Search the Site

Episode Transcript

My guest today, Neil Shubin, is a paleontologist at the University of Chicago. He’s made breakthrough fossil discoveries like finding the missing link between fish and land animals, now known as Tiktaalik. And he’s a leading popularizer of science through his bestselling books: Your Inner Fish, The Universe Within, and Some Assembly Required.

SHUBIN: My lab meetings, we go back and forth from fossils to embryos to D.N.A., but the problems are pretty much the same, right? How did fish evolve to walk on land? It’s really how bodies are built, how bodies can evolve. What are the forces behind that?

Welcome to People I (Mostly) Admire, with Steve Levitt.

One thing I find fascinating about Neil Shubin is that more than 30 years ago, he picked a question to study: How did animals transition from water to land? And he’s devoted his life to that question, mastering what seems to be an almost impossibly broad set of skills along the way. He’s not just an old-fashioned fossil hunter. He’s also become an expert in anatomy and in cutting-edge molecular biology.

*      *      * 

LEVITT: So does your fossil hunting take you to a lot of warm, sunny places?

SHUBIN: I wish. My summers are typically pretty cold because we end up going to polar places. We go to the Arctic in our summer. And in the Austral summer we’ll go to Antarctica. And I usually go into the field to answer particular questions about the history of life, so that hunt takes us to — it’s taken us to all seven continents. But, most recently, that is in the last couple decades, it’s really been up in the Canadian north, up in Greenland, up in Ellesmere Island, Melville Island, these cold places, which we think of as ice covered, but in reality, have bedrock that contains a window into the history of life and earth.

LEVITT: You must have a very understanding family. I can’t see my wife letting me go off to the Arctic every summer, leaving her behind.

SHUBIN: Yeah, it takes a village, I got to tell you that much, especially when the kids were little. It was really hard. Now that they’re older, it’s a little easier. Although the most challenging expeditions, honestly, are those to the Antarctic because the field season there goes essentially from Thanksgiving through late January. So you’re missing pretty much every major holiday.

LEVITT: Probably don’t take a boat there. Or do you? 

SHUBIN: No, we fly in. Typically I fly into Christchurch, New Zealand. And then the National Science Foundation runs a base called McMurdo Station. And you take either a jet or you take a prop plane. The prop plane’s about an 11-hour ride, sitting in the netting, you know, of this Cold War-era, LC-130, which is older than pretty much every passenger. And then it lands in McMurdo Station, which is off the Ross Ice Shelf. Which is on the icy part of the continent.

LEVITT: Is your time in these remote areas fossil hunting — is it something you do because you have to, or is it a source of joy?

SHUBIN: Oh, it’s both. I mean, for me to answer the questions I want to answer as a scientist, I kind of have to because that’s the place that has the rocks of the right age and the right type to answer the kinds of questions I’m after. However, it gets into your blood, the work up there. Being in these landscapes where few people have walked before. You know, there’s nothing in our lives that can really prepare us for the sensory experience of being in a polar landscape. Our perceptions of distance, the way we key into sound, the light and dark cycles of night and day, all these things are so different. And there’s an aesthetic joy to working in a place like this. But there’s also an intellectual one that brings me back.

LEVITT: What makes someone a good fossil hunter, other than being willing to tolerate tough living conditions?

SHUBIN: You want somebody who has sort a cognitive flexibility to be able to find things. That is, you know, we’re looking for bone on the surface where there’s lots of rock. And each place, there’s different cues. In some places the cue might be, oh, well you’re looking for white bone in a red bedrock area. But in most places, it’s not quite that simple. You’re looking for shape, sometimes different texture, you know, something that might glint in light in a different way. So there’s that piece in terms of pattern recognition that’s really important.

LEVITT: Do you actually test people on that?

SHUBIN: Oh, yeah. We have like a local field site we take them to just for fun. And, you know, you can see who’s good and who’s not so good pretty quickly. But there’s another trait that’s important and that is persistence and patience. So I can take somebody out and, in a day or two, see if they have the innate ability. But, you know, some people just grow into it really well, and those can be some of the best fossil hunters. Sometimes I have to take a leap of faith with people. Because one of the big personal traits that distinguishes a great fossil hunter is to be able to endure the days where you’re not finding fossils. Because it gets really tough. So there’s some days when the weather’s bad or you’re on the wrong rocks, or it’s just not clicking for you. You know, you’ll go without seeing anything. And you might have a whole week like that. And some people can just keep mentally in the game, and other people just turn off. And the other thing is, when you’re living in the Antarctic or the Arctic, you know, you’re living in a small tent with six other people and the temperatures might dip to minus-30. So you have to find fossils in conditions sometimes where you don’t even want to stand, you know?

LEVITT: Do you personally have a special knack for finding fossils, an inborn talent that others don’t have?

SHUBIN: Oh, Steve, if you knew me when I was like 18 or 19, you would’ve thought I was the last person who ever would lead expeditions to find fossils.

LEVITT: Why’s that?

SHUBIN: I mean, the first expedition that went on, I broke more fossils than I found. We were removing this T-Rex skull. I was in college, and I was driving a little backhoe, and I ruined the specimen — not permanently, but pretty badly to the point where the person did say, “You may not have a future in this.” And then, I grew up in suburban Philadelphia. I didn’t really camp until I was in college. And so I had a to learn a whole new skill set. And then I discovered, okay, well, I really like this stuff. It’s really fun. I really enjoy preparing expeditions and being in these wild places and being part of a team that is exploring.

LEVITT: Alright, let’s talk big picture. I’d like to get an overview of the steps that go into making a major fossil discovery. And I guess it starts with picking the fossil you’re looking for.

SHUBIN: Well, like most things in science — I think like most things in all inquiry, it begins with a question, right? What you think we really need to know, in our branch of science. My particular question was trying to understand the transition from life in water to life on land, which seems so utterly impossible, but that’s also a template to understand and to approach and explore a lot of basic biology and geology and earth history and life history.

LEVITT: And you got into that question a long time ago. What got you into it?

