Do ecologists do enough research to resolve apparent contradictions? (UPDATED)

A while back, a correspondent noted that many important advances in physics arose from apparent contradictions between established bodies of knowledge. If you’re confident that X and Y are both true, but X and Y appear to contradict one another, well, that’s a puzzle that demands resolution. And the resolution often is a deep insight into X, Y, and/or the relationship between them. My correspondent suggested that this isn’t unique to physics, that identifying and resolving apparent contradictions is a good way to advance any scientific field.

I think there’s something to this. I’m currently revising a paper I’m very proud of (we’ll see what the reviewers think!) The genesis of the paper was me recognizing what seemed like a contradiction between two things I thought I knew about metapopulation dynamics. Resolving that contradiction led me to what I think is a deep insight about how metapopulations persist.

My correspondent suggested that ecologists do relatively little research based on resolving apparent contradictions. I think that’s right, though I don’t have any data.

Assuming for the sake of argument that’s right, why is that? Is it because ecologists’ ideas about the world all are mutually compatible? And if ecologists’ ideas about the world are all mutually compatible, is that to ecology’s credit or discredit? If ecologists are unable to do contradiction-resolving research because their field seems not to contain many incompatible claims, well, maybe that’s a sign that ecology’s claims are too vague?

What do you think?

UPDATE: Our commenters always come through. See the very first comment for two excellent examples of apparent contractions in ecology and evolution–Reid’s paradox and the paradox of stasis. Also see the comments for discussion of why the paradox of the plankton and the paradox of enrichment aren’t really “paradoxes” in the sense I intended in the post.

9 thoughts on “Do ecologists do enough research to resolve apparent contradictions? (UPDATED)

  1. I think there’s certainly some truth to the thought that many ideas are mutually compatible (the “multi-causal world” mentioned not infrequently on the blog), so questions in those cases come to be about relative importance (my favourite process is more important than yours). I also think a decent bit of research has, in fact, been motivated by contradictions, each one typically described as a “paradox”:

    – Paradox of the plankton: The competitive exclusion principle says one species should win it all, but communities contain many seemingly very similar species.
    – Reid’s paradox: Well-parameterized models predict potential plant migration rates far lower than what must have occurred since glaciers started melting 20,000 years ago.
    – Paradox of stasis: most often, when someone studies a given trait in a given species, it appears to be under directional selection; but the same species has looked more or less the same for a great many generations.
    – The neutral paradox (I just made that term up): neutral models reproduce some community properties amazingly accurately, but individuals of different species are not ecologically equivalent.

    The resolution is rarely one single, silver bullet, but I do think resolution happens to some degree. I’ll be curious to know about other paradoxes (the paradox of enrichment is one other).

    (looking forward to the metapopulation paper!)

    • Good comments.

      The paradox of the plankton isn’t the sort of “paradox” I was thinking of. A mismatch between a theoretical limiting case and reality can be an interesting starting point for research. But I wouldn’t call it an apparent contradiction or even a paradox. Because at a broad-brush level, the resolution is completely obvious: reality violates one or more of the assumptions on with the theoretical limiting case is based.

      The neutral paradox is the flip side of the paradox of the plankton–a match (rather than a mismatch) between some predictions of a theoretical limiting case and reality. That one was resolved pretty quickly (though I think it should’ve been resolved even quicker): The features of reality that neutral models correctly predict are the ones that it’s hard *not* to correctly predict. That is, many different models making different assumptions all predict more or less the same thing. Again, not exactly the sort of paradox I was thinking of, though I suppose your mileage may vary on that, depending on how obvious the resolution of the paradox was to you. Which raises the point that many paradoxes are kind of in the eye of the beholder. What’s “paradoxical” to one person can be quite obvious to another person. I think the best sorts of paradoxes are the ones that seem paradoxical to pretty much everybody. Like the incompatibility of quantum mechanics and general relativity.

      Reid’s paradox and the paradox of stasis are just the sort of thing I was thinking of. What’s the resolution of Reid’s paradox (I’ve heard of it, but not of the resolution)? As I recall, the resolution of the paradox of stasis is some combination of fluctuating selection and what Fisher called “deterioration of the environment”, right?

