Jeremy recently highlighted a speech by Paul Krugman, the famous economist with a column in the NY Times here . In this speech, Krugman very casually throws out that he is aligned with the Neo-Keynesian school of thought. And elsewhere in the speech Krugman points out that the Galbraith school would think this (emphasizing the societal context of the individual thinker) and the Marxian school would think that (emphasize that decisions are made collectively and not necessarily individually). And although he doesn’t say it in the speech, Krugman’s school of thought is in a death match with the University of Chicago (or Milton Friedman) school of thought.
Lakatos (my favorite philosopher of science: Wikipedia or in his own words) argues that all science starts with each individual scientist subscribing to a core research program that has certain key assumptions that are not really on the table for debate. Although this could sound unscientific, this is just a more nuanced version of Kuhn’s paradigm shift – Lakatos’ version still allows for objective advancement of science it just changes the time scale and thought scale (science as a whole, not individuals) at which objectivity occurs as scientists select successful research programs. I know of one paper that frames an ecological concept (optimization theory) in this framework (this paper). I would argue Lakatos’ core research programs look not that unlike economic’s schools of thought.
Is Lakatos right? Are ecologists really divided into camps/schools of thought like economists – i.e. core research programs that consciously or unconsciously drive and circumscribe our basic approach? Are we just too blinded by our need to appear objective to admit that we belong to schools of thought? What schools of thought do you think exist in ecology? Do you belong to a school of thought? What school of thought is your mortal enemy?
Oh man, I’ve been waiting to comment on this since I saw it in the queue. Great idea for a post, can’t believe I didn’t think of it.
“Schools of thought” surely suggests something like an entire worldview, and implies that different schools will disagree on many things. Schools of thought in macroeconomics certainly fit that description.
Here’s an opening bid for ecology: density-dependence vs. independence. Nicholson, Bailey & co. vs. Andrewartha, Birch, & co. Basically the same school maps onto the “null model wars” over the importance of competition in the late ’70s and early ’80s.
I’m tempted to say Clementsonians vs. Gleasonians, but I won’t because, like most ecologists, I only know that debate through textbook caricatures.
I hope nobody says neutralist vs. selectionist. They might have been schools of thought in evolutionary biology. But in ecology, I get the sense that nobody (even Steve Hubbell, really) sincerely believes the world is actually neutral. Lots of people want to try to test whether neutral models can make correct predictions despite not being literally true, but that’s not the same thing as having a “neutralist” school of thought.
Picking out Lakatosian “research programs” is trickier, I think. The boundaries separating one “research program” from another are pretty fuzzy, I think. Are Phil Warren’s “macroecology in microcosms” experiments part of a “macroecology research program”? A “microcosm research program”? Both? And broader categories like “modelers” or “experimentalist” are really too broad to really be research programs or schools of thought, despite the fact that they sort of mark out real (fuzzy, but real) boundaries in ecology in terms of the interests and skills of individual investigators.
I taught about Clements and Gleason in Intro Bio on Monday. I told them that there was this big debate over it at one point, and that it was really contentious — and then immediately realized that they probably all thought it was insane that anyone could care enough about this topic to get so worked up about it.
Good for you. Whenever anybody tries to argue about whether contemporary ecology is more Clementsian or Gleasonian (which fortunately seems to be rare), I just say “neither” and walk away.
Your comments on neutral theory raise an interesting point. I think most ecologists are deeply embedded in an adaptationist school of thought, rather blissfully unaware that there are genetic constraints and phylogenetic constraint schools of thought opposed to this.
Just to make it a little more controversial than the old density dependent/independent debates (and get a rise out of you Jeremy), I would say there is a population process school of thought that says that defining things through population dynamics/differential equations is the central route to understanding. This in turn leads to a very species-interaction focused view of community ecology. Hubbell never broke out of this paradigm because he still modelled population processes. But Ricklefs has been arguing that we need to pay more attention to historical contingency but also for more organism-environment interactions (his communities don’t exist paper does put out the alternative of a Gleasonian like focus on species one at a time). Some of the vegetation ecology clearly falls in this as a competing school of thought to. Macroecologists might be starting to fall into this camp as well. I certainly to my surprise find myself evolving away from the population paradigm over the 15 year time span of my career.
Re: the “population process” school, yeah, you could be right that that’s a school. I’d certainly count myself as part of such a school, if it existed. I don’t know that it’s necessarily species interaction-focused, though (though I can see why you’d say that). Neutral models for instance are population process models, even though neutral models say (roughly) that all intra- and interspecific interactions are identical, and don’t focus at all on the consequences of those interactions (since in a neutral world those interactions don’t produce any selection). And I’m not sure if “population” process is the best name for this school, since I think the core of it is thinking in terms of state variables, inputs, and outputs. Births, deaths, immigration, and emigration are the inputs and outputs in the case where your state variables are species abundances at some location. But there are different inputs and outputs and outputs for other state variables. For instance, if you’re working with birth-death-type models of speciation in order to ask whether they can reproduce observed phylogenetic trees, where the inputs and outputs are “speciation” and “extinction”, I’d probably put you in the “population process” school. If you’re doing ecosystem ecology and measuring amounts of matter in different compartments and rates and efficiencies of matter transfer among compartments, I’d probably put you in this school.
I think Morgan’s right that the opposing school may be too ill-defined to be called a school, though I don’t know that it’s completely inchoate either. Just yesterday I actually started mulling over a post on how to better define this opposing school (the constraints school? the large-scale school? macroecology?), and push people to think about how it relates to the population process school. Basically, I think people in this alternative school are far too quick to dismiss the relevance of population processes, and do so for reasons that I don’t think stand up to scrutiny. As Morgan says, schools of thought become a problem if they prevent you from asking interesting questions. This post I’m contemplating would basically be old wine in new bottles, as I have an old post on this that I like a lot but which has never drawn any commentary.