SHUBIN: I was looking for a Ph.D. thesis, back in the first year of my Ph.D. program. And I came in with a vague idea of what I wanted to do, but I took this survey course on the history of life.

LEVITT: You didn’t come in thinking about fish?

SHUBIN: No, I didn’t. I wanted to understand how mammals evolved to climb. Don’t ask me why, you know, it’s the college brain.  

LEVITT: I assume there’s a hierarchy and the early hominids are probably the top of the hierarchy, and mammals are pretty close to the top, and fish are probably pretty low. Is that a fair assessment? 

SHUBIN: Yeah, only worms are a little bit lower. That’s about it. We’re in the basement, let’s put it that way. So anyway, I had no interest in fish whatsoever. And so, I went to, graduate school, and I took a survey course on the history of life. It was basically, we met once a week we’d have one great transition each week, you know? So first it was the origin of multicellularity, the origin of bodies. Then it was the origin of vertebrates. It was one thing after another, culminating in the origin of humans. And in week four or whatever it was, we did the transition from life in water to life on land. And I remember in that particular week, I was saying, “Wow, we’re pretty far from understanding what we need to do here.” And this problem itself, how fish evolved to walk on land, you know, the shift of ecosystems, this great revolution in the history of life, which took all these anatomical inventions and innovations to happen — I’m like, “Oh, that’s a first-class scientific problem! I want to get into that one.” And literally that’s when I just shifted and said, “Okay, I’ve got to learn about fish.” 

LEVITT: Have you ever gone back to that professor and told him how profound the impact his course was on you? 

SHUBIN: Uh, it gets even better. That professor was with me the day we found the fossil Tiktaalik roseae in the Canadian Arctic in 2004. He and I actually became collaborators. I ended up saying, “Hey, this is a great problem. What do you think?” He said, “It’s a great problem. Let’s do it.”

LEVITT: In the overall history of life on earth, it is pretty recent that land animals emerged. Right?

SHUBIN: Yeah. So the earliest land-living vertebrates, our branch of the tree of life, are about 365 million-years-old, you know? Most of the history of life on Earth is a history of single cell microorganisms. That’s probably about 2.2 billion years of history. Plants invaded land well before vertebrate organisms. They invaded land about 400-million years ago, maybe even a little bit more.

LEVITT: And what is it that made water so much more hospitable to animals than land?

SHUBIN: Water was where the action was in terms of nutrients, in terms of the environments that would support life, deep sea vents, nearshore regions. Supporting the body is so much easier. You know, you’re not dealing with gravity. Plus, you had billions of years of history of creatures evolving in water initially, and so that sort of historical inertia defined the future to some extent.

LEVITT: And there didn’t used to be that much oxygen around. Did that also make land not a good place to be?

SHUBIN: Yeah, that’s a big deal. There really wasn’t enough oxygen to support life on land for many billions of years, until blue-green algae started to do photosynthesis and pump oxygen into the atmosphere. Moreover, land didn’t really have food sources. You know, if you and I were to take a time machine and go back 500-million years ago and walk on land, there would not be trees. There would not be bushes. There would not be shrubs. There were no plants on land. I mean, there were land lichens, there were algae, there were microbes. So there wasn’t a whole ton of food. There wasn’t a whole ton of resources for creatures to live on land. And it really wasn’t until plants invaded land, invertebrates then invaded land to munch on the plants, that our distant ancestors really had anything to go to on land.

LEVITT: You said that not very much was known about the transition from water to land at the time you started looking, but I imagine there’s no shortage of fish fossils, and there’s also no shortage of fossils of early land animals. Is there a particular reason why the in-between creatures have been so elusive?

SHUBIN: You know, you really have to be looking for them to find them, and it’s high-risk effort to do that. At the time we knew that there were these fish called lobe-finned fish, which were close cousins of land-living animals. They had elements of land-living bodies. They had lungs, some of them. Some of them had fins with like one arm bone inside them, that kind of thing. So we kind of knew the right group of fish. But I don’t think people really focused on a field program to find those intermediates. You’d think they would, but they didn’t in that particular time. And that’s what dragged me, you know, go where they ain’t, right? And so, that’s one of the things that pulled me into the problem was that not only was it a great scientific problem, but there wasn’t a whole, huge community of people working on trying to find the intermediate fossils. In fact, at the time, I think we were probably the only ones.  

LEVITT: The fields are so different that way. Like in physics, I think it’s fair to say that there are a handful of good problems, and there’s a ton of physicists working on every one of those problems. And then in economics, the set of problems we can think about is enormous relative to the number of academics out there studying. So on any given problem I’ve worked on, there’ve only been a handful of other people working on it. I would not have expected that in fossil hunting there’d be more good problems to work on than fossil hunters to go find them.

SHUBIN: Yeah, it’s really great. In life sciences, it’s kind of like that. I mean, I talk to graduate students when they’re on their beginning days of trying to find a Ph.D. thesis. I said, “You could find some problems that nobody’s working on, and make some headway into these things.” And we’ve got more ability to do that with technology. But in this case, with the transition from life in water to life on land, there’s another problem too, and that is the funding. You have to make a grant application to say, “I want to do X, Y, and Z to find something. But nobody’s found it yet before, and we don’t have any bones there yet. Just the rocks look pretty good.” And it’s hard to get money if that’s your grant application, because it’s high risk. And particularly when you have something like the Arctic, which is very expensive. So we started actually very low-risk in Pennsylvania.

LEVITT: Pennsylvania?

SHUBIN: Yeah, we started at a place where I could just do the expedition on turnpike tolls and gas money.

LEVITT: Where do you look in Pennsylvania?

SHUBIN: So the place is covered with Devonian Age Rocks. These are rocks that are about 365-million-years-old. So if you drive across Route 80 in Pennsylvania, so the northern part of the state, you’ll see red rocks along the road cuts along the highway. Those are Devonian age rocks that were formed in ancient rivers and streams. And guess what? If you look carefully at those rocks, which we did, you start to find fossils of some of the earliest fish to walk on land.

LEVITT: You stand on the shoulder of the highway? Or there are better places than that?  

