Book review: Community Ecology by Gary Mittelbach, and Community Ecology by Peter Morin

As a graduate student, I had the privilege of taking Peter Morin’s Community Dynamics course. In 1999, Peter published a textbook based on his course. It was very successful; as of 2011, Community Ecology is now in its second edition.

Gary Mittelbach also teaches a graduate course in community ecology. In 2012, he published a textbook based on his course, also called Community Ecology.

Originally, I had planned to review Gary’s book. Peter’s book is older, plus Peter was my Ph.D. supervisor, so I felt like reviewing Peter’s book would be some combination of not very useful and awkward. But I changed my mind, and decided it would be interesting and useful to write a joint review. The similarities and contrasts between the books provide an opportunity to talk about where community ecology is at right now and how it’s changing. And also an opportunity to talk about the difficult choices and trade-offs involved in writing a textbook (or designing a course). So this is going to be one of those book reviews where the author mostly uses the books as an excuse to talk about whatever he feels like talking about. 🙂

Let me first say that both books are very good. You can’t really go wrong with either one, I don’t think, for any purpose (you’re teaching a graduate course, you’re studying for your candidacy exam, you want a reference book…). So if all you were looking for was my opinion of both books, you can stop reading now. 🙂

They’re similar in some ways. Indeed, if you just quickly glanced at the tables of contents, you could be forgiven for thinking that the choice between the two books might as well be made with a coin flip. Here are the chapters from Peter’s book (not the actual chapter titles; just my brief summaries):

• History of community ecology; what communities are
• Competition theory
• Competition data
• Predation data
• Predation theory
• Food webs
• Mutualisms and facilitation
• Indirect effects
• Temporal dynamics and history (phenology, priority effects and alternative states, assembly history)
• Habitat selection
• Spatial dynamics
• Causes and consequences of diversity
• Succession
• Applied community ecology

And here are the chapters from Gary’s book (again, not the actual chapter titles, just my summaries):

• History of community ecology; what communities are
• Patterns of diversity
• Biodiversity and ecosystem function
• Population growth and density dependence
• Basic predation theory
• Adaptive behavior of predators and prey
• Competition theory
• Competition data
• Mutualisms and facilitation
• Food webs
• Trophic cascades
• Spatial dynamics
• Metacommunities and neutral theory
• Coexistence in variable environments
• Evolutionary approaches (rapid evolution, eco-evolutionary dynamics, phylogenetic community ecology)
• Open questions and future directions for community ecology

Lots of overlap there! And in some ways the topic coverage is even more similar than a glance at the tables of contents might suggest. For instance, Peter’s book covers density-dependent population growth, trophic cascades, and neutral theory, but within other chapters rather than in standalone chapters. Conversely, Gary’s book covers alternative states, but in the chapter on coexistence in fluctuating environments. The books also are similar in omitting some of the same topics. Neither has much coverage of diseases, for instance.

Both books also share a similar, powerful point of view on how to do good science, which is both explicit and implicit in both. Both books cite a wide range of different sorts of empirical work–laboratory microcosm experiments, field experiments, comparative field observations, etc. Both place a lot of emphasis on testing alternative hypotheses, and on the value of mathematical theory as a source of hypotheses, and so both books include lots of explication of theory. Gary’s book is a bit more explicit on his philosophy of theory testing (pp. 153-154). He talks about how the predictions have to follow logically from the theory’s assumptions (where have I read that before?), how to test theory, and what exactly you learn by doing so. I really liked that bit of Gary’s book.

There are differences between the books, of three interrelated sorts: organization, emphasis, and details.

One of the most important functions of a textbook is to give readers a road map to what would otherwise seem to be a trackless wilderness. As a student, it’s very difficult for you to develop your own road map of a field just by reading the primary literature and trying to figure out for yourself how the papers you’ve read are related. You can’t read enough to do it, and you’re likely to get confused by people using the same jargon to mean totally different things (“stability” is one infamous example). In contrast, Peter and Gary have spent their entire professional lives reading the literature and doing research (doing research matters here; nothing teaches you the jargon like having to use it yourself).

But the appearance of order created by any textbook is just that: an appearance. It’s an order the author imposes, not one the author discovers. Now, neither Peter or Gary imposes an idiosyncratic organization on community ecology, as evidenced by the many similarities between the two books. At some level, everyone working in a field will have similar “mental road maps” of that field. Similar, but not identical. There are some differences between Peter’s road map and Gary’s, and which one you prefer is going to be a matter of taste.

Gary’s road map is “pattern first”: he starts with large-scale patterns in biodiversity, and with the consequences of biodiversity for ecosystem function, and uses the need to explain these patterns and consequences as a motivation for subsequent discussion of processes like competition, predation, etc. Peter’s road map is “process first”: after the historical overview (which includes a cautionary note on the importance of critically evaluating hypotheses inspired by apparent patterns in one’s data*), he dives right into the processes. He then concludes the book with three chapters (the last three in my list above) on what he calls “large scale, integrative phenomena” that illustrate the joint consequences of those processes.