The differences in evolutionary biology between microevolution and macroevolution seem analogous to the differences you’re picking out here. Microevolutionary types–population geneticists, people who use game theory to study adaptation, etc.–are the equivalent of the population process school. Macroevolutionary types are (mostly) analogous to the other school.
Interesting that you say you find yourself moving away from the population process school, as in comments on other posts you’ve struck me as a very population process-oriented macroecologist (saying things like “microcosm experiments are key to the future of macroecology). And in our discussion of evolution-as-seen-by-Krugman you identified with the microevolutionary types and slammed Gould. Are you in one school when it comes to thinking about ecology, and in the opposing school when it comes to thinking about evolution?
Whether these different schools of thought are “competing”, or merely “different”, is another question. Could be some of both. Note that by “different” I don’t mean “unrelated”.
p.s. Another candidate for schools of thought in ecology (concerning rather narrower issues than the ones we’ve been discussing): Tilman vs. Grime on how to think about competition. I have some anecdotes I could share about this, but I don’t know that I should share them on a blog…
Good food for thought.
Personally, I am a little adrift without a strong school of thought these days. On evolution I am definitely adaptationist and I see that spanning micro and macro. On ecology, more than anything these days I am an empiricist. I hate to say I’m an anti-modeller since I spent not only my whole ecological life but 10 years before that as a business consultant as a modeller. But I really want to see models start earning their keep in ecology more than I think they have. In evolution they clearly have earned their keep (or maybe I should say did and are now starting to become as unfocused as they are in ecology?)
I think the question of what is the opposite of the “population process” school is a really profound one. I agree the opposition is diffuse and vague. Is this a cause of a consequence of the dominance of the population paradigm? If there was one clearly defined opposition school, would it change the story? Personally, I suspect it is healthy to have not just a two-school polarity, but a multi-school competition – matches reality better. But I would like to see the non-population views coalesce into two or three clearly articulated schools. Its one of the reasons I’m blogging – I want to think out loud and told why I’m wrong. Although clearly I’m still just circling up to this issue.
Blogging is great for this sort of thinking out loud. I’m sure I speak for many readers when I say it’s great that you’re willing to take the time to think out loud about this stuff.
You and Morgan both really ought to go read that old post of mine I linked to (on “why doesn’t community ecology erase the signal of historical biogeography”). It’s terribly-titled and so may not sound relevant or interesting, but it’s totally both those things. I’ll still do a new post at some point, but the new post will be way better if I first get some feedback on the old post. (That old post was inspired by a conversation with Bob Ricklefs, by the way, whose name has come up in this thread…)
Brian — I’ve wondered about this too. I suspect you’re right that the population paradigm in community ecology has been so pervasive that there really hasn’t been time or ability for a coalescence of non-population paradigms to emerge. The non-population paradigm has been pulling from complex systems, ecosystem ecology, and macroecology. Maybe these are the alternative schools for how to study communities? As a young grad student/postdoc I saw myself as an ecosystem ecology style community ecologist…
Jeremy– speaking as someone who straddles the community ecology/macroecology divide, I (at least) don’t think that population interactions are not relevant for macroecology any more than I think atoms aren’t relevant for organisms. The question is do we have to explicitly model the population level to be able to understand and predict macroecological patterns? To me that is the core of that debate and not one that anyone has answered to my satisifaction, though it is something my group has been gnawing on the edges of recently. If you haven’t read Brian’s most excellent paper on the idea of mechanisms, I highly recommend it: McGill & Nekola (2010) Mechanisms in macroecology: AWOL or purloined letter? Towards a pragmatic view of mechanism, in Oikos.
Now, amusingly enough, I’m off to teach about Clements and Gleason, though in my case my students are actually reading those papers plus excerpts from Whittaker so we can think critically about what they are actually saying and what they actually mean for understanding communities (if anything).
Re: whether or not one needs to explicitly model the underlying microscale processes, that’s an easy one. No (or at least, not necessarily–if you can do it,, which you sometimes can, it’s often really useful)
What I would say, though, is that if you’re not explicitly modeling the underlying microscale proceses, you need to make sure that you’re not implicitly modeling them in a way that doesn’t make sense or that contradicts known information. For instance, I’ve argued in the past, and will probably have to argue again in future, that randomizing species x site matrices does not in fact remove all and only the effects of interspecific competition on those matrices. People who do that are (by their own admission) trying to avoid having to explicitly model the underlying microscale processes (which include but aren’t limited to competition). But the problem is, their randomization approach is only valid under a very particular and highly implausible model of competition (a point I plan to prove in a paper I’m planning to write…). As another example (the one I raise in that old post you and Brian need to go read), if you say that large-scale biogeography is all down to things like speciation and vicariance events and etc., well, what are you implicitly assuming about microscale processes in order for that to be the case? Because you’re clearly implicitly assuming something. A third example is local-regional richness relationships, which were long misinterpreted based on implicit, incorrect models of how microscale processes of colonization and competition work and what their macroscale effects should be expected to be.
What I’m getting at here isn’t really about macroecology vs. microecology at all. It’s bigger than that, or different than that. It’s about people making implicit assumptions, and not recognizing or pursuing the full consequences of their assumptions. Zombie ideas about the intermediate disturbance hypothesis aren’t anything to do with macroecology, or “emergence”, or whatever, but they’re another example of the same problem.
Great answer Jeremy. I’ve run into a lot of people who would say the answer is yes you have to go to microscale processes, so its refreshing to hear something different. And I definitely intend to rise to your challenge and address the question about your biogeography post.
My thought about incorporating microscale processes:
1) If you can scale up from the microscale to the macroscale then you should (even if its hard). Its just so elegant. it integrates across separate domains of knowledge, it gives us independent tests of parameters, it makes predictions at multiple scales which allows strong testing, and etc.