SHUBIN: Yes, we stand on the shoulders of the highway. So early 1990s, I was working with a graduate student at the time. We started to visit different places along Pennsylvania, along the road and just stop the car. Cops would sometimes stop to talk to us a little bit, curious what we’re doing. It didn’t take us long. We started to find lots of fossils. And in fact, one of the graduate students found, an early Tetrapod or early limbed animal from rocks 365-million-years-old. And that was a game changer for us. Proof of concept.

LEVITT: That convinced the funding agencies to send you to the Arctic?

SHUBIN: Well, it did two things. It convinced us we can do it, which is probably the biggest thing. And then, yes, then we were able to convert that into funding, that success. The big idea for us was to focus on rocks of a particular age, about 375-million-years-old. No rocket science there. If you look at the tree of life, that’s where the big gap was. But the big thing for us was to say, “Okay, look for places in the world that had rocks of that age that were formed in ancient Delta systems, much like the Amazon Delta today, that had rivers and streams that went into the ocean.”

LEVITT: Because it’s shallow?

SHUBIN: Yeah, it’s shallow. This is where the critters would’ve lived, right? They would’ve lived in the nearshore environments. They would’ve lived in those rivers and streams. And so, if you had a package of rocks that had those environments, you were in the game. So we looked around the world for rocks 375-million-years-old that were formed in ancient Amazon Delta-like settings.

LEVITT: How do you know that a rock was formed in a delta 375-million years ago?

SHUBIN: Geologists call it a package of rocks, layers of rocks, that have sandstones, siltstones, and shales that show ancient rivers and streams in a meandering plane or delta plane system. If you know the rocks really well, you can map: where is stream, where is river, where is lake, where is nearshore environment in the rocks. There’s a classic geological signature to how those rocks are deposited in those different environments. So geologists do that all the time. You can get geological maps in different parts of the world showing what the environments were that the rocks formed in, in different places and different periods in the history of life. And so, we just looked around the world. And where haven’t colleagues worked? And let’s work there. That was the idea. 

LEVITT: Why would anyone have bothered to do that in some remote Arctic area? Is that oil drilling?  

SHUBIN: Oh, yeah. You’re an economist. It’s — what’s the word — the incentives? Yeah, it’s oil and gas extraction industries. So a lot of investment was made in Northern Canada, Greenland, and so forth, I’d say, you know, late ‘60s, early ‘70s, to map these rocks to see what their extraction potential was for oil and gas and minerals. But also, there was Cold War era politics at work. That is Canada and Denmark and others wanted to make claims on the north because there’s a lot of tussle between the Russians, the Americans, the Norwegians on land up there. So economics as well as geopolitics.  

LEVITT: All right, so now you’ve picked a place to hunt fossils. You’ve got miles and miles, I assume, of rocks that looked like they were part of an ancient delta. How do you decide the particular rock you’re going to start hammering on? Is it a conscious, well-defined process or something more intuitive? 

SHUBIN: We put a lot of effort into just that piece. I’ll get aerial photos of different parts of the Arctic. I can even use Google Earth honestly to zoom in on different parts of the Arctic. But I’ll get high resolution aerial photos or satellite photos. We’ll look to see if there are geological maps, and I’ll look for several things. I’ll look for places where the streams or rivers are slow moving, because a fast-moving stream or river will destroy whatever fossil — if a creature died in there, it’ll all be broken up. I’ll look for places that have very fine-grained sediments, where the layers were deposited gently on top of one another. I’ll look for lakebed sediments often. So I have a catalog in my head of different kinds of ecological conditions that would be ideal to preserve the fossils. And then, once I have that, I’ll say, “Okay, where can I set a camp?” Because remember, there are no roads or anything like that. Where can a helicopter pop in a camp, where we can access, within say a 10-mile radius, the right kinds of rocks. I’ll make a list of all that before I go in the field. And then before we actually put a camp down, I’ll do a recon flight of the whole area in a helicopter or a bush plane. So we’ll fly over the sites, if we have time, and then actually set down on some sites that are provisional camps. And I’ll look around. And I’ll think about: Is there water? Are we out of the wind? Are we away from polar bears? You know, life is first, then comes the science. And then, after doing the recon flight and after doing my work at home in Chicago, we’ll pop into a camp or two for a field season.

LEVITT: Did you ever run into polar bears?

SHUBIN: Actually the day after we found Tiktaalik roseae, one came oaring through camp, but they showed little interest in us, fortunately. We take a lot of preparations for them. But no, we’ve never had a real incident with them.

We’ll be right back with more of my conversation with Neil Shubin. 

*      *      *

LEVITT: So eventually your team did find that missing link, and you named it Tiktaalik. From the time that you packed up your station wagon and started looking on the turnpike in Pennsylvania to the time that you found Tiktaalik, how many years had passed?

SHUBIN: Twelve years. That was 1992 when we started in Pennsylvania. And then July 2004 when we found Tiktaalik. And it’s not just a single specimen, by the way. It’s a whole bunch of them. It’s not rare.

LEVITT: It’s not rare, but you still couldn’t find it.

SHUBIN: No, we didn’t. Although, after we found it, we were like, “Oh, wait, wait. I think we saw one of these a couple years ago.” Sometimes you find bits and pieces of things, like a back of a jaw. And there were a couple backs of jaws we found back in 2000 that were probably Tiktaalik. 

LEVITT: So after 12 years, tell me about the moment, the unfolding of the discovery that would change your life and science.

SHUBIN: Yeah, it was pretty funny. So, the season before we had found this layer of rock that contained lots of fossil skeletons of fish. So we knew here we had a site, maybe about 20 feet long, etched in the side of a hill, where six people can work, you know, sitting next to one another cracking the rock, looking to see what skeletons emerge. The joke was, “Gosh, what would be your dream in this quarry?” We called it the quarry. “What would be your dream?” And, the dream we all said was, “What if we found a snout of a flat-headed fish like sticking right out at us and that the whole rest of the animal with fins and body was like embedded in the cliff? That would be the dream.” And we’re all laughing about that. So we’re, you know — we’re cracking rocks.

