What is wrong with theoretical ecology?

The status of theory within ecology is still uncertain. While it is no longer fashionable to declare ecological theory completely useless, many ecologist still harbour uneasy feelings toward it. They must have a point. We theoreticians should not just say that field ecologists don’t understand maths. Theoretical ecology, as they see it, does not motivate them to learn it.

Theoretical ecology is generally considered as a collection of independent models. Usually, questions about relationships between the different models are not even asked. They are just different models with different assumptions, so the results are different, as well. It is a very rare situation, that one of the possible models matches reality in a very convincing way and allows us to make reliable predictions. While the so called “strategic” models [1] are supposedly wide in scope, but far from empirical details, the “tactical” ones may describe a specific situation, but fail to provide wider insight and predictive power. Neither of them build a connection between general concepts and specific field situations. Fundamental issues of ecology remain unresolved for decades, as one can support any proposal with models.

Many theoretical ecologists are excellent mathematicians and use their proficiency in a subfield of ecology. However, their results are not synthetized into the general culture of ecology. Ecology is still a verbal science; the subfields of, and the questions for, theoretical ecology are defined by the verbal (and often confusing) discourse. Current theoretical ecology books are usually collections of chapters written independently by different authors [2,3]. (Ted Case’s “Illustrated guide” is the beautiful exception [4].) Chapters of Scheiner & Willig’s book [5] are also written by different authors, but not independently: they had to obey a common structure. Still, all chapters are verbal summaries of models and the chapters are related only through that structure.

We should do better. We need to have a different mind-set. It is not enough to teach and learn the very few most elementary models by themselves. We have to teach how to build a model and how to introduce any complications we wish to include with a reason. This way we develop the feeling that models describe reality – with the level of elaboration we want. Of course, fidelity of the model will depend on our detailed knowledge on the real thing – but this is the empirical side of the issue. This kind of familiarity with theory building allows us to assess whether the conclusion depends on this-or-that assumption/simplification, or not. We can develop a sense of related models. Model B can be a more specific version of Model A, so any conclusion from A automatically applies to B, independently of the added specifics. Or, B is an approximation of A, valid in a limit, etc. This way we can reach conclusions at very different levels. Some of them are very general and robust, others are more restricted in scope.

My personal motivation comes from the adaptive dynamics culture. Adaptive dynamics was to a large extent motivated by the theory of structured populations. This theory is the prime example for a high-level ecological framework theory. The point is, that all complications of the life history of an organism contribute to the fitness in a mathematically well controlled way – which must be the starting point for a theory of evolution.

Note, that it is very rare even in physics that we accept a model just because it fits data very convincingly. Sure, finding Neptune at the predicted place was a moment. Quantum physics became an established science when a single numerical prediction of quantum electrodynamics matched measurement with precision of 10 digits. Such things are rare and always related to the simplest possible situation. In the more typical case, we have a more-or-less accepted scientific framework – mostly learned from the simple cases and tested by many prior applications. We ask questions within that framework. The first check of a new theory/model is whether it is consistent with the already established knowledge. If not, then it is wrong almost surely. (Well, rarely, we have to replace the framework.) The established knowledge has a structure. There are very general laws, and more specific ones, the latter ones have to be consistent with the former ones. Then, we have even more specific understandings on different classes of systems. All of this prior knowledge informs us when developing very specific assumptions and models about the specific system of interest.

While ecology is very different from physics, it also needs a kind of structured theoretical framework. The goal of ecology should be the fundamental understanding of ecosystems [6], instead of just to reproduce (“explain”, see Axel’s recent post) some observed correlations. For this purpose, we need a theoretical approach in line with that goal of fundamental understanding. We know, how to model the simplest ecological situations. It works, and theoretical ecology should be built on this experience, instead of trying to describe (understand?) the more complicated systems with arbitrary models and (often vague) theories. A few of us have written a book that may illustrate our hope for an ecology based on a consistent theoretical framework [7].

[1] Czárán, T.: Spatiotemporal models of population and community dynamics. Chapman & Hall, 1998
[2] Hastings & Gross (Eds.): Encyclopedia of theoretical ecology. University of California Press, 2012
[3] May & McLean (Eds.): Theoretical Ecology: Principles and Applications. Oxford University Press, 2007
[4] Case, T. J.: An illustrated guide to theoretical ecology. Oxford University Press, 2000
[5] Scheiner & Willig: The theory of ecology. The University of Chicago Press, 2011
[6] Courchamp, Dunne, Le Maho, May, Thébaud & Hochberg: Fundamental ecology is fundamental. TREE 30: 9-16, 2015
[7] Pásztor, Botta-Dukát, Magyar, Czárán & Meszéna: Theory-based ecology: a Darwinian approach, Oxford University Press, 2016

Leave a Comment

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