2) But so often you cannot (no matter how hard you try) scale up. I have a whole post in the works for why you cannot. But for now I’ll say two phrases: chaos and Jensen’s inequality. Here is where it gets tricky. As I see it there are a few opinions about how to proceed:
a) if you cannot scale up and just look at the emergent macroscale processes and phenomena it is bad science and not worth doing
b) science is the art of the possible – you should proceed if you can do “useful” science (make predictions, understand things, or pick your favorite version of useful)
c) you seem to be trying to stake out a middle ground here – go ahead and work at the macroscale but beware of being bitten by glossing over the microscale. Or maybe you’re saying you are making assumptions about the microscale whether you mean to or not so you can’t avoid it?
In general I agree one needs to be clear about assumptions and work them through. But to me a statistical-mechanic like assumption of there are so many details going on that they cancel out and a law of large numbers/central limit kind of pattern can emerge seems in many cases like a safe assumption and license to ignore the details. We ignore details all the time. Quantum mechanics is of limited value for modelling anything much above the Bohr atom. We jump to macro patterns and processes like the ideal gas law or Michaelis Menton/Arrhenius chemical dynamics with little consequence or penalty.
Re: your 2c: Yes.
Re: statistical mechanics, part of what this future post I’m planning will try to get at is what *exactly* it means to say that complex underlying microscale processes all just “cancel out”, “average away”, or just create “random noise” around some aggregate “signal”. Also when *exactly* we expect that simple picture to work. Also how *exactly* to interpret the “signal”. For instance, insofar as the “signal” is itself a function of the underlying microscale processes, then scaling up doesn’t really buy you the ability to *ignore* the underlying microscale processes; all it’s done is *summarize* them. Of course, summaries can be *very* useful–but if you forget that they *are* summaries, or forget what they’re summaries *of*, there are going to be tears before bedtime. But before I can write a post about that, I need to read up on statistical physics so I actually know whereof I speak.
In that future post I’m also going to try to get to grips with other sorts of arguments that people tend to deploy in order to justify the putative independence of the macroscale from the microscale. Like separation of timescales arguments, for instance.
Re: Jensen’s inequality, yes, nonlinearities at the micro level make it hard to scale up to the macro level. But they’re also precisely why it’s important to do so. Insofar as the microscale processes are nonlinear, your verbal intuitions and implicit assumptions about how the macro and the micro are related are pretty much guaranteed to be absolute crap. Further, I strongly suspect that nonlinear microscale processes are precisely *not* the sort of processes that one would expect to just neatly cancel out or average away at the macro scale.
Re: the ideal gas law, interesting you should bring that up. One of my very first posts for Oikos Blog, now archived here (https://dynamicecology.wordpress.com/2011/05/03/why-macroecology-needs-microecology-revisiting-an-oikos-classic/) is about precisely that. Put briefly, I don’t think the ideal gas law example has the implications it’s sometimes thought to have. The ideal gas law is actually a paradigmatic example of the power of using explicit microscale assumptions to predict macroscopic behavior. Indeed, extensions of the ideal gas law to cases like small volumes (where the original ideal gas law breaks down) illustrate the importance of *not* neglecting seemingly-unimportant microscale details.
On a different but related note: multilevel selection models in evolution, based on the Price equation, are a nice case study where people have thought hard about the relationship between what happens at two hierarchical levels of organization (among groups, and among individuals within groups). For instance, one can show quite neatly how what a group-level evolutionary model would call “transmission bias” (i.e. bias in “group-level inheritance”) can be redescribed as the effects of within-group selection. See Samir Okasha’s very fine 2006 book on evolution and the levels of selection.
Jeremy – sounds like we are about to have dueling future posts. Game on! 😉
More seriously, I agree building up is a good idea when you can. All in favor of it. It should be mandatory when possible. I just believe the space where you cannot scale up from micro to macro (and I mean literally cannot, maybe even in some cases mathematically provable that you cannot) is large and interesting and I am unwilling to avoid that space. I refuse to just throw up my hands and leave that space and I also refuse to wait 50 years for our math to evolve. Scaling up is a nice-to-have, not a must-have.
Are you really saying that the central limit theorem doesn’t give me permission to ignore details under some circumstances? Because once we acknowledge we can ignore details in some circumstances and that we cannot in some cases (which I gladly stipulate) then the whole tenor of the conversation changes. It is about how big each of these spaces are. And which space is more interesting. And …
The ideal gas law is now derivable from microscale phenomena (first so derived in 1856 according to wikipedia) which then allows generalization into “non-ideal” conditions (one of those aforementioned benefits of building up that I agree exists). But that is not how it started. It started out in a highly phenomenological pattern-based fashion assembled from smaller pieces like Boyle’s law (1662). The fact it isn’t totally general doesn’t invalidate its utility. Newton’s three laws are pretty useful even though you have to go to a vacuum in outer space to find where they’re completely true. And I dare say the Ideal Gas Law was useful for a long time before it got a micro-scale process foundation. I just don’t see the “tears before bedtime” in using the ideal gas law for a long time without the micro-scale-process-safety-belt you seem to think essential. I see a lot of early-mid 19th century chemical engineers fueling the industrial revolution.
Sure, the central limit theorem is useful. My question is, precisely what cases of interest in macroecology are cases that are central limit theorem-like? It seems to me that one often sees the central limit theorem trotted out in rather a broader array of circumstances than are actually justified.
Re: the ideal gas law, I did not mean to deny that it is a useful empirical law, and that it was useful before microscopic derivations of it existed. Sorry if I wasn’t clear.
Part of what I want to try to do in that future post (which in my mind is getting more ambitious, and so vanishing farther into the distant future, by the minute) is bring this discussion back to concrete ecological examples. Analogies from other fields can teach us much–but I think at this point it’s useful to return to concrete cases in our own field. I think this thread, and previous posts, do a nice job of laying out the relevant general principles and considerations. Next I want to use those general principles to think about particular ecological cases (besides the few I keep harping on because I know them well).
I’m sure everyone will enjoy our dueling posts when we write them in 2025.