Then, sort of mid-July rolls along. And my colleague Steve, who’s now a professor at Brown, cracked a rock and he says, “Hey, hey, guys, what’s this?” I came running over. And it was a snout of a fish. I could tell by the structure of the bones. It was a flatheaded fish, not just any fish. And it had teeth inside that suggested it was one of the close relatives of early tetrapods, early land-living animals. And here I had exactly what we were joking about finding the whole time. And the snout was sticking out. So if we had any luck whatsoever, the rest of the skeleton would be deeper in the cliff there. And indeed it was. So yeah, that moment, I had a couple emotions. One is relief, because this was almost certainly going to be our last field season up there. And, yeah, one of great elation, but then curiosity. What is this thing going to look like? What’s the fin going to be like? And then, we started to find three more that season right next to it. And so, all of a sudden, we started boom, boom, boom, popping them out. Then it got real quiet. Because we had to extract these things now. And here we are, mid-July, we had two weeks before the August weather comes, which basically means we can’t really work. It starts to snow and stuff. So we had two weeks to get three-and-a-half, four skeletons out, which is no mean feat because these rocks are pretty hard. They get embedded in ice. So we had to get busy pretty quickly to get them out.

LEVITT: And everybody made sure to keep you away from the backhoe.

SHUBIN: Yeah. There was no backhoe up there this time fortunately. No power equipment for Shubin on this one.

LEVITT: But somebody could blow it or not? Could someone have destroyed this? 

SHUBIN: Yeah, it would’ve been hard. We were using hand tools. You know, if you break it, it’s okay. Just save the pieces. The trick with something like that is to keep it really clean. Here we are in this rocky quarry. We’re sweeping more than digging. It’s just kept totally clean, honestly, as you can call a rocky ledge that. So that if you do hammer into any bone, you can identify the little pieces and collect them separately so everything can get home.

LEVITT: Were you planning on looking forever for the missing link, going to different places as long as it took, or were you going to give up at some point?

SHUBIN: Oh, no, no. We were going to stay on the hunt. We were going to give up this spot in the Arctic because we had already done four seasons. We had another idea to go to Western Canada or Western United States.

LEVITT: We’ve talked a lot about quitting on this show and how people don’t quit soon enough. What’s your view on quitting?

SHUBIN: Well, if I quit soon, we wouldn’t be sitting here. You know, one thing that kept me coming back was we were always pushing the curve, learning something new. So the amount of time we spent up there, there was always something new coming out, whether it was a new site with more bones, whether we’re refining more diversity of species. So there was always a reason to come back. But we felt that if we gave that one layer a good season, a good four or five weeks of cracking and didn’t find any new species, then that would be it. But you know, I’m glad we stuck with it.

LEVITT: I find this research so interesting for sure, but what’s less clear to me is whether there are practical applications of what you discovered. Is it mostly about the joy and wonder of understanding history or are there other dividends? 

SHUBIN: Joy is a lot of science for me, honestly. But no, there’s more. Because understanding our connection to the rest of life on earth, that is being able to compare fish to tetrapods, fish to limbed animals is really critical. Fish turn out to be great models for human disease. And really understanding the continuity and the connection between fish structures and our own becomes actually increasingly important as we begin to leverage an understanding of the molecular biology, the developmental biology, and the physiology of fish to understand human disease states. One area where this is particularly relevant is understanding the genetic basis of birth defects in limbs. Because we have all these genetic tools to understand how fins are built in fish, which we don’t have for limbs in humans because we can’t do experiments in them the same way. And so, being able to connect the dots of which anatomy and fossils and developmental biology are all apart becomes really essential. And one of the things to keep in mind is, as we begin to compare fish to people, there are all kinds of biologies in fish that we should be very envious of. Fish have an ability to regenerate in ways that our own tissues can’t. And so, understanding the evolutionary relationship of fish to people really gives us the tools to be able to compare experiments in one taxon or species to experiments in another species.

LEVITT: What you said about regeneration is really interesting because it seems like once you had that from evolution, doesn’t seem like the kind of thing you’d want to give up.

SHUBIN: Yeah, no kidding. 

LEVITT: Why do you think regeneration disappeared? 

SHUBIN: All of biology consists of trade-offs, everything. Something may look great, but there’s the dark side. Salamanders and fish don’t have great immune systems. We have ever vigilant immune systems. If there’s a foreign body in us, it’s identified, and it’s removed. And if it’s not removed, sometimes it’ll cause disease state, but more often than not, it’s removed. And it’s thought that the kinds of cells that are necessary to have full regeneration, they would get plucked off by our ever-active immune system. The other answer is the cost of regeneration. It’s really expensive to do. And most of the creatures that regenerate their appendages and things tend to be aquatic. So when a fish loses a fin — let’s say a neighbor bites off its fin. You know, it might have three fins. It could swim around. It might swim in circles a little bit more, but once the wound heals, it can still use its back and its fins to get around. If a cheetah loses a fore limb, it’s a dead cheetah. Why? Because it’s essentially unable to function in its environment. So, the aquatic medium allows for creatures to survive certain trauma that would kill a land-living animal. I’m just making a generalization. So one is the environmental difference. And the other is immunity, which may be an ever-increasing piece, as we’re beginning to understand.

LEVITT: One thing that’s always puzzled me about evolution — it’s easy to understand why being able to fly is a wonderful thing, but for arms to evolve into wings must take thousands of steps. And it seems like having half-formed wings would be a huge liability. Those creatures would get weeded out, so we’d never actually see wings emerge because they can’t emerge wholesale. So reading your books was actually the first time that I ever understood what was wrong about that logic. So maybe you can give an example of lungs in fish. It doesn’t seem like fish have any reason to have lungs, given that they breathe through their gills. But you explain exactly how it came to pass in a way that finally helped me to understand this conundrum.

