The Unified Neutral Theory of Biodiversity and Biogeography, published as a monograph in 2001, by Stephen Hubbell, is considered one of the most important recent developments in Ecology. What is, probably, less well-known is that the ideas in the book were originally laid out in a paper published in the journal Coral Reefs in 1997. Nineteen years after that paper was published, I spoke to Stephen Hubbell about his motivation for developing Neutral Theory, the story of its publication in Coral Reefs, and his reflections on the impact of the theory on the field of Ecology.
Citation: Hubbell, S. P. (1997). A unified theory of biogeography and relative species abundance and its application to tropical rain forests and coral reefs. Coral reefs, 16(1), S9-S21.
Date of interview: 23rd August 2016 (via Skype)
Hari Sridhar: This paper is a precursor to your famous book The Unified Neutral Theory of Biodiversity and Biogeography, but it is relatively unknown . So, I wanted to start by asking you to talk about the link between the book and the paper.
Stephen Hubbell: There is a paper I wrote about the difficulty of publishing the original paper, which was written up in a book on tropical forest ecology edited by Walt Carson and Stephen Schnitzer. And it talks about some of the ugliness, the hate mail, I got from reviewers. I have a file of it; it’s buried somewhere in a file cabinet. When I moved to California, I kept them all, because it was pretty amazing. Did you know that I’m the anti-Christ of ecology? Things like that. I tried to publish the neutral theory idea everywhere. And then what happened was Nancy Knowlton, a marine ecologist and evolutionary biologist, who was at the STRI [Smithsonian Tropical Research Institute] at the time, and her husband Jeremy Jackson, were organizers of the eighth International Coral Reef Symposium, which was to be held in Panama. And that was in, I think 1997 or 96. It was 96. And they asked me to give a plenary in which I brought up-to-date Joe Connell’s old 78 paper on non-equilibrium dynamics of coral reefs and rain forests and to make a comparison between the two. And all the plenary speakers were promised that their papers would be accepted, with minimal making sure that it followed the format of the journal and so forth. I was, at this point, several years into frustration over publishing my neutral theory. And so I jumped at the opportunity, but I told them point blank that I wasn’t going to do what they suggested. And they said, “Oh, that’s okay, whatever you want to do.” I said it’ll have something to do with rainforests and coral reefs, but kind of tangentially; it will be more of a theoretical thing. And I didn’t go into detail because I didn’t want them to go, “Oh, well, I don’t think that’s really what we want for the journal.” And so I gave my talk and there were a huge number of people, I would say several hundred people, in the audience. It was held in Panama, in the Atlapa Conference Centre, which is this gigantic convention centre that’s down near the waterfront. The plenary sessions were all in this big seating area and it was just absolutely a sea of people. And so I gave this paper, which is what the Coral Reefs paper is, and the response was kind of like “Oh, wow, what’s all this stuff?” And afterwards I could tell Jeremy and Nancy were kind of pissed off, but they didn’t really say much. They, you know, kind of rolled with it. I had relatively little interaction with people about it. I think it just went totally over their head. They had no idea, really, what I was talking about, because I had to get through it in 40 minutes or something; it wasn’t very long. And so I was zooming. So I submitted the paper to the editors and I got back this terse note saying, “Yes, we will publish your paper”, but it was not a friendly note. It was kind of like: we will honour our previous statement that we are going to publish your plenary. And that’s how I got it published. And you know, the kind of sweet irony is that all of the journals that rejected the original paper, started publishing on neutral theory all at about the same time, you know, Nature, Science, PNAS, American Naturalist. They all did it.
But the book didn’t come as a result of the Coral Reefs paper. I don’t know if it’s possible to find Google data on citations rates back then. I’m not even sure that we were doing Google Scholar in those days, and we didn’t have Web of Science. But my recollection was that the paper was pretty much un-cited until the book came out. And then people went back to look at the historical roots of neutral theory, and they go back to my 79 Science paper, which is the first simulation of neutral models of tropical tree communities, but it didn’t have the superstructure of island biogeography theory, or metacommunity and dispersal, really, in that. It only came together with immigration and speciation rate for that Coral Reefs paper; I had not done that before. That’s the history as I have reconstructed it, but the philosophers think somewhat differently. They think that I thought about neutrality as early as 79, but that was when there was a huge debate going on about neutral community assembly. Dan Simberloff and Don Strong of Florida State caused a big brouhaha between the sort of, you know, the Jared Diamond sort of community assembly rules, kind of niche assembly arguments, that there were permissible slots for certain kinds of niches and when you filled those up you had a community at carrying capacity for the island. This was actually a rather radical departure from the old island biogeography theory. I mean Jared Diamond’s ideas were not really compatible with the original model of island biogeography, even though nobody pointed that out. And I didn’t point it out at the time either, because I didn’t want to get in trouble. So after my paper in Science came out, I said, “Whoa, look at this controversy. It’s really ugly. I’m going to just forget about this for now; maybe I’ll come back to it someday.” Well, I pretty much forgot about it. I started doing the big plots. The philosophers got one thing wrong here -they thought that I was doing these big plots to test neutral theory; that was not it. I was going back to my roots on niche assembly and Janzen-Connell kinds of mechanisms. I was looking for stabilizing mechanisms, density and frequency dependence. And so a lot of my papers were about that, but early on, I was having trouble finding that evidence. So you know, I wrote that Biology, Chance and History paper, which was in the Case and Diamond Community Ecology book. And that got a huge amount of attention because I talked a lot about chance and history being major factors structuring tropical forests, which was kind of a neutral argument, but I didn’t use the word “neutral”, even in the Science paper, except in reference to Caswell’s old paper that was in Ecological Monographs, I think it was, on applying neutral models from population genetics borrowed directly into ecology but without the superstructure of island biogeography theory either. So I talked about Caswell’s paper, but I didn’t use the word neutral. I don’t even think I used it in the Coral Reefs paper; I can’t remember if neutral comes up. You’ll have to check that. But I was reluctant to call it neutral because of the neutral wars of the early 80s. And finally, I think the thing that caused me to start calling it neutral was I was shocked when Graham Bell kind of scooped my book by a few months by publishing his paper in Am. Nat., on his neutral model, which didn’t have immigration, it was just essentially a reconstruction of Caswell’s model but looking specifically at relative species abundance, He called his model neutral, and I thought, “Okay, the cat’s out of the bag, it’s probably time to start using the word neutral because people will understand it in that context”. Then I decided that in my book, I need to point out that MacArthur and Wilson’s graphical model is in fact a neutral model at the level of species neutrality. This was not widely recognized, even by them. Because the latter half of the book is all about non-neutral processes that determine the arrival and extinction of species on islands. But their graphical model doesn’t work, as I pointed out, unless you assume equivalence, because they just count number of species. They don’t say who they are. So by not naming them, it doesn’t matter what combination of species caused the equilibrium on the island; that’s not determined by their mathematical model and their graphical model. So partly that was written out of defensiveness: it is neutral, and all I’m doing, in essence, is adding a speciation rate and recasting the neutrality assumption at the individual level. I made that much clearer in the book than I did in the Coral Reefs paper. As I recall, in the Coral Reefs paper, maybe I mentioned that the neutral assumption was at a different level of organization. I can’t remember; you have to check.