Re: your unwillingness to give up on studying cases where scaling up is hard, I certainly wouldn’t say you should. All I’d say is, don’t then claim, explicitly or implicitly, that you know how to scale down in such cases. I know YOU wouldn’t, but others would and have. We don’t disagree on this. Like you, I’m all for learning as much as it’s feasible to learn about whatever it is we’re studying–and no more. Like you, I’m all in favor of squeezing as much information as we can out of the available data–and against fooling ourselves that there’s more information there than there actually is.
Absolutely – even if we (referring to all ecologists here) could just agree that you can’t generalize from large scale down to small scale nor (and here is where you may disagree with me) from small scale up to large scale without a lot of work justifying that such a leap is valid (i.e. default to you can’t carry results across scales assuming they are unchanged), I could leave the field happy. Scale matters!
2025 sounds about right for our posts! This is hard stuff. I agree at some point we have to turn the conversation to how this works in ecology.
I’d agree that there is a population process school of thought. Though I don’t think there is a single school that is an alternative, but maybe that’s because I’m in the non-population process school of thought and all I can see is the diversity within my school and not the bigger picture!
I would also say there is an experimental and a non-experimental paradigm in ecology. I don’t know if that falls into the realm of what you’re envisioning, because it’s not a conceptual school of thought that directs how we conduct science, but an approach to science that I think can end up directing our questions.
At some point with my students, we talk about how everyone is indoctrinated in a way of doing science and thinking about our field. Maybe it’s drilled into you by your advisor, maybe it’s just a pervasive culture in your department or field. I think the important point is to realize that this is happening and to avoid the worst paradigm of all: only my way/viewpoint/tool results is actually a valid way of doing science. And of course, in that conversation I admit that I’ve been indoctrinating them in my view! 🙂
Hi Morgan – I agree with what you said.
You raise some interesting questions:
1) I agree self-awareness and openmindedness are the keys to being a good scholar in a world where everybody has a bias or a school of thought. I wonder why economists are so open about this and ecologists are so hidden about this? As an outsider, the economist approach seems so much more rationale.
2) And I agree that some of the biggest divides in ecology may be methodological. What constants convincing evidence. How do we advance the field. Jeremy’s redefining population ecology to differential equation ecology fits this too.
OK, here’s one which I feel like I encounter a good bit – Reductionism versus Emergent-Properties. How necessary is a reductionist worldview for understanding large-scale emergent properties? I feel like I run into this argument all the time whenever I present food web or BEF work. Half of the audience seems to think there is little value in looking at the behavior of emergent properties. Half does not.
Have you guys experiences this, and where do you fall in this argument.
(n.b., my typical answers to whether I fall on one side of a dichotomous argument or the other is “Both!”, but, well, this seems like a thread where one comes to have an argument).
Oh no it’s not.
And to continue, I think you’ve drawn up a false-dichotomy between reductionist and emergent-property schools, but you’re probably just characterising your audience with this, rather than a strongly held opinion (which your “Both!” comment suggests).
I agree that the answer is usually both (also see Morgan’s comment on open-mindedness) but also agree that occasionally its more fun to argue a side!
I definitely see the reductionist/emergence contrast all the time. And I see it not just in the properties or patterns as you mention but it really comes out in the processes to. I get told all the time that I have to take things down to population processes to be mechanistic.
I’ve had the same experience as Brian re: the reductionist/emergence contrast and it often involves the explicit or implicit statement that mechanism is found at the population-level or below (depending upon the question and what the person demanding mechanism studies).
I don’t know why the concept of schools of thought is not more widely discussed in ecology. Are we avoiding discussing them purposefully or are we generally not aware? My discussions with a wide range of scientists over the years makes me suspect that latter.
To throw out another one (at least in Community Ecology) – Top-Down v. Bottom-Up. I seriously thought we had moved past this by now, but, I keep being shocked to find that many people have not.
(and, Mike, that IS a prime example where the answer is “both”)
Wait, top down vs. bottom up is still a thing?
Although even if it is still a thing, I definitely wouldn’t call it a matter of opposing schools of thought. People who disagree on one issue, however vociferously, aren’t thereby members of opposing “schools of thought”.
I really wish it wasn’t. But I keep running into it in the world. Like I said, I’m shocked every time. Maybe it’s more of a marine thing that we keep seeing it? Although, I don’t think so.
You clearly need to hang around different people. Next time someone brings this up as an issue, tell them that 1993 called, and it wants its vociferous debate back.
And I do think it is about schools of thought. Do you fundamentally believe that the world can be explained via resource supply, or do you fundamentally believe that the world can be explained by trophic control. It leads to a very different worldview on everything from population dynamics to evolution – if you think it is an either or thing.
“And I do think it is about schools of thought”
Really? I don’t recall top down and bottom up folks, even at the time, disagreeing about much else besides top down vs. bottom up, and I don’t know that that one issue itself is really so fundamental as to comprise an entire worldview. I don’t know, I guess it could be. But it seems like such a narrow thing to define one’s worldview. I mean, there’s all sorts of important issues in ecology that have nothing to do with resource supply or predation. Something like density-dependence vs. independence (=biotic vs. abiotic, =endogenous factors and feedbacks vs. exogenous factors and no feedbacks) seems to me to be much more broadly important. Not to say that framing those things as opposing alternatives is a good framing–it’s not–just that those things seem to me to be much more fundamental than top down vs. bottom up.
You definitely need to hang around different people!
p.s. As someone who’s argued that there are vociferous, unsettled debates that we nevertheless have no choice but to keep fighting, let me emphasize that top down vs. bottom up is NOT one of those debates!
I mentioned in a comment on Jeremy’s earlier that freshwater ecology folks do talk about having different schools of thought. The two that get mentioned the most are the Wisconsin school (which tends to be more ecosystem-oriented, and more focused on really big scale experiments) vs. the Hutchinson school (which tends to be more evolutionary/population-oriented, and more focused on small scale, highly replicated experiments). (A case can be, and has been, made that there is a third school, the Needham/Cornell school.) It does have a big impact on how science is done, and I think the key is, as Morgan mentioned, recognizing your biases.