SHUBIN: The point you bring up actually was a legitimate scientific controversy the day that Darwin published his first edition of The Origin of Species. There were some very serious scientists bringing up that exact same issue. George Mivart was one of them, a very famous curmudgeon, actually. He got excommunicated by both the Catholic church and by science at the same time. It was a pretty remarkable feat. Anyway, let’s just take the transition from life in water to life on land. You think of all that has to happen for that to transpire. You’ve got to have lungs. You’ve got to have arms. You’ve got to have wrists. You’d have to evolve at the same time all kinds of structures across the body to enable creatures to make this great leap from life in water to life on land. But the thing about it is, that’s not how evolution works. Almost every innovation that was necessary for the revolution of the invasion of land or the origin of flight, as you opened the conversation with, happened well before the revolution itself. Lungs and wrists and arms and all this other stuff, all those inventions came about well before in fish living in aquatic ecosystems as inventions to help them live in aquatic ecosystems, not to invade land.

So let’s take lungs for example. Obviously, we have lungs, and all land-living creatures have lungs. But if you look at our closest fish relatives in the ones who are alive today, they all have lungs. And in fact, lungs are primitive in the history of fish. If we map them on the tree of life, they appeared in fish well before their descendants ever took the first steps on land. And the reason for that is these fish used lungs as sort of an accessory organ to breathe air when the oxygen content of the water is unable to support their way of life. You think about water, it actually has a very variable concentration of oxygen in it throughout the year. And there are some times of year the fish needs to be active. There are some times of year the fish can get by by being relatively sedentary. So when a fish has high metabolic needs and the oxygen content of the water is low, all of a sudden gills aren’t going to cut it. That’s when lungs become this great accessory organ. And you can see this in creatures like lungfish or bichirs and others that have lungs and gills. And they go to the surface, and they take gulps of air. And then they dive again. And they’ll just be happy sitting underwater for awhile while they metabolize the oxygen.

And indeed, if we look at the lungs of these fish, they’re very similar to our own lungs in terms of their structure, and indeed even in the genes that form them during development. And that’s true with arms and legs. Primitive, seen first in fish living in aquatic ecosystems, walking on water bottoms, living in shallow streams and so forth. So, fish being good fish, living in the ancient Devonian time, say 380-, 375-million years ago, evolved many of the huge inventions that were necessary for life on land. So the idea is if you think that lungs evolved to help animals walk on land or feathers evolved to help animals fly, you’d be in really good company, but you’d also be wrong. And we’ve known that for over a century. So the take-home message is: the inventions necessary for great transitions in the history of life almost always appeared in different contexts earlier. And so, the big change isn’t necessarily to evolve a new structure, but to utilize ancient structures in new ways. All of a sudden, these great transitions make a ton of sense. You don’t have to wait for hundreds of different mutations genetically to happen to invade land. You already have everything that’s necessary. Now all it takes is the environmental stimulus or the opportunity.

LEVITT: Can I ask you another question about evolution? And this one I don’t think I’ve ever seen a clear answer to. If you look at the human body, just as an example, the complexity is truly amazing. Let’s just start with D.N.A. I recently had the mathematician Steven Strogatz on this podcast. And he has worked on the problem of how nature manages to stuff a few meters of D.N.A. into each of the incredibly tiny billions or trillions of cells we have in our body. And then for that D.N.A. to be read, to create proteins, and to reproduce itself in every new cell, and not to mention sexual reproduction and all the difficulties of that. And this is the tip of the iceberg. Our hearts beat for 80-years straight. We remember things from our childhood, even though there’s not a single cell in our body that was around back then. Evolution seems like such a crude tool. Most mutations are bad, and any one mutation is likely to have a small impact. Even putting aside how life got started in the first place, doesn’t it seem miraculous that we got from simple life to things like humans?

SHUBIN: It’s miraculous that you and I are sitting in this room. We would’ve never predicted this in 1959, right? There wasn’t a Neil Shubin or Steve Levitt, and here we are. There’s all these things that appear incredibly unique when they finally happen. But the reality is with evolution is that so much of evolution is derived from copying, repurposing, modifying things that already exist. And that turns out to be an incredibly powerful way to generate complexity over millions, let alone billions of years. You can have a section of R.N.A. protein that duplicates itself to become a much longer strand. And then those duplicates can be repurposed and evolve in new ways. So you can get great complexity happening from simplicity. By the way, we’re not the most complex critters on the planet. Yes, we have 26-trillion cells, but there are a lot of creatures that have more genes, and they do some amazing things that we would be jealous of doing.

LEVITT: What’s a good example? 

SHUBIN: Regenerating limbs would be a really good one. If we come back a couple million years in the future, maybe it’ll be cephalopods, octopuses, that have taken over the world. Maybe they’ll be the overlords. They can do incredible things. They can solve puzzles. They can camouflage their bodies in seconds. It’s really remarkable.

LEVITT: I had Sy Montgomery, she’s written about octopuses. And the parents die when they give birth to the new one. So there’s no social learning. Every new generation’s got to start from scratch. And that seems like the biggest problem the octopus has. 

SHUBIN: Yeah, that’s a pretty big deal. We’re the total opposite. We have extended childhoods that go, you know, well — I mean, they go for a long time, depending on the human.  

LEVITT: Too long. I have seven kids, too long.


LEVITT: Are there some cases where you think we did get really lucky with evolution? Things happened that maybe wouldn’t be likely to happen again? And let me throw D.N.A. out as a leading candidate because it seems really surprising to me that every complex creature out there does D.N.A. in almost exactly the same way. 

SHUBIN: You know, when you think about luck and contingency and evolution, I mean we could have the best D.N.A. in the world and, the greatest ancestors, but if they were in the wrong place at the wrong time and an asteroid hit 65-million years ago, it wouldn’t have mattered. A lot of evolution is just surviving and being in the right place at the right time. It’s one thing having an innovation, but it’s another having an innovation when the timing is right for it to flourish and gain traction. Time and time again we’re the story of that. There were creatures that had limb-like fins before Tiktaalik, but their lineages died out. It really wasn’t until the opportunity and the necessity to walk on land that one lineage with that invention flourished.  

You’re listening to People I (Mostly) Admire with Steve Levitt and his conversation with Neil Shubin. After this short break, they’ll return to talk about genetically modified fish.