HS: I think you only mention it in the context of the earlier work. Otherwise, you don’t use the word neutral in the paper.
SH: Yeah. I think that the moment that I realised it was okay was after Bell’s paper,
HS: At the time when this paper came out, in 97, you already had a draft of the book…
SH: No. I had written and formatted it in different versions for different journals. So I had lots of versions that were formatted for Nature, for Science, for PNAS, which were essentially the Coral Reefs paper. Perhaps you don’t know, but there was actually an earlier paper in which I outlined the basic theory, before the Coral Reefs paper, but it was truly buried in a book of papers from a symposium held at the University of Iowa in 1995. The book was titled Preparing for Global Change: A Midwestern Perspective, edited by Carmichael, Folk, and Schnoor. After 25 years, it has been cited just 55 times.
HS: But you cite your book as being in press, in the paper…
SH: Oh, really? You know, that’s an example of where my memory failed me.
HS: The reference for the book provides the date as 1997 and as” in press”. The title is also slightly different.
SH: I guess what had happened was that I had sent a précis for the book to Princeton, and the editor, Sam Elworthy, said, “Yeah, we’d love to do the book.” And I sent some chapters as they were being written to Princeton, but I don’t think I had the fully assembled book until 1999. But the outline certainly was there, in terms of what chapters would be in it. So I think it was hubris on my part to say it was “in press” because I don’t think it was finally accepted until 1999. It took them two years to typeset and proofread it because I don’t know how to use LaTeX. I really need to learn how to write in LaTeX. I don’t like having to learn typesetting language for math because my equations are often complicated. And it’s not wizzywhig; you don’t see what you’re going to get until you actually print the PDF. And so I’ve struggled along for years just using the Word equation function and it’s really klugey and awful. I hated it. So they had to reset everything in LaTeX and that took them a long time because we had to go back and forth to proofread all the equations; it seemed like forever. That’s going to happen again with the new theory book I am doing, on mating theory with Dr. Patricia Gowaty, which is even harder and more mathematical.
HS: The title you use for the book in the paper is “The Unified Theory of Biogeography and Relative Species Abundance”
SH: Sam Elworthy at Princeton University Press was responsible for the final title. He’s the one who stuck “Unified” in it, and I said “Why are you putting that in, it will just make everybody mad.” And he said, “Well, what you’ve done is unify the theories of biogeography and relative species abundance. That’s an important theoretical unification and if you don’t mention it in the title, well, you’re kind of… putting your light under a blanket. You should say what it is. It’s much more accurate than your original title.” So the editor at the press had a big hand in choosing the title.
HS: I’d like to step back a little bit and talk about how your interest in this came about. Your PhD was on ecological bioenergetics of a terrestrial isopod, and after that, for a while, you continue to do theoretical work related to insects, competition, food webs etc. And then you started doing empirical work on the 50 hectare plots and forest dynamics. But theoretical work on forest dynamics and diversity seems to have come much later in the mid 90s. What was the trigger for the interest at this point?
SH: In the 70s, I was really interested in behavioural ecology and I still am. In those days I was especially interested in optimal foraging theory. And my wife then, Leslie Johnson, was an animal behaviourist, and we collaborated. I met her in an Organisation for Tropical Studies (OTS) course in Costa Rica, and when we married, we moved to University of Iowa. We both had jobs, and we worked together on stingless bee foraging behaviour. These are interesting organisms; you have them in India. They are social bees. There are lots of stingless bee species in Central and South America and they have very diverse foraging strategies. Some of them have very few scouts and when they find a concentrated mother lode of nectar or pollen, they recruit a whole lot of bees that guard and monopolize the food source, driving off rival bees. Other species are gleaners and they go out after most of the resource is gone when it doesn’t pay to recruit bees any longer because there isn’t enough resource left in any one spot. So they go around as individual bees, and collect the remains of nectar and pollen after the aggressive bee species that came in first have left. We studied a whole community of stingless bees in Costa Rica, and there were really interesting coexistence questions about how many species you could pack into a bee community with these different foraging strategies, who would exclude whom, and that kind of thing (e.g. 1, 2) . So we mapped the colonies, marked bees at flowers and looked at which nests they came from so that we could actually find out their foraging ranges and territories. And then I realized, “Well, we don’t really know where they’re getting their pollen, and so we are going to go out and map a chunk of forest” where we were doing the bee study. The mapped forest was 13 hectares. We recorded which trees were flowering and which bee species were on the flowering trees. This meant we could find out where the bees were getting their pollen.” So we used this mapped forest to find out how far the bees flew when they foraged, and to see if there was a relationship between the quality of the resource, and how far they would go for it. Do resources farther from the nest attract fewer recruits, all those kinds of questions. And so at this point, I had a map of the forest and then I realized that, you know, these bees are really hard to count as they forage. They’re flying and you don’t know exactly where they go. So, since I was interested in foraging behaviour, why not work on a species that doesn’t fly? And I turned to leafcutter ants which have very nicely mappable foraging trails that go through the forest. They go to specific trees and you can record how much leaf material they’re bringing back on every trail. So you can really quantify their foraging behaviour; but you still again need a map of the forest and the trees whose leaves they are cutting. I had mapped the forest for the bee work on a private ranch in Guanacaste called Comelco. We had mapped13 hectares on this ranch for the bee work and then the ranchers told me, “Steve, We’re really sorry, we are going to clear the forest, sell the timber and plant grass so that we can increase the size of our cattle herd.” And I was devastated. You know, I’d spent two years mapping that forest, so I figured, “Okay, well, I do have a map of this forest. Is there a paper I can get out of the mapping work we had done even though the work is gonna stop? So, I analyzed the spatial dispersion of the trees in the forest in relation to hypotheses about supposed spacing mechanisms. The wisdom of the time was that seed predation interacting with seed dispersal would result in orchard-like spacing of the adult trees of a given species, a pattern called “overdispersion”. Trees of a given species were expected to be more regularly spaced than at random and not aggregated or clumped. But the big surprise was that all tree species were clumped, not overdispersed.