball
So what else is out there? What tissue types might cells be able to form that we haven't seen? Because, you know, evolution just hasn't got there, hasn't had a need for them. It seems utterly incredible to think that what evolution has found is exhaustive of all the possibilities ourselves can do any more than evolution will have exhausted all the possible proteins that exist or, you know, all the possible gene sequences. So what else is out there? And this is it, you know, absolutely what Mike is doing. You know, are there other states have in his case, you know, he started with frog cells, other states that these cells can have that can be liberated that we can find our way to? And he found, you know, his way to these ones that he calls xenobots, which are, you know, collections of cells. So they're absolutely things that are formed. They're not just individual cells doing particular things, they're things that are formed that seem to be collective agents of some sort. And that have, as far as we can see, they don't correspond to anything that we find in nature. I mean, we've seen, you know, there are observations going way back beyond Mike's work for structures of this sort, that, that frog cells in particular seem to be able to form. But like, by by looking at those, rather than seeing those, as, you know, bits of tissue that have just kind of come loose or whatever, looking at them as potential agents and seeing what they do, if left to their own devices, you know, it revealed all this kind of behavior that had, presumably, been overlooked before. So that's absolutely the kind of thing that Mike's experiments, you know, got me thinking about and. of course, he was there way before me in saying, "What else can cells make that they haven't made already? "And, you know, that leads us into whole areas of tissue engineering, and, you know, straight biotechnologies, but multicellular biotechnologies, what else can we build with cells?
Marty Martin
Yeah. And I guess I should mention that why we call them agential, why Mike refers to them as agential is that they're not just conglomerations of cells that sit in a dish. I mean, these are new entities, entities that never existed in the evolutionary past, as the conglomerations they are. They're from frogs, but these are different kinds of cells. You put them in mazes with resource gradients, and they can use the cilia to sort of follow these resource gradients. I mean, that's what makes it agentia. They're not dying, as long as you're giving the resources. They're behaving, they're doing
Art Woods
And to me, I think, you know, one of the mind blowing takeaways here is that, you know, agency emerges very naturally, when you put cells together into agglomerations, right? It doesn't take some magic evolutionary wand that operates over a long period of time, but rather, there's almost this capacity of cells to form higher level agents just directly when they come into contact with one another. I think that's kind of amazing.
Phil Ball
It is, because it's not at all obvious that they should do that. You know, even if we admit that individual cells are agents, how do they collectively, you know, decide on some sort of group objective, if you like. You know, I should say that there are people who say, of Mike's work, "Well, what we're seeing in this behavior, it's not actually purposeful, it's not goal directed, it's not really agential, it's just, you know, the playing out of the things that the cells do, you know, the waving about of cilia, which happened to create this sort of motion, that isn't a motion, you know, with any sort of goal or purpose." So there are, you know, there are people who sort of push back on that idea, but I think, it seems to me, hard to understand some of the things that Mike has seen xenobots do unless we bring an agential picture to it.
Art Woods
Mhmm. Yeah, agreed great. Well, we're getting pretty close to the end of our time. But before we go, I wanted to ask you about some really interesting thing that you had just at the very end of your book, you had this sort of extended box at the end of the last chapter. And in it, you start to explore some of the consequences, or even just the question of are there consequences for how we understand evolution of agency. And this is quite interesting, of course, in light of this ongoing controversy over the past couple of decades about what should the scope of modern evolutionary theory be? How much of an overhaul does it need? How inclusive can it be of new ideas? I guess I just want to put the question to you directly, like after having written this book and thought a lot about agency, do you think it changes in any fundamental way how we should understand the evolutionary process?
Phil Ball
Yeah, all I would say on that is it might do I mean, I rather ducked this question, Marty from you earlier on, you know, what does it mean for evolution, and I do decayed in the book, and to some extent, as well, because as you say, I leave it all to this box at the end. And there are two reasons for that. One is that I think I know enough about evolutionary theory to know how little I know about it, and to know how vast my ignorance of it is, and also to know how subtle it is. So I don't feel I am well informed enough about evolutionary theory to be confident of saying anything about that. But the other reason, perhaps a better reason, for me being circumspect about that is I don't think anyone knows well enough, at the moment, what these new perspectives on biology are going to imply for evolutionary theory. I think it's absolutely conceivable that when we start to think of organisms as agents that have their own purposes and goals, including ones that enable them to change themselves and to change their own makeup in some way, the question of is that going to impact standard neo-Darwinian evolutionary theory that was predicated previously on seeing what organisms just as if you'd like passive vehicles for genes. You know, they're no longer passive. They're actually top down control agents. Is that going to change the theory? It seems an entirely reasonable, in fact, an unavoidable question to ask. But I don't think we know the answer to that at the moment. And in fact, you know, you mentioned that there are some people who are going to push back on some of the things that this book says, I absolutely anticipated that, and I'm sure I'm going to see more of it. But I think of myself as quite conservative in these respects, because I'm totally agnostic about whether we need something like this new evolutionary, extended evolutionary synthesis, or whether, you know, we just need to tweak standard neo-Darwinian theory or whether, in fact, a lot of these things are already in there, you know, whether niche construction already incorporates this idea of, you know, agential behavior from organisms. I'm totally on the fence about that, because I don't think I understand the issues well enough to decide. But also, if anyone does, I think you've probably spoken to Dennis Walsh about this.