*      *      *

As I mentioned at the beginning of this episode, Neil Shubin has an astonishing breadth of skills. If you’re a regular PIMA listener, you’ll remember our episode with Nobel Prize winner Jennifer Doudna, who developed the gene editing technology CRISPR. Well, Neil Shubin had the foresight midcareer to recognize that CRISPR was a game changer, and he somehow managed to turn himself into a cutting edge molecular biologist. I’m excited to hear how he did it.

LEVITT: I’d like to go back to when you were a graduate student. I assume when you came to get a Ph.D., you were already imagining that a big part of your career would be fossil hunting, right?

SHUBIN: Oh indeed. Yeah. That’s why I went to graduate school.

LEVITT: But I know once you got to grad school, you also became obsessed with studying embryos under the microscope. 

SHUBIN: Yeah, in a big way. For graduate school, they made me take this course in developmental biology and embryology, which I was taking at the same time that I was taking the course where I’d discovered the passion for the fish transition. And then, in developmental biology, I remember I was sitting there in the lab, and we’re watching chicks develop. What you do is you, like, take the chicken egg, and you put a little window in it. And you can see the embryo inside. And you can actually put a little dye on the embryo, and you can watch it develop over a period of days. And I’m preparing to be a paleontologist to see the great transitions in the history of life and travel around the world and crack rocks. And here, in a period of three days, I’m watching great transitions in the history of life inside of an egg, you know, going from a single cell all the way to a body — to watch a body being built. And there is both an aesthetic joy in seeing this just body emerge from this amorphous mass of the early embryo. But there’s also the intellectual challenges. How do those cells know how to become hearts and limbs and spinal cords and brains? What are the instructions that enable this to happen? And how do changes in that recipe relate to evolution? Evolution results from changes in development of different creatures. Changes in the recipe by which the body of a fish is built resulted in the body of a land-living animal. So how do you understand that recipe? That’s been profound, honestly.

LEVITT: Reading your books where you write about your graduate student experience, it reminded me of what an amazing thing it is to be a grad student if you love what you do. So few formal demands, so many choices about how to invest in oneself and carve out a life path. You strike me as having done an incredibly good job of squeezing the most out of grad school. And the thing that impresses me most that you write about is that you had weekly visits with an 80-something-year-old biologist named Ernst Mayr. How’d that come about?

SHUBIN: Mayr was an intimidating character because he was one of the founders of what’s called the New Synthesis of Evolution. This was a period in the 1940s where population genetics and systematics in ecology and all these different disciplines were pulled together in a new view of evolution. And Mayr was one of the architects of this. So he was just like this God. And I ran into him once in the hallway. And I was reading his great tome, Animal Species and Evolution, which was an 800-page book on how new species form, which was very influential in the field. And he looked at me and he said, “Hey, you like it?

LEVITT: Oh, you were literally carrying his book?

SHUBIN: I was literally carrying his book. And I had little, they weren’t post-it notes, but I had like little bookmarks everywhere. And he says, “You like it?” He looked at me. And I was like really intimidated. I said, “Yeah, it’s really stimulating.” And he invited me to a weekly coffee with graduate students, about 15 of them. And he’d hold court, tell old stories. And then, he ended up having a separate session with me every Thursday morning, where I’d bring a paper up to talk to him. And we’d meet for about an hour, 45 minutes. And he’d just wax about the history of science, evolutionary biology, great people in science, great feuds he had, and he had a lot of opinions about not only the science but other scientists. So I’d just sit there and listen the whole time. It was a real treat.

LEVITT: Did you learn a lot from him, or was it just fun?

SHUBIN: Oh no, I learned a ton. He was a great historian of science as well as a great biologist. He was one of the people who really brought the science of taxonomy or systematics to the fore. You think about systematics at the surface, you know, categorizing species could not sound more dull. But he actually put it front and center to understand diversity and speciation and so forth. And he brought the sensibility of a field biologist to the study of evolution in a new way. I learned about how science was done by one of the greats of my field. These were papers you’d read as part of, you know, first-year seminars in graduate school. And here I’d have one of the architects of the whole modern synthesis of evolution commenting on those papers. One thing he had, and some people have this, very profoundly, was they put their finger on the problem. Some people just work on science, collect a lot of data, and publish some cool papers. But some people have the ability to really find the problem that moves the field. And Mayr had that ability.  

LEVITT: So I visited your lab on campus for the first time in 2008. And it was full of fossils and industrious young people working on the fossils, building reconstructions and so on. And then I came back for a second visit in 2016, eight years later, and I couldn’t believe the transformation. The same physical space, still some fossils, but now it seemed like the activities were dominated by cutting edge molecular biology, projects using CRISPR to genetically modify embryos to see how genes express themself. I left that day in awe of you. And there aren’t many people, I think, deep into their career who have the insight and the energy and the talent to so fundamentally alter their approach to a problem. Was that a transition a lot of researchers made, or were you unusual among your fossil-hunting peers?

SHUBIN: People are making it more now, but at the time, no, it wasn’t. The science of D.N.A. and genetics and developmental biology got to the point where it really can inform the kinds of questions that led me to become a paleontologist in the first place. Here I’m studying my whole career to be a paleontologist — and it remains a very powerful way to know the world around us, a very unique dataset. But now, all of a sudden, D.N.A. technology, our understanding of how D.N.A. and R.N.A. work in development, got to a point where it can provide a whole new source of data on those exact same questions. The puzzle I want to solve is the same. But now, I have this whole new universe of tools to apply to it. So that was a very conscious shift into that. Where the field is now, I think, it’s pretty routine for labs to be both in developmental biology and paleontology. But at the time, it was sort of a leap of faith on my part. I knew it would be informative, but whether I could pull it off, because it’s a little bit removed from my training.

LEVITT: Yeah, totally different skills.

SHUBIN: A lot a bit removed from my training. Totally different from my training. That was the understatement of the year. Yeah.

LEVITT: But that’s what I take away, going back to your grad school approach. We’ve talked on this podcast a lot about being a specialist versus a generalist. And in one way, you’re a specialist. You’ve been asking the same questions your whole life. But fundamentally, you’re a generalist. You studied embryos. You studied fossils. You used CRISPR. You really have been tremendously varied in the approaches you’ve taken to asking the same question. 