–That was the main finding of the 79 Science paper. In that paper, I did a bit of modelling because I thought it would be nice to have – I was thinking of it as a way to ask, what patterns of relative tree species abundance will arise just by chance, by random demographic fluctuations? That was kind of an add-on, it wasn’t really a central part of the paper. At that point I was still doing foraging theory, and I moved my operation to a national park in Costa Rica – Guanacaste National Park – which was further towards Nicaragua. And that’s where I started intensively studying leafcutter ants. And I mapped the forest there too, but I never published that map. I did some work on the leafcutter ants and then gave the data to George Stevens, who assisted in the leafcutter ant research. Stevens then gave the data to Brian Enquist, who subsequently published it. Then I realized, you know, foraging behaviour is very interesting, but I wanted a larger set of questions to ask. I was not satisfied with just foraging theory. That’s when I made a conscious decision to go into forest ecology, also because I was getting interested in conservation in those days, although only avocationally. I mean, I was seeing forests cut down everywhere. And I thought, you know, we don’t really know very much about tropical forest biodiversity. We got all these theories about diversity but nobody’s really testing them very well. If you look at the existing literature, it’s all driven by forestry schools that are interested in timber production. They’re not interested in biology very much or in understanding the population ecology of trees. They just want to know how many board feet they can get out of a hectare. And I said, if you’re going to ask those kinds of ecological questions, then you need to design it specifically for asking those kinds of questions. So I then realized that I wanted to study more diverse tropical forests than we had in the dry forest. And I wanted a protected site, so that what happened at Comelco would not happen again. I realized that BCI [Barro Colorado Island] was a possibility because it was a protected reserve that had been managed by the Smithsonian for decades, under treaty between the U.S. and Panama. I’d never been there before, so in 1979 I went down to visit BCI. I knew Robin Foster. He was at the University of Chicago at the time, and I was at the University of Michigan. And we taught a couple of field course exercises together in southern Michigan forests, and I asked Robin, “You know, how would it be to do a big plot study on BCI that has adequate samples of all the tree species?”He said, “Ah, that would be great. You know, why don’t we go down there?” So we went down to BCI in the winter of 79 and walked around on the island. I wanted Robin on the project because he had done his PhD there. He is an expert in tropical botany and tropical forest ecology. Tom Croat from the Missouri Botanical Garden had just published the Flora of BCI, and he could also have done it but Tom was really a taxonomist, not an ecologist. I knew Robin was well-liked at STRI and would be perfect to introduce me to the folks at STRI. So we walked around all over the island, looking for the site for a large plot.
There is a now-famous true story about how the size of the large plots was set at 50 hectares. I asked Robin how big the plot needed to be. I told him that I had mapped 13 hectares in Guanacaste and that we found 120 species in those 13 hectares. We knew from Croat’s flora that BCI had roughly 450 species over one centimetre in stem diameter. So I said, “Robin, you know, we’re going to have to do a much bigger plot than 13 hectares, because we just don’t know if we’re going to get enough individuals to study of each species if we don’t do a large plot. And so I said,” I think it’s essential that at a minimum we do a plot at least twice as big as the Guanacaste plot, maybe 25 hectares. And Robin without any delay, shot back, “No, we’re going to do 50.” And that’s the entire scientific rationale for 50 ha plots in the global CTFS network! We had no idea what the abundance of species would be in 50 hectares when we started, and as it turned out, there were lots of really rare tree species in the plot, too rare for us to analyze their species-level demography. Fifty hectares is now seen as a compromise, allowing good sample sizes of 150 to 200 species or so, but not big enough for the remainder of the roughly 320 total species in the plot.
So, basically, before I switched to forest ecology, I was following my nose on questions that interested me in the 70s. I describe myself as having done cute biology. For me, cute biology is biology that is natural history story-telling in the context of natural selection thinking, i.e. in an evolutionary context. It’s lots of fun but often doesn’t address the really big, critically important questions in ecology and conservation in a global change world. I had followed my nose having fun with science for 10 years. But at the tender age of 35, I realized that my career was not going to be infinitely long as I had tacitly assumed in my youth. So I consciously decided to try to do science that mattered. So my shift towards forest ecology was really motivated by a personal need to get serious about the ecology that I did.
HS: You were talking about the debate between Diamond and Simberloff in the late 70s and that you made a conscious decision to stay away from it, and then come back to it later. Did that happen? I’m guessing the mid 90s was when you started thinking about these ideas more seriously
SH: I don’t know exactly when it was. I was teaching a course entitled Biogeography and Community Ecology, or something like that at Princeton. I’d been teaching this course for a number of years, and was going through island biogeography theory, when a student in the class – I think it was a female student but unfortunately, I don’t remember the name of the student – said, “but this theory is incomplete as a theory of diversity on islands because there’s no new species that can arise. They’re all species that are pre-existing from the mainland.” And I said, “Yeah, you’re right. You know, that’s a limitation of the theory. And she said, “Well, what would happen if there was a speciation rate?” And I said, “You know, that’s a really good question. I have no idea.” So that’s when I started thinking about it. I had this idea of a metacommunity by that time, and of immigration from the metacommunity that would balance local extinction. So I knew that I couldn’t get any diversity maintained permanently in a local community without immigrants coming in from outside. But what I hadn’t realized was that that’s also true globally: if you don’t originate new species through some process, the whole metacommunity diversity, you know, decays away; you don’t have anything left. Extinction is the only game. So you know, at the end of the day, I realized, “Hey, metacommunity diversity is an equilibrium between speciation and extinction on global scales, not immigration and local extinction.” So I realized that I needed to put speciation in there.