Obviously, either organization can work; they’re two different routes to the same endpoint. And I think the distinctions between “pattern first” and “process first” approaches are more important when it comes to doing ecology than when it comes to reading about ecology that’s already been done. There are smaller differences in organization, too. The two books group the same topics slightly differently, like two grocery stores that disagree on whether to shelve the oils with the baking goods or with the salad dressings.** But I don’t think the differences in organization are all that important.

As an aside, I’m not sure that either book is 100% successful at linking the processes that occupy most of the pages to the large-scale patterns and integrative phenomena. I think that reflects the limitations of the field of community ecology; it’s not a failing on the part of Peter or Gary. Research on large-scale patterns of biodiversity, and on biodiversity and ecosystem function, really is mostly disconnected from research on other topics in community ecology. Or, to put it in another, equivalent way so as to avoid offending anyone: research on other topics in community ecology really is mostly disconnected from research on large-scale patterns of biodiversity, and on biodiversity and ecosystem function. 🙂 We have lots of old posts discussing this disconnect with respect to large-scale patterns (here, here, here, here, here, here, and here; note that these posts are among my all-time favorites and many have great comment threads, so you should totally click through). Reflecting this disconnect, many of the topics in Gary’s first two chapters, and Peter’s last three, come off more as self-contained topics than as leading into or integrating the other material. Although Peter’s last chapter, on applied community ecology, is fairly successful as integration, I think (and while it’s a brief chapter, it’s notable because there’s no equivalent coverage of applications in Gary’s book, unless I missed it). Textbooks can’t avoid reflecting some of the limitations of the fields they cover. I do hope the next time someone writes a community ecology textbook, they’ll be able to write one that better integrates macroecology and “microecology”.

I think the biggest and most important differences between the books are in emphasis and details. As a very broad-brush generalization, I’d say that Peter’s book places more emphasis on the classics, by which I mean both classic topics, and classic studies of those topics. Gary’s book is stronger on coverage of recent “hot” topics, and places more emphasis on recent studies and the latest wrinkles. I emphasize that this is a really broad-brush generalization. Peter’s book certainly isn’t outdated in any way, and has lots of examples from the recent literature. Conversely, Gary’s book refers to many classic studies.

Obviously, there’s a trade-off between coverage of classic topics and coverage of currently “hot” topics. Trying to find the optimal point on that trade-off curve is really hard for textbook authors. It requires you to make judgement calls about what classic topics retain their interest and value. And it requires you to make judgement calls about what currently-hot topics will continue to be of interest years from now, and what currently-hot topics are just trendy bandwagons. I doubt any two ecologists would agree 100% on how to make these judgement calls (and Gary in particular writes at length about how difficult he found it to decide what topics to cover).

For what it’s worth, here are my purely personal opinions on some of the bigger differences of emphasis and detail. I like Peter’s decision to only give a couple of paragraphs and no figures to neutral theory better than Gary’s decision to devote half a chapter to it. I also like Peter’s decision to ignore phylogenetic community ecology better than Gary’s decision to devote part of a chapter to it. In making those decisions, I think Peter avoided enshrining some trendy bandwagons in his textbook. I also like Peter’s decision not to structure his whole discussion of spatial community ecology around the currently-popular metacommunity framework suggested by Leibold et al. (2004), the way Gary more or less does. On the other hand, I like Gary’s decision to talk about eco-evolutionary dynamics, which Peter doesn’t cover. And Gary’s book is much stronger than Peter’s on modern coexistence theory (a topic on which I have an old series of posts). I love that choice of Gary’s. I think that choice is going to look prescient, as modern coexistence theory becomes more widely understood and appreciated (something that’s happening already). Don’t misunderstand, Peter’s book does talk about modern coexistence theory, but does so more briefly, and more as one topic among many others. In contrast, Gary structures most of a chapter around the modern theory of coexistence in variable environments.***

Indeed, I kind of wish Gary had gone even further with his emphasis on modern coexistence theory, though if he had it would’ve been a very different sort of book. You wouldn’t realize it from Gary’s book, but modern coexistence theory isn’t just about coexistence in variable environments. It actually includes as special cases a lot of the stuff Gary talks about in other chapters–coexistence via resource partitioning, keystone predation, etc. Modern coexistence theory is about coexistence, period. I think it would be interesting for someone to write a community ecology textbook that’s structured quite differently than either Peter’s or Gary’s. One big chunk of the book (several chapters) would be on coexistence and diversity, with modern coexistence theory as the organizing principle.

In terms of other stuff, the two books are almost exactly the same length, and they’re similar in price. Gary’s book has color figures while Peter’s doesn’t, although in most cases the figures would’ve worked equally well in black and white, so I don’t think that’s a major difference.

Having trouble deciding which one to buy? Do what I did: buy both. 🙂

*Interestingly, Gary’s historical introduction includes the exact same cautionary tale: the failure of “Hutchinsonian body size ratios”, a failure which illustrates the impossibility of advancing community ecology solely with observational data and poorly-grounded hypotheses.

**What can I say, I’m a grocer’s son. 🙂

***And as part of that chapter, he says pretty much all the same things about the intermediate disturbance hypothesis that I’ve been saying on this blog (and that folks like Chesson and Huntly said in the primary literature years ago). Which made me happy. 🙂 Sadly, Gary’s book doesn’t include zombie jokes, though. 🙂