So Meg, do you think that’s basically just a local, aquatic manifestation of ecosystem ecology and population/community/evolutionary ecology? And if so, is that really schools of thought (meaning roughly, different worldviews that address the same questions but give different answers, like monetarists vs. Keynesians in macroeconomics)? Or is it more just different subfields (i.e. different, possibly-related-but-nonetheless-distinct questions, pursued by different means)?
EDIT: Contrast density-dependence vs. independence. That was clearly a within-subfield thing. It was population ecologists vs. population ecologists.
I guess I’d have to think about this more, but I think it’s sort of both of the things you were getting at (though I will admit that I know very little about economic theory and worldviews!). To me, I would count having different views about what are the important questions to tackle in nature, and how to go about tackling them, as having different worldviews. But, in that sense, it is sort of like the ecosystem vs. population/community/evolutionary ecology thing. Have I successfully argued myself in a circle yet? 😉
species richness on the x-axis vs. richness on the y-axis. two schools of thought? anyone think so?
Nah, that’s just different questions. Like most things, species richness can both affect, and be affected by, other things (including the same other things). The world is full of feedback loops.
As somebody who had a comp question about whether diversity caused stability or stability caused diversity, I am going to vote yes with SB on this one. Although I think to turn it into schools of thought one would have to generalize out a bit. Might this boil down to an abiotic drives biotic, vs biotic drives biotic competing world views? It might also be a question of space and time scales. (Stability – here meaning climatic- driving diversity is very global scale; diversity driving stability – here meaning biotic community – is a much more local scale question). I think a lot of conflicting schools of thought really boil down to a “favorite” scale to work at and a belief that things at other scales are uninteresting.(I call it “scale snobbery”). This overlaps with the reductionist/emergent paradigm Jarrett & Morgan are talking about too.
Brian: I also had to wrestle diversity~stability demons for my comps! Species richness rarely appeared in literature as the x-axis till the diversity and ecosystem function debate of the ’90s (and later). does anyone know of the first paper that put it there?
SB: The notion that biodiversity would affect ecosystem function traces back past Darwin, though not using those terms, of course. Indeed, the very first “ecology” experiment ever conducted asked whether more diverse combinations of grassland plants produce greater total biomass. See Hector and Hooper’s Science paper on “Darwin and the first ecological experiment.” Insert your own joke here about how everything old is new again, or about how there’s nothing new under the sun, or about the rate at which ecology “progresses”. 😉
Recent interest in BDEF traces to a 1993 book (edited, if memory serves, by
Peter VitousekHal Mooney) and Tilman and Downing 1994 Nature (with Naeem et al. 1994, 1995 following close on its heels).
Many thanks to everyone for the really interesting conversation (and to Jeremy for the link to his older blog post).
I think that Brian’s most recent comments match my casual observations of “schools of thought” in ecology. In particular, I’ve noticed real divisions between ecologists who give priority to either abiotic or biotic controls over ecology. I work with terrestrial vegetation models, and in my field there are people who view water, or soils, or interspecific competition, or something else as being the “real” control over the communities and ecosystems they study. These views then shape how they design their models, field/greenhouse experiments, and studies. Often, I think this division can manifest itself as “scale snobbery”, but I think that “scale snobbery” emerges from the particular control of interest (e.g., moisture availability) and study system (e.g., montane forests vs. hummock wetlands). I have worked with many talented and successful ecologists who tell me that grassland community dynamics in the Pacific Northwest with be driven largely by: soils; plant pathogens; changes in the jet stream; etc. And each ecologist can point to a body of research, and her/his own work, supporting these assertions. Of course, each research project was designed with a particular ecological control in mind, so maybe we shouldn’t be surprised that a study across a steep moisture gradient finds a strong effect of moisture on plant growth rates, community dynamics, etc.
To stir the fire just a bit more: I think there’s a big difference between “schools of thought” and “scientific paradigms” a la Kuhn. In The Structure of Scientific Revolutions, Kuhn defined a paradigm as a way of viewing the world that is fundamentally incompatible with any other way of viewing the world. Too often (like the comments above) I see people referring to any new idea in science as a paradigm shift. That’s sexy and exciting, but not accurate. Sure, everyone would like to be responsible for a paradigm shift, but the reality is that no one person (not Copernicus, Galileo, Kepler or Newton, for example) is singly responsible for a paradigm shift. A paradigm shift is much bigger than the work of an individual scientist. So, Brian, when you wrote above how you used to work in a population paradigm, but now you don’t, that seems to miss to the point of what a paradigm is. (And I merely intend to use this as an example, not an ad hominem attack.) If there really were a “population paradigm”, but now there were some new paradigm, it wouldn’t be possible for your work to evolve from the old paradigm to the new. Your work could shift, but that shift would be abrupt, and would invalidate your old work – hence Kuhn’s use of the term “revolution.” The reality is that these are just different schools of thought within the same paradigm. What is that paradigm? Beats me – and we won’t be able to tell until some time after we’ve shifted to a new one. But part of why I enjoyed this thread so much is that people are trying to identify different extant schools of thought, and actually calling them just that, without trying to turn them into bigger conflicts than they actually are.
Thanks Gabriel. Obviously much I agree with.
Just to be clear, I personally think Lakatos got it closer to right than Kuhn. Because you are right that Kuhn’s paradigm shifts are just too big and rare in time. But there is something about Kuhn that then sort of implies no real science is happening between rivet-popping “paradigm” shifts. I know he calls it normal science and validates normal science, but it is still sort of the implication. Lakatos talks about research programs instead of paradigms and I think it is much closer. Lakatos specifically allows scientists to switch research programs within their career and they are less grandiose and more gradual.