      The paradox of enrichment isn’t really a paradox or contradiction in the sense of this post, or even in the sense of the paradox of the plankton and the neutral paradox. It’s only a “paradox” in the looser sense of “Enriching this predator-prey model would lead to high-amplitude cycles, which would put both species at increased risk of extinction in the real world. So more resources equals greater extinction risk. Huh, that’s kind of paradoxical, you’d think more resources would always be a good thing.”

      “(looking forward to the metapopulation paper!)”

      Me too. 🙂 That one’s kind of interesting example for purposes of the post, because it resolves an apparently contradiction between two *theoretical* claims, as opposed to two empirical claims (as with Reid’s paradox and the paradox of stasis).

      • Not sure about the paradox of stasis, although Njal Rollinson gave a great talk on it at CSEE last year, specifically about offspring body size (larger always favoured by selection, but huge tradeoff with offspring number). For Reid’s paradox, my understanding is that there are two components to the resolution:

        (1) Rare, long-distance seed dispersal (in vertebrate guts, or updraft winds above the canopy) – not accounted for in early models – makes a massive difference, and brings model and data in line.
        (2) Some (many?) species had populations closer to glacier edges than we thought, so the initial estimates of migration rates by Margaret Davis and others based on paleo data were too high. (I don’t think this one does the job completely, but it helps bring prediction and observation closer). Genetic evidence from Jason McLachlan and others show this.

  2. Just remembered an old post in which I discussed an interesting example from physics of fundamental insight (and practical advances) arising from resolution of an apparent contradiction:

    Though as noted in the post, this is one of those cases where the “contradiction” is somewhat in the eye of the beholder. It’s more a case of “what seemed like a contradiction to some people but not to others led to insight as the latter group of people explained to the former group why they (the former group) were confused. The latter group then realized that their explanation had interesting implications and applications.”

  3. One contradiction that our lab works on has to do due with predator-prey systems and cycling. Theory tells us that it is quite easy to get cycles in predator-prey systems, yet only about 30% of predators and prey exhibit cycles. I have heard some people argue the opposite, that cycles should be uncommon yet occur in about 30% of predator-prey time series. Either way you look at it, there is some contradiction between theory and our data on time series.

    • Hmm…not sure I’d say that theory predicts that cycles are “easy” to get. I’d say theory identifies the circumstances in which you get cycles, but I wouldn’t say it predicts how common those circumstances are.

      Also, 30% is probably an upper bound on the proportion of natural populations that cycle, because probably at least some of the populations in the GPDD compilation were published *because* they cycle, or were chosen for study because they were thought likely to cycle.

      • “Easy” probably wasn’t a good way of putting it. I agree with you that theory predicts the circumstances rather than commonality of cycles. Perhaps this is a more specific and better characterization of what we are looking at; 1) saturating predator functional responses should be common in nature, 2) saturation is destabilizing in predator-prey systems and can cause cycles, 3) why do most predator-prey systems appear to be stable?

        I also agree that 30% is likely an upper bound. I’m sure that many predator-prey time series that went into the GPDD were collected because they were interesting, e.g. show predator-prey cycles, single generation cycles, etc.

  4. I think part of the problem is that, in ecology, the contradictions are seldom between rigorously-defined theoretical models, and instead based on verbal arguments. I sometimes warn students away from working on topics perceived as ‘controversial’ because there’s probably a good reason no-one has resolved them. For example, whether species in communities should be more or less related than expected by chance. Another vote cast for either side is not going to answer the question, even though studies continue to pile in and claim that they have provided ‘support’ for one or the other. Better to step back and ask whether people are asking the right question, at the right scale, or with the right mathematical framework. But that’s hard to do, and less fun than running out for more data.

    • Yup. I should probably add this to my old post on “weak reasons for choosing a research project”. On it’s own, “X is controversial” is a weak reason to study X, because as you say maybe the reason X is controversial is that nobody really knows how to study it. And yes, if you think you’re going to contribute even in a small way to resolving the controversy over X by conducting one more study of the same sort that others have already conducted, you’re just wasting your time.

Leave a Comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s