Ask us anything: should ecologists devote more effort to framework-building and unification?

Every year we invite readers to ask us anything! Here’s today’s question, from Andrew Krause (paraphrased and summarized, click that last link for the original):

Are there particular ways forward that can help consolidate various different strands of work into unified conceptual frameworks, or is it best to focus on our own specialties and not worry so much about big picture things? Should more or less effort be given to framework building and unification?

From Jeremy:

Depends what you mean by “conceptual frameworks”. If by “conceptual frameworks” you mean “opinion/perspectives-type papers in which the only figure is a bunch of colored blobs with words in them, connected up by arrows or arranged in some 2-D space”, then ecologists spend too much effort developing them. I don’t think those sorts of “conceptual frameworks” have much value. They’re too vague.

If by “conceptual frameworks” you mean unifying mathematical frameworks that subsume a wide range of different special cases, then ecologists should put more effort into developing them. Those sorts of conceptual frameworks are super-useful and I think they’re an underused approach in ecology. Think modern coexistence theory, for instance. See here (scroll down to #4) and here. One important use of unifying mathematical frameworks is to allow everyone to work on their own specialties while also contributing to the big picture. Everybody can work on their own case studies, using whatever specialized, case-specific techniques they need to use–and yet we also know exactly how to compare and contrast among all those case studies, because they’re all covered by the same unifying framework. But depending on exactly what you’re trying to unify, unifying mathematical frameworks might not exist or might be hard to develop.

There are other things one might mean by “conceptual framework” or “unification”, of course (again, see this list). I have no strong opinions on whether ecologists should put more or less effort than they currently do into pursuing those other routes to “unification”.

More broadly, there certainly is a widely-shared (but not universal) sense among ecologists that their field isn’t a unified, coherent discipline. But one can argue that that’s a good thing on balance. That it’s on balance good for ecology as a whole to be done by ecologists with a diversity of interests, goals, and approaches.

From Brian:

In general I support more conceptual frameworks that unify diverse threads and I appear to be more open to different versions of conceptual frameworks than Jeremy puts forth.

The area of conceptual framework I have personally worked on the most is finding unifying principles under seemingly disparate work. So my paper Towards a Unification of Unified Theories is something I am really proud of. Several authors had worked to show that diverse biodiversity patterns like species area relationships, species abundance distributions and other patterns were all interrelated. That was a big step. But then different authors showed they were unified by methods as diverse as neutral dynamics, fractals, maximum entropy, stochastic geometry, sampling theory, etc. I argued that these diverse theories all really shared 3 underlying assumptions. But then I am highly biased towards synthetic work that pulls strands together.

I think just having a box and arrow diagram of causal influences can be helpful. Certainly its testable (do these arrows usually exist or not?). Having the list of forces that act on a system is no small thing. See this blog post on multicausality. I certainly think mathematical models are useful when they’re unifying diverse threads – mathematical models that dig down into the implications of ever more obscure details are not.

In principal verbal conceptual models can advance science, but so often either the terms or the logic are too vague and cause enormous confusion. Mark Vellend’s 4 forces of community ecology is a nice example where that didn’t happen and the verbal model did advance the field (of course that is partly because there is math underneath that prevented the logic from going too far astray, but the terms are still vague yet successful).

So to your last question of what should ecology do. Ecology is subject to immense centrifugal forces. Everybody works on a unique question in a unique system. I would come down unhesitatingly on the need for more effort to unify ecology and pretty much some form of framework is the only way I know to do that.


5 thoughts on “Ask us anything: should ecologists devote more effort to framework-building and unification?

  1. Jeremy’s response resonated with me. I have been surprised by recent conceptual papers in *certain* high-profile journals that try to be both predictive/insightful and broad. But often failing by omitting a mathematical framework, especially when they synthesize ideas built on well-developed quantitative theories.

    The 1990s-ish seem to have been a prominent era of mathematical works in high-profile ecology journals, followed by an (editorial?) emphasis that theoretical papers also integrate/connect to data. I liked that emphasis, but I sometimes wonder if the emphasis has swung too far, so that there are too few predominantly (say, 85%) mathematical content in high-profile ecology journals, especially with respect to reviews/syntheses. I know Jeremy recently did a nice synthesis on this where Am Nat stood out*, but curious what the “sweet spot” ratio might be for ecology.

    • I keep meaning to go through the literature and count up how many theory papers leading general ecology journals publish these days, compared to how many they used to publish. Did leading general ecology journals actually publish more theory back in the ’90s? It would be interesting to know for sure!

  2. Question: what are the most cited or most influential “synthetic” or “unifying” ecology papers from, say, the last 10 years? And what if you don’t count citations occurring in the first paragraph of the citing paper? You might want to exclude first paragraph citations so as to exclude “boilerplate” citations, as when a paper begins by stating the broad topic and then cites whichever high-profile recent papers happen to be about that topic. Boilerplate citations aren’t a sign of “influence”.

    Obviously, the answer’s going to depend a lot on exactly what you call “synthetic” or “unifying”…

    • It’s interesting to do a metric counting the number of “specialized” citations for papers – that is, how often the paper is cited in isolation (likely in reference to a specific point/idea) rather than as part of a 2-3 paper list. Of course would strongly follow overall #citations, so would be comparing residuals.
      Being a text mining whiz is a minor superpower I wish I had. If you set species=paper (or concept?) and abundance=citations, do Allee / Priority / seasonality [bandwagon] effects arise? 🙂

      • Being a textmining whiz definitely would be a blogging superpower! Man, there’s an endless supply of fun blog posts I could crank out if I knew how to textmine JSTOR.

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 )

Google photo

You are commenting using your Google 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 )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.