Now as to how I slipped into using the word paradigm instead of school of thought: a) I’m not using it in Kuhn’s sense because I agree most of what we are talking about is not so grand (not sure much of science is so grand) and b) population paradigm just has a nice alliteration 😉
One of the main differences among ecologists’ worldviews concerns conceptual generalizability. One perspective, a view held primarily by physicists, is that the identification of generalizable predictive laws governing ecosystems is a feasible and worthwhile research goal. The other group, taking cues from 18th and 19th century naturalists, seeks meaning in the unique aspects of a particular system, and the differences among them. I think the people whose work brought about ecology’s most successful principles (eg., Darwin, Malthus, Hutchinson) balanced these two schools of thought very well, requiring the latter to elucidate the former.
Yes, you’ve reminded me that John Lawton had an old View from the Park column on this, in which he borrowed an acronym from Mick Crawley for the latter school: the WIWACS (the “world is infinitely wonderful and complex school”). Also sometimes known as the “but in MY system” school. The same school that Robert MacArthur implicitly (but memorably) slated in Geographical Ecology when he defined science as the search for generalities (or something like that) and then said that his book was aimed at those wanting to do science.
And in evolution, one can see a similar division, between what someone once called the “lean and mean” school (folks like Haldane, Fisher, Hamilton, Maynard Smith, Dawkins, and Price, who favor simple, elegant models of the effects of natural selection), and the “let a thousand flowers bloom school” (exemplified by Gould, with an emphasis on the complex interplay of many evolutionary forces and unique historical events). Interestingly, both schools can probably legitimately claim Darwin as their founding father.
The division you identify runs somewhat orthogonally to some of the others we’ve been discussing. For instance, I have an old post about an essay of Jim Brown’s (written for NCEAS, soon after its founding) on the importance of searching for and explaining general patterns in ecology, and a response from Peter Kareiva. Kareiva’s response argues, not for the importance of the particularistic as opposed to the general, but for the importance of a focus on dynamics and processes (whether system-specific or general) as opposed to patterns. Kareiva also calls for a science of ecology capable of addressing whatever questions we might want or need to ask, rather than a science focused only on questions relating to general patterns (he of course thinks such a science will be process-oriented).
Whatever you think of the correctness of Kareiva’s position, I think it’s interesting because of the way it illustrates how different ways of dividing ecologists into “schools” can cross-cut and intermingle. A Venn diagram of ecology with all of the various “schools” so far proposed on this thread might be close to unreadable…
Jim Brown himself is of course an interesting example to raise in the context of this thread. Founding father of macroecology and all about focusing on big, general patterns–but also very much a field experimentalist. And a guy whose most famous and highly-cited work has to do with a highly reductionist, mechanistic model of the adaptive evolution of circulatory systems as an explanation for a “macroecological” pattern (3/4 power allometric scaling of metabolic rate). I’m sure Jim’s fine ecumenical example is why Morgan isn’t dogmatic about her own way of pursuing macroecology. 😉 As another example, Darwin himself wasn’t merely a well-traveled natural historian and keen observer–he was also an avid experimentalist (that’s what he spent most of the last 40 years of his life doing), and was even prepared to collaborate with mathematicians (as in the passage in the Origin on how honeybees can build hexagonal honeycomb arrays by following simple behavioral rules). E. O. Wilson is another example. He may have called his autobiography “Naturalist”–but Wilson is far more than just a naturalist. He also wrote a classic textbook on population biology that includes a lot of math, and he’s published a lot of modeling work, not least the theory of island biogeography. Not that every titan of ecology and evolution was a polymath of no particular school–but the examples of these polymaths are instructive, I think. May have to post on that sometime–“great polymaths in ecology and evolution”…
I guess I’ll stick my neck out there before I’ve read all the comments.
The most important grouping that I see is not very novel, but I still think it’s the most important one: the difference between modelers/theoreticians versus empiricists. They often look at the world fundamentally differently and with a degree of distrust. Not saying anything new, I realize, but then novelty’s not really my gig.
Yeah, I know what you mean Jim, but I’d suggest that’s more of a continuum than distinct schools. There are too many people who do both (like me) for them to be distinct schools, I think.
OK, seems a fair point Jeremy. I wondered myself how distinct that dichotomy really is as I was typing that
I do think the ecosystem ecology ideas that the Odums came up with were a fundamentally different way of looking at ecological systems. I mean “emergy” as an ecological currency, I gotta believe a lot of folks said WTF? to that some of his other ideas at the time, or even still. Surely there’s a huge difference in looking at the ecological world largely as collections of individuals and populations whose dynamics are all about birth rates, reproductive patterns, longevities, death rates etc., versus as conduits for the transformation of energy and geo-chemicals. Very different senses of spatio-temporal scale and constraints on the system and what dynamics and constraints are the important ones.
But then, once again, I’m hardly stopping the presses with such an observation.
Interesting and focusing question. Are ecosystem vs population/community ecology just different subdisciplines studying different things or are they schools of thought. Can I get away with saying sometimes one, sometimes the other? As I think about it, I think to find a case where they’re world views/schools of thought, one would have to find a scenario where they tackle the same question differently. I’m actually having a hard time coming up with one. For example, if one asks “what are the effects of climate change on the biosphere” you will get very different answer one about rates of decay and nutrient cycling and one about biodiversity loss. But these are really just different questions hinging on different emphases on the interpretation of what is the biosphere. Can anybody think of a truly common question where it is just the approach that is different?
The one that comes most immediately to my mind Brian, is the work by Borman and Likens at Hubbard Brook 40 years ago, trying to understand multi-annual/decadal forest demographic/community change, but not just through strictly population processes divorced from watershed bio-geochemical, micro/meso-climatic, and biophysical dynamics. And similar IGBP research programs as well, such as at Coweeta and elsewhere.
Although, I’m not quite sure that suits what you’re describing or looking for.