SHUBIN: I think this comes back honestly to Mayr to some extent. The thing I always respected about him was his finding the problem — that the problem, the intellectual puzzle, is important. Then what you do is you go out and find the best tools to answer the problem or to learn to ask powerful new kinds of questions in different ways. And that’s where the embryos and the genes and the fossils really have come together, they merge. My lab meetings, we go back and forth from fossils to embryos to D.N.A., but the problems are pretty much the same, right? How did fish evolve to walk on land? Or really, it’s a little deeper than that. It’s really how bodies are built, how bodies can evolve. What are the forces behind that? How do we compare bodies of different kinds? And, how can we compare a fin to a limb? Well, I can do it with fossils. But it becomes incredibly powerful when I can compare the recipe that builds a fin to the recipe that builds a limb of a human and asks what’s the same and what’s different. And what are the fundamental biological processes at work that drive those differences and similarities?

LEVITT: Could you give an example of a project where you used CRISPR? What question were you trying to answer? And what did the research design look like to get to the answer?

SHUBIN: If you look at the skeleton of our arms and legs, you have one bone that sits at the base, the humerus or upper arm bone. You have two bones in the forearm, radius and ulna. Then you have wrists and digits. So you have these regions of the appendage, right? One bone, two bones, little bones, fingers. Well, it turns out there are a number of genes, pieces of D.N.A., that define those regions, that are active when those regions form. And then, people knew this from work on mice, when those genes are mutated, you affect those regions in particular ways. Okay, so if you compare a fish fin to a human limb, for instance, or a mouse limb, what region is not there? Well, the region not there is fish don’t have fingers. They have bones in the fin, but nothing that compares to that. So we asked the question: Are those genes sculpting those different regions in fish fins? You know, the ones that we see in mice and humans, what are they doing in fish?

Well, it turns out in fish that those genes that sculpt those three different regions in our limbs and mouse limbs, they’re present in fish. And not only are those genes present in fish, they’re present during the development of the fins of fish defining the different regions of the fish fin, such that when we use CRISPR to knock out the genes that form the wrist and digits of a mouse limb, and we do that same experiment in fish, it turns out in fish, we’re missing the terminal end of their fin. And in mouse, they’re missing the wrists and digits. So the terminal end of a fish fin and a big section that holds the fin webbing, when we hit these genes with CRISPR, we’re missing that entire region. So it’s showing that there’s a continuity there, that the origin of the wrists and digits in people and tetrapods really didn’t always involve new genes, that the region was already there in fish fins. They really came down to another subset of genes that didn’t define the region, but that made new kinds of bones in the region. So it’s a long way of saying that the basic architecture of fins and limbs have a shared genetic activity, despite the fact that the fish and the mice have been separate evolutionarily for almost 400-million years.

LEVITT: I mean, you’re all about the transition from water to land. Have you tried to do experiments where you changed the genes, and you get fish fins to look more like wrists and fingers?

SHUBIN: A colleague from Spain did an experiment like that a number of years ago. And people have tried that. You have a series of genes that differ between mice and fish. And if you insert those novel sequences in there, do you get more of a limb-like structure? That’s sort of a holy grail for people to do, honestly. But the genetic context is very different, so it’s kind of hard. What you’re going to get is not necessarily a limb. What you’ll get is, at least in the experiments that have been done, you’ll get a fish fin with like longer bones at the base, with more bones at the base, and smaller fin webbing. Something like that, which is more transitional. But remember, the milieu that genes are active, in fish and in mice is very different sometimes. So if you introduce a different gene in them, something that’s not present, it may not be able to interpret that gene very well. It doesn’t have the substrate to do that. So it may be too novel. But when we take the equivalent genes in fish and mice, we can swap them around pretty easily.  

LEVITT: You’re a little bit older than me. Are you starting to wind down, or do you feel like you’ve got some huge discovery still ahead of you?

SHUBIN: Oh, no, I hope I have a few discoveries ahead of me. We’re going into the Arctic this summer, new places in the Arctic. We haven’t been able to get there since Covid. So I had planned to go back there in 2020, 2021, 2022. So hopefully 2023 will be the trick for us.

LEVITT: What are you looking for in the Arctic? 

SHUBIN: We’re going deeper in time to find proto-Tiktaaliks, going to more recent in time to find other more tetrapod-like Tiktaaliks. So the search goes on. We’re looking now at the origin of vertebrates, trying to think about the origin of skeletons themselves in the fossil record. My lab is actually getting into regeneration, so how do we regenerate organs and repair organs like lungs and limbs and other things like that. You know, that’s why I’m in this business. It’s just never-ending questions. And what’s great is you take the never-ending questions with an ever-expanding repertoire of tools to answer them. Hey, I’d be crazy to retire.

Talking with Neil Shubin has made me reflect on my own academic career, which is so different from his. Unlike Neil, who found a big problem and hammered away at it for decades, I’m an academic dilettante. I’ve always gravitated to little questions that I could wrap my head around. And I’ve tried my hand at all sorts of different topics within and at the edge of economics. But even more fundamental difference between us is that Neil kept investing in new skills, like learning CRISPR, whereas I didn’t. When I was young, I really admired the older economists who stayed up to date with the latest advances in the field. I marveled at how Gary Becker in his sixties and seventies, he went from knowing nothing about empirical research when I arrived at the University of Chicago to asking questions about data analysis that were so insightful I would do a double take. But interestingly, as I started to become an older economist myself, I didn’t find the new innovations in my field very inspiring. The new techniques being introduced seem to me to be only marginal improvements at best. And more often than not, the increased technical sophistication of the tools actually got in the way of understanding what could be learned from the data. I therefore made a conscious decision not to make the big investments it would take to keep up with the field, ultimately turning myself into an academic dinosaur. I don’t regret it. I did it consciously, but I tell myself that if a breakthrough as amazing as CRISPR had occurred in economics, I would have invested and remained on the cutting edge. Now that’s probably not actually true. What Neil Shubin did to reinvent himself is both hard and rare, but it seems like a pretty harmless lie to tell myself. If you want to learn more about Neil Shubin’s research, check out his popular books, especially Your Inner Fish. And if you want to learn more about CRISPR, the breakthrough technology in molecular biology, you can listen to my conversation with Jennifer Doudna, the Nobel Prize winning scientist who developed CRISPR. It’s episode 67 of People I (Mostly) Admire from March of 2022. So now is that time when we take a listener question and let me welcome my producer Morgan on to help us with that.