Marty Martin
Yeah, yeah we have
Phil Ball
Dennis has a fantastic book. Yeah, yeah. You know, I think he makes a great case for why we really do need to revise, you know, lots of our ideas in evolutionary theory. You know, it's a really interesting case. But as I say, I'm agnostic about how much of that is needed, or how much is understood at the moment. But these are definitely questions that this new sort of perspective on living things as agents, they're definitely questions that that perspective raises.
Marty Martin
Yeah. Okay, let's question sort of last question was. So first, I appreciate, you know, the sort of caution that's really refreshing to hear, because for a long time, we did hear, there's a need for an extended evolutionary synthesis, and that just led to a lot of argument. It feels like people talking past each other. So that take was refreshing, and your book very much conveys that perspective. One of the other things that it says is that, you know, we should maybe look for or provide better metaphors. Besides agency, I think you're clearly behind agency is one of those. But do you have others in mind? Or do you have specific ways that you think that agency can be a useful metaphor?
Phil Ball
Yeah, well, that notion of metaphors in biology is also pretty central to what I talk about in this book, because it became clear to me, first of all, how much everything we say depends on metaphor, right? All language depends on metaphor. Certainly everything in science depends on that. But I think the life sciences are particularly reliant on metaphor, because life is such a complex thing. And so that's essential, it's necessary, it's useful, but it's also dangerous, because these metaphors become so strong, "the blueprint," you know, for the genome, "the selfish gene," we introduce them for a reason, but then they get stuck. And they become what is often called "dead metaphors," which means not that they're actually dead or not used, it means that we forget that they're metaphors. And we just think this is how the way things are, and then that's when they become really dangerous. And there's no good way for uprooting metaphors. We know how to deal with theories, we test them against experiment, you know, if the theory gives a wrong prediction, then eventually they get chucked out. So we know about that. But metaphors, there's no mechanism in science for uprooting them. And I think that that's a real problem. So I really wanted to push back on some of the metaphors we use. The machine metaphor in, you know, in biology, too, I really wanted to say, you know, they're very good arguments, I think, for showing now that that's not a good way to think about living things, nor is it good to bring in computer metaphors too much.
Phil Ball
So yeah, we need new metaphors. And I don't yet know what some of those are. But one thing that did become, I felt, clear to me in the course of this book is that perhaps life needs to be, living systems need to be their own metaphor, biology needs to be it's to be the source of its own metaphor. Rather than drawing metaphors from our technologies, we don't have any technology that works in the way life does, none. And so we need to get rid of those obsolete metaphors. And for example, you know, I talked about and this wasn't my idea, either, but how the combinatorial systems that we see working in gene regulation where I you know, talked about all these committees of molecules working together in some fashion, a great metaphor for that is the way our own factory system works, where we have, you know, just this kind of few hundred sensory receptors, and we have millions and millions of different odor signals that come from those, because they work combinatorially. So there's, you know, a metaphor where we can use one living system to refer to another. The whole notion of living systems as cognitive beings, cognitive agents, cognitive systems, that is, in a sense, a metaphor that comes from cognition comes from the way our brains work. It may be something more than that, but certainly again, I think, at this stage, it's a more useful metaphor than a sort of machine-based metaphor. So that's as much as I can offer at this stage that actually, to get good metaphors for living systems look for other living systems to draw them from.
Marty Martin
That makes sense.
Phil Ball
Now, well, that's just so beautiful. And I think it's a great place to stop. Phil Ball, thanks so much for the conversation. It's been a total delight. We always ask our guests at the very end, is anything else you'd like to say? Anything we didn't cover that you'd like to have the last word on?
Phil Ball
I think you've kind of gone, it's gone in my mind from the start to the finish of my book. So I think you've been very comprehensive in doing that.
Art Woods
We won't do the appendix.
Phil Ball
So we don't need to do the appendix. So there is nothing I feel I need to add. Thank you so much, both of you. It's been fantastic talking to you.
Marty Martin
Thanks, Phil. Thanks a lot.
Art Woods
Thank you.
Marty Martin
Thanks for listening. If you like what you hear, let us know via X, Facebook, Instagram, or TikTok. Or leave a review wherever you get your podcasts. And if you don't, we'd love to know that to write to us at info at Big biology.org
Art Woods
Thanks to Steve Lane who manages the website and Molly Magid for producing the episode.
Marty Martin
Thank you to Dayna Dela Cruz for her amazing social media work. And Keating Shahmehri produces our fantastic cover art.
Art Woods
Thanks to the College of Public Health University of South Florida and the National Science Foundation for support.
Marty Martin
Music on the episode is from Podington Bear and Tieren Costello.