And there’s a big irony in putting speciation into island biogeographic theory. When I put speciation in, I used the point mutation model for speciation, namely that each new species is founded by a single mutant individual from some pre-existing species. In modelling speciation in this way, the irony was that the mathematics was identical to something Warren Ewens had derived years before, in population genetics, to describe the steady-state abundance and diversity of neutral alleles undergoing mutation at a locus in a population. And here’s where confession time comes up – I was such a poor scholar that I hadn’t read Warren Ewens’s paper published 23 years earlier. I didn’t know much about population genetics. So I wasted probably six weeks redoing that mathematics by myself, not knowing the solution was already out there. All I had to do was look it up. And you know, the only silver lining of that story is that I went through the punishing exercise of actually having to derive it. It was non-trivial and is at the limit of my mathematical ability. Later, I gave a talk at the University of Pennsylvania, after I’d developed neutral theory but before it was published, and Warren was in the audience. He’s a very gentle man. By that time, he was already working on human genetics and doing other stuff. He hadn’t done population genetics in quite a while. But he came up to me after the seminar, and he said, “Steve, you know, that equation you showed for speciation and the relative abundance of species looks really familiar. I’d like to check it out and see if it’s something that we have in population genetics too.” And he gave me a copy of his 1972 paper where he derived this result I was mortified. I was devastated. But you know, I got to know about it, it was a new application, and it really was nice that I was able to derive it myself. And in the long run, I was glad that I had done it because solving the math and finding the result was thrilling to me, a very educational moment. But I also learned that it’s important to be a good scholar and read the literature and go back and figure out what you don’t have to reinvent. But I was pleased that what I had done was mathematically correct. And so that’s how that happened. I had the speciation issue resolved obviously before the Coral Reefs paper.
HS: The paper was published in 97, but I’m guessing that the model and even a draft of the paper was ready much earlier, because you had submitted it to a number of places before that.
SH: Yes. It was finished by early 95; maybe even 94. It was two or three years before the Coral Reefs paper. Check out the paper in Preparing for Global Change: A Midwestern Perspective.
HS: Do you remember how long it took you to develop the model and write the paper?
SH: The mathematics was slow. It’s interesting. I wrote the paper, but then I started writing the book alongside the paper, because I was frustrated by the fact that I wasn’t getting it published. And so the idea was that if I can’t publish it, at least I will have a book. I say five or six years because, after I got reviewer comments back from Princeton University Press, the final version went back to the press in September of 2000, and then the published book appeared in April or May of 2001. So I think I was working roughly 94-2000 and the bulk of the book took about two and a half to three years to write. And each chapter was kind of a voyage of discovery because, although I had a topic and a chapter title, I didn’t know what was going to happen -what neutral theory would say about that topic. There is kind of a breathless tone to the book. And it was the excitement of the discovery of what it was doing. Each chapter I was writing was brand new, so I was very motivated and excited by it. So I think that’s something good for students to read, the way you get when you’re sort of totally turned on by something. And I think that’s part of its success. you know, even when I read parts of it again, it’s kind of inspirational, even though I’m much more laid back about it after all this time. At the time it was totally intoxicating. It was just an amazingly engrossing experience. I thought about it in the shower, eating lunch, trying to go to sleep at night. I mean, my poor wife! She was a book widow.
HS: Were you always mathematically minded?
SH: I came within a hair’s breadth of going for a math degree as an undergraduate. I eventually got a biology degree, but I was taking complex variable and I had real analysis and I had number theory – all that kind of stuff. I liked mathematics, yeah, but I really loved biology. And I figured, “Why am I taking all this abstract math that I’ll never use?” And so I took some engineering math because I wanted to apply things, like, I wanted to do linear systems analysis. So I did a lot of, you know, Laplace transforms and Fourier transforms and that kind of stuff. So I learned engineering mathematics as a graduate student. And I took an individual seminar from Professor Evelyn Fix, a professor of statistics at Berkeley, when I was a graduate student, on two subjects, numerical analysis, and probability theory, because I hadn’t had any statistics as an undergraduate. This seminar had a rather tragic end in December, 1965, when Dr. Fix died of a heart attack. Yeah, so I had a lot of math. But I don’t do math all the time, And I’m also not nearly as good as some other people. There is a joke about my statistical physics friends. I went to a meeting at the Trieste Centre for Theoretical Physics to talk about neutral theory after my book was published. The meeting was attended by a bunch of my theoretical physics friends who were working on neutral theory . We all gave talks and Patty, my wife, said, “Oh, now I get it. I thought they were your pets, but you’re their pet.” It’s true. I mean, I’ve learned a lot of statistical physics and mathematics from them that I never knew. I’m not very good, but I can understand it. I wouldn’t have known what a Bessel Function was as an undergraduate. I know what they are now. I know things like moment generating functions and other things in statistical theory but, you know, statistical physics is differential equations of stochastic processes -master equations and all that kind of stuff, and that was all brand new to me. I didn’t know any of that; and now neutral theory getting into quantum field theory. And well, it’s, it’s tough sledding. Now I have to know how to solve for detailed balance and all this kind of stuff. And it’s, it’s challenging. When I retire, one of the goals that I have is to write a “Neutral Theory for Dummies”, meaning I’m the principal dummy. You know, I’d like to read through all these papers on neutral theory, of which there are more than a 1000 now -it’s unbelievable – and try to write a primer for ecologists. Otherwise, it’s never going to make it into ecology textbook except as a kind of cartoon of its true intellectual content and merit. It’s just too mathematical. This is a shame because it’s a very useful body of theory.
The other thing that I want to say about neutral theory is that it’s one of the few truly honest theories in ecology. By honest I mean it gives you the mathematical tools for its own falsification. Not very many other theories in ecology, mathematical or otherwise, tell you how to falsify themselves. And that to me is very a Popperian thing to say. I’m very proud of its honesty in this regard. The philosophers tell me that Popper’s dead; it’s all Kuhn. I’m not sure I agree, but I must admit that I don’t always falsify stuff. I look for supporting evidence. So I’m kind of a corrupted Popperian. I believe in inference by induction. But I do think that we also make much progress in science from deductive reasoning. One of the wonderful things about neutral theory is that it generates many analytical results that follow as mathematical proofs given the assumptions. I love deduction when you can do it, but induction is essential in biology. We can’t get away from it. Maybe in physics you can sometimes get away from induction, but not in biology.
HS: I was curious about a couple of terms you use in the paper: niche assembly and dispersal assembly, which are commonplace in community ecology…
SH: I invented those terms in my book, and they caught on. I realized that Ed Wilson was very good at inventing terms, you know, going back to “pheromone” and “sociobiology” and lots of other terms, and I needed descriptive terms to capture a concept and make people think about it. One way to do this is to create terms that everybody finds useful as shorthand for the concept. They’re not perfect, because they are too short and they can’t really capture the variety of meaning. But, you know, dispersal assembly and niche assembly captured some of the elements that I needed to help organize discussion and thinking about community ecology. Another one of those, which I didn’t create, which I think is really useful is Weiher and Keddy’s environmental filtering hypothesis. That is used by everybody, and it’s a really useful concept. But it begs the question of coexistence, you know – if you filter all these species that are identical into the same habitat, because they have the adaptations to abiotic conditions to prosper there, how do they coexist? Weiher and Keddy don’t answer that question at all. Also it seems to me that the functional trait people also ignore coexistence questions almost completely; I mean, it’s off their radar. And I think this is really kind of the elephant in their room.