Thanks Brian for the interesting post and thanks to the rest of the commentators for the (as always, lively) debate. Reading through the comments suggests that the answer to your question is: not really. Or at least, not in the same way as in economics. While there have been many interesting “schools” proposed, I think most people would agree that things are probably somewhere in the middle. I don’t think many readers of the blog (or at least the commentators) would identify themselves as belonging to any of the schools in the same way that Krugman identifies himself as a Neo-Keynesian. At least in my field (evolutionary biology) there have in the past been intense debates between schools of thought where people strongly identified with some approach (admittedly some are still going on–group selection versus kin selection folks and, at least the remnants of the parsimony-likelihood debates are still there) but I think most people have generally moved beyond these to a more pluralisitic approach to evolutionary biology. The same appears to be the case with ecology. Perhaps I am naive, but I would be surprised if for example, someone came on the comments section and claimed “I am a top-down ecologist” or whatever. The world is complex and i think most people are willing to ackowledge the advantages of various approaches and are not too dogmatic about which “school” they belong to.
You mentioned group vs kin selection. It reminds me of a seminar when I was at U Arizona. David Sloan Wilson gave a scathing talk about how badly and ridiculously group selectionists had been treated. During the questions a junior faculty member raised her hand and said “I don’t see what the big fuss is – everybody I know in evolution agrees sometimes group selection happens and sometimes it doesn’t” (which is my read too). But then we see the giant Nowak & Wilson Nature slam-down of group-selection in 2010 and I wonder. Is this a case where Kuhn is right – old scientists have to die for debates to leave? One generations battle royale is the next generations “can I have a bit of both please?”
I will put a bit of a stake in the ground and claim that the “population” school in ecology was every bit as strong as some of the schools you mention in evolution. In the 1970s some of the most famous ecologists actively fostered the view that population processes was the only place where you could find rigorous science in ecology. People like EO Wilson and MacArthur. It had this nice logic – # of individuals changes only through birth, death, emigration and immigration. And fitness unifies population growth and evolution. This led to all sorts of offshoots like the founding of the Population Biology graduate group being founded at UC Davis which probably still to this day the most prestigious of the many different ecology programs there.
A personal anecdote. The very first ever full 1 hour seminar I gave as postdoc after my dissertation defense was on my macroecological work. At the end a famous name in evolution and also in ecology (and not coincidentally trained and came of academic age in the 1970s) stood up and said “I was taught that anything that doesn’t involve population processes is not good science”. He was trying to be nice and hence his passive-voice phrasing. But in the end his judgement was so absolute and so dismissive of my work I’ve never forgotten it.
So I submit that the “population” school is alive and well. Although I think more tolerance and diversity is found in ecologists trained in the 90s and 00s. Which raises an interesting question. What are the schools prevailing now that 20 years from now our students will be waiting for us to hurry up and retire so they can move on with in a more realistic, more balanced view. Or are we so post-modern that we don’t do this polarization any more?
A colleague from China and I were talking about ecological schools of thought and she stunned me by saying when she learned ecology, it was taught as “American Ecology”, “European Ecology”, “British Ecology”, and so on. Her understanding of schools of thought was at a very different scale (and coming from a unique angle) compared to what was discussed above, but something I found valuable as it revealed another blind spot in my biases as an ecologist.
I didn’t realize that the reach of the “Silwood Mafia” was such that it was taught in China. 😉
Sorry, insider joke. Silwood Park in the UK, where I was a postdoc, was a longtime stronghold of the sort of process-oriented ecology I do, and was connected with some of the most prominent British ecologists of the last few decades (Bob May, John Lawton, Charles Godfray, Mick Crawley, Mike Hassell…). That crew was sometimes referred to by folks not associated with Silwood as the Silwood Mafia, for various reasons I might get into if you ever happen to run into me in a pub… 😉
You don’t carry that axe with you into pubs do you?
Or is it only by the T-shirt and axe that we would have any chance of recognizing you?
For mystified readers: Jim is referring to this picture of yours truly.
I totally understand what this person means. It certainly exists in macroecology. Even the definition of what macroecology studies differs (American’s have a much narrower view tending to involve comparative analysis across species and Jim Brown’s book specifically sets aside biogeography and paleoecology as distinct – albeit allied – fields, while Brits tend to expand macroecology to clearly include at least the ecological parts of biogeography).
A quick perusal of citation patterns shows that people tend to cite much more often within their own “geographic school” than outside in many areas of ecology.
One that just occurred to me is that the Europeans have a fairly different philosophy regarding how to sample and describe plant communities than North Americans do. They seem to be much more geared around static typologies than we are, and to have almost no regard for the idea of random or systematic sampling. Presumably this reflects the general lack of truly wild vegetation and more stable land use patterns in Europe compared to here, where things are more homogeneous and constant. However, I’ve found similar approaches to large surveys in N America as well, such as a massive vegetation mapping and typing project in California in the 1930s. That thing is a statistical mess frankly.
Man that was entirely confusing. The phrase “where things are more homogeneous and constant” was meant to refer to Europe, not N America.
It just occurred to me that I’m as interested in the question of whether there *should be* different schools of thought for particular topical questions in ecology, as whether there actually are such. In other words, are there “right” and “wrong” (to make something of an artificially black and white issue out of it) ways of addressing various different questions. And if there are, then why isn’t everyone following them?
Good question! In some ways, this gets back to the issue of opposing schools of thought, like density dependence vs. independence (different answers to the same question) vs. different schools of thought (interested in different questions, though perhaps also prepared to argue about whose questions are the “best” ones).
Definitely an important question. The way I see it, anytime you are at the frontiers of human knowledge (i.e there is uncertainty and room for difference of opinion), then human nature will organize itself into schools of thought. In short this is really a question about human nature and the fact that scientists are human. I get very annoyed sometimes with my humanist colleagues who carry this to far and imply there is no objective reality that emerges out of science. But I do agree strongly with their basic point that we ignore the fact that scientists are humans at our peril. For me the *should* part of the question is what should we do about it and here I’m squarely in Morgan’s school – the best thing to do is to flush it out in the open.
Pingback: Scaling up is hard to do | Dynamic Ecology
An interesting post and some great follow up comments. It’s been hinted at by a few folks, but I think it’s worth pointing out explicitly (given the macroecological flavour of much of the discussion) that different “schools” span different scales. We can measure scale here by size of field (number of scientists researching that area) or how “important” that field is considered (which may correlate with number of scientists!) rather than ecological level (biome/ecosystem/community/population).