LEVEY: Hi Steve. So a listener named Francis wrote in. He says that today several fields such as sociology and public policy seem to be conducting numerical experiments on historical data — research whose methods and material generally resembles economics. Do you believe that the barriers between academic fields are breaking down and is that a good thing or a bad thing?

LEVITT: So I think that Francis is correct to a point, because I think fields like sociology and political science are more and more using the kinds of empirical tools that economists have used: natural experiments and regression analysis. But I think that the similarities in the analysis usually stop there — that if you look a little deeper, a field like sociology has a completely different worldview than the field of economics, and that deeply influences both the questions that people ask, but also the way that they interpret their results and write them up even if they had done the same analysis. I’ll tell you a story about when Sudhir Venkatesh and I would write papers. He was a fantastic co-author and we would start the project — I would do empirical work. He was not an empirical sociologist. He did ethnography. So he would do ethnography. And we would start with the exact same material. He would explain to me what the ethnography said and I would explain to him what the empirical work said. And then I’d write one paper for economics journals and he’d write one paper for sociology journals. And if you looked at those two papers, they had nothing to do with each other. They were so different. It was remarkable how we would take the same raw inputs and then translate them into completely different statements about the world.

LEVEY: Were you making different conclusions? Or was it the same conclusions coming at it from different perspectives? 

LEVITT: Yeah, different conclusions. What we stressed would be totally different. I would have a lot of tables and figures and he would use just four or five numbers in his entire paper, where would have 200 numbers in a paper. And he would spend a lot of time describing the details of the building exteriors and the context in which these people were living. And economists, we pretend like everything we do is universal and we abstract away from a lot of the specific stuff. I think it’s valuable that these different disciplines see the world so differently, and ask different questions and sometimes come to different answers. And it’s good for public policy. I used to think that economists had a monopoly on every smart, good thing in the world. And I’ve really come to realize that’s not true at all. And I’m always impressed by how powerful disciplines are for shaping — and limiting — the way people see the world, myself included.

LEVEY: Did you ever publish something in a non-economics journal?

LEVITT: Yes, usually though with co-authors who were from the other discipline. There’s a lot of secret code in academic papers about which papers you should cite and exactly what words you use. I learned the secret code economics, but when I would try to publish, say, in political science journals, which are a lot like economics journals, the ones I was trying to publish in, I just couldn’t get the code right. And without a political scientist as a co-author, I would get rejected every single time.

LEVEY: What comes to mind is like Daniel Kahneman. You know, he’s a psychologist, but he has a Nobel Prize in economics, so he was publishing in economics journals from time to time. Is that common?

LEVITT: Kahneman is one of the rarest animals you will ever find in that he was not trained as an economist, but what Kahneman managed to do was psychologize economics. He really brought ideas from psychology into economics and changed the field to be more like psychology, but it was no easy road for Kahneman getting his papers published. He had to fight like crazy to get his papers into economics journals, and it was only really a generation later that this idea that behavioral economics — what Kahneman really started — that that was legitimate economics.

LEVEY: Are there journals that are cross-disciplinary and is there a case for journals to not quite be so siloed? If it’s so hard for academics to switch fields or publish outside of their specific academic field, is that harming science in some way, do you think? Is that a drawback? 

LEVITT; My sense is there are a number of interdisciplinary journals, but none of them turn out to be very important because nobody knows how to write good papers that really cross disciplines. That’s been my sense. It’s so hard to write a good paper in economics. The idea that I could be good enough at economics and at sociology to write something that informed both disciplines, that’s always seemed out of my personal reach. But there are very prominent journals that publish many different kinds of scholarship, like Nature or Science, would draw from all sorts of different areas of science. But they aren’t cross-disciplinary in the sense that you bring authors who are coming from different perspectives together to try and write pieces which are intentionally bridging the lines. There has been movement towards the blurring of lines between disciplines over time. I think some of the best unanswered questions are the ones that are on the edges of disciplines and require different kinds of thinking. It would be fantastic if we could do more high quality, cross-discipline research, but I also think, frankly, it’s just really hard and it takes special people with special talent to be able to pull that off.

LEVEY: Francis, thank you so much for writing. If you have a question for us, our email address is That’s It’s an acronym for our show. Steve and I read every email that’s sent and we look forward to reading yours. 

In two weeks, we’re back with a brand new episode featuring Greg Norman, the former number one golfer in the world, now in charge of the renegade LIV golf tour. LIV golf is without a doubt the biggest thing to happen to golf in at least 50 years. And there’s been a lot of heated debate about the politics and the morality of LIV, but very little thoughtful discussion of the fascinating economic model that LIV is built upon. And that’s what I want to talk about. Thanks for listening and we’ll see you in two weeks.

*      *      *

People I (Mostly) Admire is part of the Freakonomics Radio Network, which also includes Freakonomics Radio, No Stupid Questions, and Freakonomics M.D. All our shows are produced by Stitcher and Renbud Radio. This episode was produced by Morgan Levey and mixed by Jasmin Klinger. Lyric Bowditch is our production associate. Our executive team is Neal Carruth, Gabriel Roth, and Stephen Dubner. Our theme music was composed by Luis Guerra. To listen ad-free, subscribe to Stitcher Premium. We can be reached at, that’s Thanks for listening.

SHUBIN: Initially it’d be like, you know, Dr. Shubin, what kind of doctor are you? You know, are you a cardiologist or a neurosurgeon? I’d be like, no, I’m a fish paleontologist.

Read full Transcript


  • Neil Shubin, professor of organismal biology and anatomy at the University of Chicago.



Episode Video