HS: You said that this paper got rejected a number of times before it was published in Coral Reefs. What were the grounds for rejection?
SH: The reasons given for rejection ranged all over the place. Some of them were ad hominem attacks; a lot of them were of the gist, “I can’t believe this result; it just is not possible.”The Nature review was, “If you accept this paper, I will never review for you again.” One was, “Has Hubbell lost his mind? His theory cannot be correct because it’s based on a false assumption.” Things like that. They were mostly incredulous and highly skeptical and completely unhelpful. I mean, like, nobody even wanted to consider the idea.
It was pretty universally rejected, and it was very discouraging. With the wisdom of hindsight, I’m much more mellow about it now. And if you consider this a paradigm shift in a way, that’s exactly what you expect from true believers. I think there’s a lot of true belief in ecology, as there is in any field. My feeling was that community ecology based on niche assembly theory had become rather ossified. Many of the niche theory papers that were published were confirmatory and they didn’t lead to any new research. They were like paintings. They were art forms, not science. What bothered me most was that many were not testing alternative hypotheses. So one of my motivations for the book was to propose fresh challenges to classical ideas, with testable, alternative predictions. I wanted to get them to test their assumptions, and not just presume they were true. So how do you falsify neutral theory? I mean, that’s one of the fundamental questions that’s almost never asked. How do you test whether differences among species are actually due to competition? How do you actually falsify that hypothesis? Do you actually look for data that would falsify it? Where I think all this came from is that everybody, me included, was into Darwin’s “tangled bank” – the idea that everything matters, that every detail is important to the coexistence of species. Everything is adaptive. Each species is perfectly co-adapted to every other species in their community, and the coexist because they are the best competitor in their specific ecological niche. If a species is missing from a community, it must be because it was competitively excluded. And, you know, I was getting kind of tired of this argument. I mean, I’ve been teaching it in my Ecology 101 courses for many years, and one day I just said, “You know, this is b.s.! Why am I teaching this stuff?” Or maybe it isn’t b.s., but I don’t know whether it is or not. Now what I do in my community ecology class is I have one lecture devoted to the canonical theory of ecology. I go through Lotka-Volterra equations and all this classical stuff. And then I say, at the end of it, so is this theory predictive? What conclusions do you draw from these models? I mean, the classic problem with Lotka-Volterra equations is that you can’t tell whether the species are going to coexist or not, until you grow them together and find out what they do, because you’ve got to measure the alphas – the interaction coefficients, which you cannot measure until they grow together. This is hard to do experimentally in the field, and it is impossible for communities of hundreds of tree species, such as a species-rich tropical tree community. So it’s not a predictive theory at all. It’s retrospective. It’s a curve-fitting exercise. So this is kind of my mindset. I’ve always been kind of a physicist at heart. I start from the assumption that you want as few free parameters as are absolutely required. Sort of Einstein’s dictum, you know – a theory should be as complicated as necessary, but no more so. This is just my philosophy. There’s nothing to prove there about it being the best way do science, but it is my personality, that the more complicated models get, the more uncomfortable I get. Because I don’t know whether I’m just playing games with my brain, you know, I’m mentally often in the ozone and not really doing science. So I guess if that’s a strength it’s also my handicap, because if something is really important I tend to ignore it as long as I possibly can get away with it. I only put things into a theory if I have to, to explain the results to some level of accuracy. That’s another thing. I believe in approximate theory. Most ecologists don’t. They want to put everything into a theory, complexify it to the point of untestability. Neutral theory is an approximate theory. It says, to a first approximation, species in the same trophic level in a community (for example, trees), can be regarded as nearly equal. John Harte, from UC Berkeley, wrote a commentary on our 2003 paper in Nature on neutral theory, and said it right about the value of approximation. He said that many theories in physics are approximations. They work well enough even if they are not precisely correct. He said that neutral theory in ecology is an approximation the way that Boyle’s law, PV = nRT, is; but, you know, there are no perfect gases. And the law isn’t perfect, because you don’t have Van Der Waals forces and other things. But it’s pretty damn good. So tell me what level of accuracy you want in your result, and then we’ll decide how much complexity we have to put in the model. That’s the order you go in, not the other way around. You don’t start from a very complex model and say, “Does it predict reality?” because we all know from AIC that the more complex the model is, the more free parameters, usually you have more wiggle room and you can fit data better. But more complex models are exponentially more difficult to understand and increasingly difficult to test. In ecology there seems to be a widespread hope that greater model complexity leans to greater realism and precision in prediction. But, in my experience, this hope is almost never realized; more often the opposite is true. Incidentally, this view seems to pervade the champions of Bayesian statistical methods in ecology. One of the problems using Bayesian methods in ecology to deduce causality is that it can lead to overly complex models. One big problem is there’s no clear stopping rule for how many layers of priors you pile on. What’s the stopping rule? I asked a prominent proponent of Bayesian methods in ecology that question once, and he said, “Oh, you stop when your prior is flat”.“Well,” I responded, “If you go that far, you’re not gaining anything with a Bayesian approach because there’s no information in a flat prior. That’s no better than maximum likelihood.” Anyway, what does this really tell you about what’s going on biologically? You’re not really rejecting any hypotheses. What’s missing here is strong inference, critical tests that eliminate alternative explanations.. That’s what you need to advance the science. So I’m kind of a curmudgeon about this stuff, Fortunately, I’m married to a scientist who’s also similarly inclinedand we get along famously. But I think ecology is in real deep doo-doo in terms of a lot of its methods of doing science. Much of it is confirmatory.
HS: Do you remember how many journals you went through before Coral Reefs?
SH: Let’s see – I know I did Nature, Science, PNAS, Ecology, and American Naturalist. I think I did those five, maybe a couple more. I’ve asked these journals for the reviews from them, and they don’t have them. It is too long ago. So I can’t retrieve that, which is too bad. But as I said earlier, I have copies of the letters in a file buried somewhere in my archives.
HS: Did the paper attract a lot of attention when it was published in Coral Reefs?
SH: I wish that I had tracking methods for past citation rates, but I don’t. But I don’t think it was cited more than 50 times before the book came out.
HS: So between 97 and 2001 this wasn’t really an idea that was being discussed a lot. It was only when the book came out that it became controversial.