The ferocity of the disagreements seems not to correlate with size of field or ecological level, however. I’m sure we can all think of small-scale “schools”. For example, some of the debates that have gone on in my own field of pollination ecology over the past 15 years or so have been centred on levels of specialisation versus generalisation in pollination systems, how one might measure specialisation, the relative importance of pollinators versus other factors in the evolution of floral phenotype, the predictive value of pollination syndromes, etc.
These debates have sometimes been intense and bad tempered (I have a good few personal anecdotes of my own…….) but they have largely been confined within the subject have not spilled over into adjacent fields such as biotic seed dispersal. Do we count such disagreements as “schools”? I pointed out in a paper a while ago that some of the disagreements seem to me to stem from the training of the scientists involved and whether they coming at pollination ecology from a background of botany, ecology, population genetics, zoology, or whatever. Perhaps pollination ecology is unusual in that respect (though I suspect not).
Thanks Jeff. As I think I noted in an early comment, I wouldn’t count disagreements over things like specialization of pollinators as “schools of thought”. It’s far too narrow. A school of thought should shape your views on pretty much every issue in some broad field. I think something like density dependence vs. independence qualifies, or something like adaptationism vs. whatever-you-call-what-Steve-Gould-believed.
To quote something you said above, Jeremy:
“Another candidate for schools of thought in ecology (concerning rather narrower issues than the ones we’ve been discussing): Tilman vs. Grime on how to think about competition.”
This is an important debate for plant ecologists but less so for animal ecologists. So how narrow does it have to be before it it ceases to be a “school of thought”? Hence my comment about scale. The debates in pollination ecology have some important implications for consideration of macroecological patterns of species interactions, such as whether tropical taxa are more specialised than their temperate counterparts. see this recent paper by Schleuning et al. in Current Biology, for example: http://www.cell.com/current-biology/abstract/S0960-9822(12)00943-8
Fair point Jeff, although I actually think that Tilman vs. Grime is about much more than just how to think about competition. That debate, I think, is merely one symptom of deeper conceptual disagreements, and I think it’s those deeper conceptual disagreements that really divide those schools. For instance, people in Grime’s school of how to think about ecology believe in zombie ideas about the IDH. People in Tilman’s school don’t. The two schools have opposing views on biodiversity and ecosystem function. All those
Their disagreements aren’t merely, or even primarily, empirical–they’re conceptual. They’re disagreements about what sort of questions one should ask, what sort of studies one should conduct to answer those questions, and what counts as an answer to those questions.
Would you characterize the debates in pollination ecology to which you refer as ultimately conceptual, or basically just empirical?
They’re a mixture of both conceptual and empirical, Jeremy. The main conceptual disagreement is about whether the traditional pollination syndromes (which have been used to categorise plants and acted as a conceptual framework for the field for about 150 years) are still fit for purpose. There’s a school of thought that says they are, that we can accurately predict pollinators from floral form most of the time. There’s another (which includes myself) that is of the opinion that the traditional syndromes are far too broad and imprecise, and based on limited taxonomic and geographic distributions, to be robust for current use. A paper we published in 2009 concluded that the pollinators of about 30% of the c. 350.000 species of flowering plants can be predicted from the traditional syndromes. I would suggest that this is too low to make the syndromes generally usable. Others disagree.
Related to this and in part empirical is how we define “specialised”. Is a plant that is pollinated by 30 different species of medium sized bees a specialist or a generalist? On one level it’s specialised (all pollinators belong to a single taxonomic/functional group). On another it’s generalised – any of 30 species can act as a pollinator. We have introduced more nuanced terms (functional versus ecological specialisation, for example) but the field is far from in agreement as to their use.
Pingback: Happy Birthday to us! | Dynamic Ecology
Pingback: Lost causes in science | Dynamic Ecology
Pingback: Modelling cyclic populations: thoughts on the workshop | Dynamic Ecology
Pingback: Ask us anything: how do you critique the published literature without looking like a jerk? | Dynamic Ecology
I wouldn’t say having contrary ideas necessarily means being in different schools. Meg’s example about freshwater ecology sounds like my idea of different schools. I’m not so in ecology scene but from broad evolutionary biology I could eg. say that evolutionary developmental biology and population genetics are two schools of thought which are very much contrary to their research methods and conceptual core, thus they would give very different answers to how evolution works. (Could it also be partly Gouldian vs. Dawkinsian dichotomy – might be.) In systematics, cladistics à la Hennig seems to be very strong school of thought, not in numbers but rather in their conviction.
(Especially when speaking about evolution, the difference normally is what people mean by saying ‘gene’. But that don’t always mean different schools of thought.)
BTW, it seems that the ones who are in minority school of thought normally have a better understanding of philosophical and conceptual differences as they are working against the main school.
Pingback: On progress in ecology | Dynamic Ecology
Pingback: Is ecology a single coherent scientific discipline? (includes poll) | Dynamic Ecology
What about the divide between modern hypothesis-testing large-data studies with complicated analyses-school versus the old-school observational ecology? As John G. T. Anderson so precicely described here: https://www.americanscientist.org/article/why-ecology-needs-natural-history
I am old-scool and get very frustrated on the strong focus on elegant statistics and maybe reduced focus on the relevance of findings. Who wants to join a club of old-school ecologists to work for the resurection of simple-statistics empirical fieldwork ecology?
Relevant old post: https://dynamicecology.wordpress.com/2014/01/28/stats-vs-scouts-polls-vs-pundits-and-ecology-vs-natural-history/
tl;dr: the statement that natural history (or observational work, or field work, or exploratory work, or empirical work…) is undervalued in ecology needs a lot of unpacking. I think it’s better to focus on narrower and more precise claims, because they lend themselves better to objective evaluation and to practical solutions.
Pingback: Understanding that, why or how in ecology? • International Initiative for Theoretical Ecology