SH: Yeah, there was kind of a dead period right after publication. There was a review by Peter Abrams in Nature and another view in Science. And those came out in the fall six months after the book was published. But I don’t think the first papers on neutral theories came out until a year or even a year and a half later. And I think it was just partly a shock effect, you know, people just needed time to absorb it. One of the earlier ones was Ricklefs’s objection to my speciation rate estimate. He argued that there were too many rare species, and the common species were too long-lived. I think we’ve now got an answer to this objection from James Rosindell of Imperial College. Rosindell realized that new species take time to speciate and also may not be abundant enough to discover immediately. He called this process protracted speciation. And Rosindell also came up with a potential solution to the excessive longevity of common species. He argued that once new species arise, they gradually loos relative fitness to later arising species, hastening the demise of the earlier-arising species. This mechanism invokes selection and fitness differences among species, so it is not a neutral model. However, the model can be close to neutral, a so-called “quasi-neutral” model, in that the fitness differences between species can be very small.
HS: And you said that after the book was published the paper also started getting cited a little more. Why do you think that is?
SH: You know, I haven’t done the research, to read those papers to find out. I did cite the Coral Reefs paper in the book, and also the 1995 paper in the symposium volume Preparing for Global Change: A Midwestern Perspective. So readers interested in the history of neutral theory would be able to track down some of it from those citations. But then, for example, people who do Wikipedia articles on you, they start looking at your bibliography. My CV’s out there and people have started looking at it. You know, it’s weird for people to be writing about your work when you’re still alive. Isn’t this something people do when you have passed away? It feels strange, I have to tell you, Hari.
HS: When this paper came out you were already fairly well established in your career. But did this piece of work have a major impact on your career and also, you know, in determining the future trajectory of your research?
I guess what I would say is that when the book came out is when the shit hit the fan. I don’t know if you know that expression? It’s an American expression, but basically it means: all hell broke loose. Not many people read Coral Reefs. It is a journal for specialists working on reefs, you know, and it had a circulation among them, but I didn’t get a whole lot of emails from people based on this paper. I don’t know whether the theory would have fallen into total oblivion had I not written the book, but my guess is that it would just be another strange ecologist writing another strange paper I can’t do that experiment but the Coral Reefs paper was pretty dry, whereas the book had a lot more attitude and excitement. So maybe just the stylistic difference between the two created more interest in the book than in the paper. The way I got queried later was: “Hey, how did you come up with this crazy idea? – the way you’re sort of doing, you know, where did this idea originate? I don’t really know, precisely. One of the things that I’m mad at myself for is that I’ve never been a diarist. Except for reading my papers, I don’t know what I thought and when I thought it, you know? I just don’t have detailed personal notes to remind me of what my thought processes were at different times in my career. And I am convinced that much of my personal history, which I remember, is probably not really very accurate. And so, I joke that if I ever write my autobiography, it’ll be “My Life: a novel”.
HS: It’s been 23 years since this paper was published. Would you say that your thinking on the main conclusions and the model hasn’t changed much from the time of this paper?
SH: Oh, I think I’ve matured a lot since I wrote both the book and the paper. Right around that same time, and I didn’t pay as much attention to it as I should have, Steiner Engen and Russ Lande wrote papers on stochastic population models and eventually they wrote a book on population biology and conservation, in which they introduced environmental variance into the models and showed that there should be a certain power law relationship between population size and demographic stochasticity but not between population size and environmental variance. This offers a way to distinguish whether population fluctuations are due to environmental causes or demographic causes, the latter being what neutral theory posits. This got picked up in the past several years by Ryan Chisholm, who’s been publishing a lot on environmental variance. And so, whereas before, I didn’t make any distinction between stochastic demography and stochastic environmental effects, we now have tools to parse those two. And in many situations, environmental variance will outweigh demographic stochasticity. That doesn’t mean that demographic stochasticity isn’t going on, it’s going on just like, you know, drift goes on in population genetics. And it’ll be very much more important in species-rich systems that have lots of rare species, because they will be more drifty than the common species. But I now believe that the reason common species are so dynamic is because of environmental variance. However, I’m not “nichey” about this. You know, there’s a lot of work by people thinking that biodiversity is rescued by rare species under environmental variability because they have different niches, and so there’s kind of an insurance effect. But our data from the CTFS [Centre for Tropical Forest Science] plots pretty much falsifies that. Basically, species richness declines monotonically with environmental variance, and the reason is that rare species are more subject to both environmental and demographic stochasticity. They go locally extinct more often. I think I am much more aware of and savvy about the importance of environmental variance now.
I think what’s happened to neutral theory is that it’s just become incorporated as a routine part of the ecologist toolbox now. It’s mostly the old people who are, you know, imbued with the classical view of community who have trouble with it, who struggle to accept it, but for the new generation, as far as I can tell, it is just part of the mix of tools they can use to understand community assembly. You know, you have to think about neutral theory to see how far you can explain things and then you can add on to that, but it’s no longer nearly as controversial as it was. And I also think that the people are moving strongly away from the idea that it’s either niche or neutral. You can have stochastic models that have a deterministic component. So that’s where I think the whole field is going. There are a lot of people publishing ideas for how to produce mixed models and this is exactly what happened to population genetics. We have all these models for selection, on top of which we have, you know, drift and mutation going on, and that’s fine. So, I think it’s reached that level of kind of acceptance that I thought would happen. You know, there are four classic phases of grief – first there’s incredulity, then there’s anger, then there’s acceptance, and finally it’s “yeah, I knew it all along”. And it’s kind of the same thing here. A lot of papers are coming out that don’t even mention my book anymore; it’s just become part of the landscape. And that’s fine. That’s exactly what I always hoped. One of the interesting things though that I don’t understand: my book got to the stage of hyper citation, you know, where people don’t even cite the original book because they haven’t read it; they cite somebody else’s writing. But the bizarre thing about my book is that, since the second year after publication, the citation rate is growing, essentially, perfectly linearly. It is absolutely bizarre. I can send you the graph. I’ve been going to Google Scholar and making a spreadsheet. It’s cited 1.5 times a day and that has been going on for about 11years. I don’t know, maybe it means there’s one small community of people who are publishing a lot and they always cite it. II would have expected it to tail off or to grow exponentially or do something different, but linear is totally weird.
I think the Coral Reefs paper was basically my need to get the thing out. But, you know, even Graham Bell didn’t know about it. That’s what I mean. It had no impact on the field, and Graham’s somebody who probably searched to see what else is published on neutral theory beforehand, and didn’t get that paper; probably because I didn’t use the word neutral. But in any event, maybe, there are yin and yang aspects to the word neutral, maybe that got enough people angry. Maybe it was a good thing to use the word neutral because it made them really interact with the material. Otherwise it would have just been yet another population model, which assumes no differences among species.
HS: I want to ask you about specific sentences in the paper. In the abstract you say, “Because of the higher intrinsic vagility of corals, the theory predicts greater regional similarity in coral reef communities than in tropical tree communities.” Is that something that has been tested?
SH: Yes, there have been a number of people who’ve tested that. There was a critique by some coral reef people on that idea. And we had a rebuttal that talked about exactly why our result was correct and I can send you the references to those papers The rebuttal is: Volkov, I., R. Banavar, S. P. Hubbell, and A. Maritan 2007. Patterns of relative species abundance in rain forests and coral reefs. Nature 450: 45-49.
HS: You discuss comparing species-area curves for different taxa and you say, “Although I have yet not taken the analysis very far, there are signs of interesting differences in the species-area curves for tropical trees in different plant families.”Is that an analysis that you worked on further after this paper?
SH: Yes. It’s in the book. There’s a figure that shows z-values for different plant families.
HS: Later on in the paper, you say that where there are high dispersal rates you are likely to have high local community diversity but low meta community diversity, and vice versa when you have low dispersion rates. And you say, “Without data on mean dispersal rates and metacommunity species abundances, it is difficult to quantify and test these predictions for either rain forests or coral reef communities.” The prediction you’re making is that, because there’s more dispersal in coral reefs you’re likely to have lower metacommunity diversity and higher local diversity. Is this something that’s been tested?
SH: Yes. That is part of what I just said. I think that is kind of the reverse pattern, that is you have more of a metacommunity like distribution locally than you do regionally. Check out the Volkov paper.
HS: In one of the figures you say you got the data from somebody called SJ Wright.
SH: That’s Joe Wright at STRI. He’s an STRI staff scientist. He’s been there for 20 years. When he was a post-doc at STRI, before he became a staff member, he went out to the Pearl Islands, off Panama, and collected data on the islands out there. I used some of the data that he had. I don’t know if he ever published that; he may have.
HS: In the conclusions of your paper, you say, “In the end, I suspect that the answer will come back: “not many” in spite of some notable success.” You are talking about the answer to the question of how many processes that are important at local scales will scale up to larger scales. What are your thoughts on that today?
SH: That’s wrong, because, you know, Bob Ricklefs has made a career in saying that large-scale continental diversity drives local diversity. And he and Fangliang He published a paper several years ago, showing a beautiful regression between plot alpha diversity and regional species richness. It’s a one-to-one relationship. He’s long been arguing that local diversity is largely driven by regional diversity. That doesn’t take into account dispersal limitation and the loss of rare species locally, but you can still have a proportionality between common species, being those that are found locally, and the overall abundance of species in the region. It does look like there will be strong correlations between the two.
HS: I wanted to ask you a little bit about the acknowledgments, about who these people were, how you knew them at the time and how they contributed to this study. The first name is John Bonner.
SH: John Bonner was a cell biologist who was at Princeton. He was a very wise man. He wrote many popular books about science. He was a friendly critic of some of these ideas, but mostly on how to write science so that it’s understood by the general public. And so he gave me feedback on a couple of my chapters in terms of making them more readable. And that’s how he really contributed. He was in the National Academy. His fame was he discovered cyclic AMP, which is the attractive chemical that makes slime moulds aggregate and form stalks.
HS: Rick Condit
SH: Rick Condit was a post-doc of mine. He had been in behavioural ecology, working on California sea lions and then he went to work for Bruce Levin and did molecular genetics work on transposable elements. And then he came to my lab and we wrote the first paper on tandem repeats, which then became known as microsatellites. We wrote the first paper ever written on microsatellites in angiosperms; it became a very well known paper. But then he got interested in tropical trees from the BCI data. He got a job at STRI first as the data manager for a while, for the beginning plot on BCI, but he was so talented that he rose up in the ranks of STRI people. He took over managing the BCI plot. I did the first three censuses and then he organized all the rest. I did the grantsmanship, but he was the resident scientist that organized the ground crews and supervised data quality. He started publishing a lot of papers on plot results, but he didn’t do neutral theory stuff in those days. But he got interested in the theory as the book came out, and published several papers on it, one of which, with Ryan Chisholm, showed that the rate of immigration and extinction of rare species in the BCI plot was accurately predicted by neutral theory.
HS: Warren Ewens, you have already spoken about. Bert Leigh
SH: He’s a theoretical ecologist who retired a few years ago from at STRI; and he has moved back to the States. He was one of my hardest critics, but I greatly respect his intellect. He initially hated neutral theory, or so he said, and he subsequently published several papers on it, mostly critical. One very important paper was to show that tree species turnover on islands in Lake Gatun in the Panama Canal are turning over and changing in abundance much faster than neutral theory predicts. He’s a strong believer in niche assembly and the Janzen-Connell effects. And we agreed to disagree, you know. He’s someone whose opinion I always entertain, but I respectfully differ from him on lots of things. He lived on BCI and so I interacted with him a lot.
HS: Bob May
SH: Bob May, I knew at Princeton. He left the year before I joined the faculty at Princeton but he was apparently supporting of my hiring at Princeton. However, he had nothing to do with neutral theory. He wasn’t objecting to it; he just didn’t know I was working on it until after the fact. He made some brilliant observations in his papers that ended up being cited in my book. One observation what that relative species abundances on large landscape scales should follow Fisher’s log series distribution, an almost prescient remark that I don’t know how he thought of it, but it is true.
HS: Steve Pacala
SH: A brilliant theoretical ecologist. I had read a lot of his papers on forest ecology and he read some of my chapters as well..I respect him greatly. And we had a fair number of conversations about ecology and the state of theory in ecology when I was at Princeton. He and his students have published some seminal papers on the BCI plot over the years.
HS: Ed Wilson
SH: Wilson was a mentor of mine. He got me a Guggenheim Fellowship, when I was working on ants – leafcutter ants. He’s always been a deep sponsor of my science. He read a draft of my book, and wrote me a lovely letter. He knew my father who was an entomologist, and he remembered my father coming up and asking him about a paper that Wilson had given at a meeting, and my father apparently had said to Ed Wilson, “I wish I’d had that thought.” Or words to that effect. Ed Wilson’s letter said “And I recycle those words with great pleasure to Steve and say, I wish I had written this book.” He described the book as “magisterial”; I loved the term.
HS: Have you ever read this paper after it was published?
SH: I probably read it before the book came out, but not afterwards. Over my career I haven’t been very interested in what I used to think. I don’t tend to read many of my old papers. You know, what I can’t stand is when some person comes up to me and says, “You said this in 1980.” And I say, “Well, that’s what I thought then. I don’t think that anymore”. Science marches on—or should. Although, recently I have had a number of people, some of them philosophers, interested in my thought processes as I was thinking up neutral theory, and this has made me wish lately that I had been a diarist. But I never kept one. Now I wonder, when did I think that, and why? I will never write my autobiography because I don’t trust my memory. I have forgotten so much. If I ever wrote my autobiography, it would have to be titled: “My Life: A Novel”.
HS: That’s exactly what I’m doing! Asking you about things you said many years ago. Do you think the way you write has changed since then?
SH: I don’t like writing for journals that have very restrictive styles. I care about control over my style. So I’ve been happily invited to write chapters for lots of books. And many of my papers that are best-cited are book chapters. I’m willing to write an invited paper for an issue of Ecology on Neutral Theory or something like that, but even there, I tend not to adhere too rigidly to the classical way of writing. And I was advised by someone… who was it..I think it was my statistical physicist friend, Jayanth Banavar, who told me, “Steve, go back and read Einstein‘s papers.” And I did and they’re brilliantly written. They’re very approachable. And virtually every one of them starts out with some paradox or puzzle that there’s some conflicting empirical evidence of something that begs for resolution. And so when I wrote my book, I remember Einstein’s approach. And so I tried to start every chapter with some mystery or some unsolved dilemma, or some overlooked interesting fact in the topic of the chapter. And the idea, quite honestly, was like scripting a movie, you know; it’s like, what’s the hook? What’s going to make this person turn the page? I wanted my book to be a page turner. And so I consciously thought about ways to make it grab people so that they want to know what the next page says. I don’t know that I did it completely successfully. But that was conscious; I was trying to do that. And I think I do that much more now than I did when I was young. So much is about communication. Science is a social enterprise. I don’t think entertainment ought to rule -it shouldn’t be infotainment –but, by the same token, if your paper’s boring, nobody’s going to care about it; nobody’s going to read it. I tell my students, “Make your papers interesting. Why does anybody want to read this thing? Separate your paper from all the others by making it more interesting to read.” I advise my students to do that all the time and I critique them on that. But there’s a fine line, you know – you can step over into getting too informal. There’s a point at which you have to back off and say “No, that’s too informal.”
HS: Would you count this paper as a favourite among the papers you have published?
SH: I am totally grateful that I got it out. I mean, I already knew all the results that were in it, so it was kind of like, “I gotta get this out, I gotta get this out, I gotta get this out.” So, in a sense, I was grateful that I had the paper and relieved that it was published. But in terms of where I was going to express a view of this subject matter: the book was coming up, and I thought, “Okay, this is a place-holder. I’m glad it’s out there.” It’s important that I got it out there when I did. But you know, it wasn’t published prominently, and for that reason, it got very little attention. In those days, search engines were not very good. So finding it was not a trivial thing for somebody. And if you didn’t know it existed, you wouldn’t have found it. So I think it had pretty much zero effect other than reducing my panic. It contained a lot of the important nuggets of the theory, there’s no doubt about that. It is a benchmark. It’s a time when I could say, “See I had it all there.”; but in terms of impact, not much. That was really the book. A bigger paper for impact was probably my 79 paper.
HS: What would you say to a student who is about to read this paper today. What should he or she take-away from this paper written 23years ago? Would you add any caveats they should keep in mind as they are reading this?
SH: That it was milestone in the development of neutral theory and ecology, in the sense that it was one of the first publications of neutral theory, even though the word neutral doesn’t appear on the paper. You can see subtle changes in the evolution of the ideas that have occurred since then, by comparing it to the book and subsequent papers to that paper. But I guess my advice to them would-be, “Although you need to be familiar with the more recent literature, also read the older literature in order to trace the history of ideas in ecology.” In the case of neutral theory, for those who are interested in the history of the ideas, it is useful to read. As an aside, with students I advise I rarely talk about what I know. I talk about what I don’t know. I don’t think of science as this giant mountain of knowledge. I think about the next ledge I have to climb and where I have to put my hooks and pitons. I guess I think that, most of the time, I feel pretty dumb. And the reason is that I’m at the forefront of science. And if I knew the answer I was going to get, why do a particular piece of science? So, part of feeling dumb comes from being right at the forefront of science, where the answers are not known. I tell my students, “If you already know the answer to your experiment, why on earth are you doing it? Are you just adding a brick to a wall that’s already 30 feet high, putting up yet another example of something already know? Life’s too short for that; find out something new.
This is why I’m not very good at reading my old papers. Number one, they irritate me, because I say “How stupid I was then. Why didn’t I think of this?” and then I say “I could have gotten here much faster had I known this, this and this then, but I didn’t.” So it just pisses me off what I didn’t know then. So I don’t get a lot of mileage out of reading my old papers. But I do get a lot of mileage about reading other people’s older papers, just not my own.
The only reason I do it sometimes is if I want to know whether I need to cite a certain paper because I said something there. So when somebody comes out with a paper saying, “This is the first time somebody ever did this”, I go, “Hey, I think I’ve said that before”, and I go and look it up. That’s the kind of thing I’ll do -I ‘ll see if they weren’t very good scholars and hadn’t cited it properly.
Here’s what I would say to a student. In science, you’re really lucky if you get to think of something that nobody else has thought about before, because almost everything has been thought about. And so one of the things that is bad about the modern era is that most students think that the only things that they have to look at are what they find on, you know, Web of Science, in the last five years. If it’s before that it’s like, in the Pleistocene, you know; it’s not worth reading. I would urge students to take a more archival approach to science, and go back in time. I’ve been reading Alfred Russel Wallace, for fun. And his writings are surprisingly modern for biogeography. He got an awful lot of things right. He didn’t have a lot of data, but he had a lot of impressions of things that he saw that were right on the money. Absolutely right. And, you know, you read all these modern papers that are discovering all these principles, but they’ve been around for 150 years now. And it’s like, wait a minute, you know, that’s not new. I think it’s really important to eat some Humble Pie, read older literature, really see what the intellectual history of ideas are, you know, and I think that’s kind of what you’re doing. So that’s where my respect for this comes from. And for a student, not in my position, but a student, that’s much more useful now, than my wanting to read my own papers again. So, just because I don’t want to read them, doesn’t mean that someone else shouldn’t.
0 Comments