Peter Abrams on ratio-dependent predation as a zombie idea

Peter Abrams has a paper in press at Biological Reviews criticizing the idea of ratio-dependent predation. Briefly, this is the idea that the feeding rate of predator individuals should be modeled as a function of the ratio of prey and predator densities, rather than as a function of prey density only, or as some other function of predator and prey densities. Peter has had a long-running, fundamental disagreement with the advocates of ratio-dependent predation, Roger Arditi and Lev Ginzburg. They laid out the points on which they agreed, and those on which they agreed to disagree, in a joint review paper (Abrams and Ginzburg 2000). But Arditi and Ginzburg’s 2012 book on ratio-dependence, to which Abrams’ paper is a response, seems to have revived the argument.

I don’t usually post on individual ecology papers. But I can hardly avoid it in this case, because Peter’s paper starts with this quote:

Ideas, once they take root, are hard to kill.…they
persist not just in spite of a single inconvenient fact, but
in spite of repeated theoretical refutations and whole
piles of contrary facts. They are not truly alive—because
they are not true—but neither are they dead. They are
undead. They are zombie ideas.

Yes, you read that right: Peter’s quoting my original blog post on zombie ideas in ecology. He uses it as a framing device for the entire paper, closing with another quote from the same post, and citing the post in the references. So while I won’t comment on the dispute over ratio-dependent predation*, I do want to comment on the roles of blogs and rhetoric here.

Obviously, it’s very flattering that someone I really respect would take a blog post of mine seriously enough to cite it. I hope that in future this will become more common. Blog posts aren’t substitutes or replacements for peer reviewed papers. Peer-reviewed papers have long been and will continue to be the most rigorous form of scholarly communication in science, and rightly so. But blog posts can be substantive scholarly contributions too, and so it’s appropriate to recognize and treat them as such–including by citing them.

Of course, it’s not a conventional citation–Peter’s citing me as a source for rhetoric rather than for, say, an empirical claim. I have mixed feelings about Peter’s rhetoric here.

On the one hand, I really do think that lots of good scientists can end up believing in things they shouldn’t, and that whole fields of scientific inquiry can get stuck and fail to progress for far too long, without the field even realizing that it’s stuck. That’s a serious problem when it happens. Recognizing the problem is the first step to fixing it, and to preventing it from happening in future. But such widespread problems often go unrecognized, precisely because they are widespread. The point of the “zombie ideas” rhetoric is to use a silly slogan to call attention to a serious problem that otherwise might go unrecognized. So I’m happy to see the “zombie ideas” meme spreading; hopefully it means people are worrying about this issue.

On the other hand, I disagree with Peter that ratio-dependent predation is a zombie idea, or if it is I’m not sure it’s one that should much worry those who disagree with it. It’s true that it’s an idea that’s persisted despite repeated criticisms.* But as far as I can tell it’s not a widely-held idea. According to Web of Knowledge, there hasn’t been a paper with “ratio-dependent” or variations thereon in the title published in Eco Letts, Ecology, Ecol Monogr, Am Nat, Oikos, JAE, Proc Roy Soc B, Oecologia, Ecological Applications, Journal of Applied Ecology, Journal of Ecology, Ecosphere, or any other empirical journal beginning with “Ecol” since 2006, except for one in 2012 in Ecosphere that had Roger Arditi as a co-author. As far as I can tell, most papers on ratio-dependent predation these days come from a relatively small number of theoreticians and appear in theory journals like Ecological Modelling and Ecological Complexity. Nor could I find any talks or posters on ratio-dependence in the programs for the last three ESA meetings (I didn’t look any further back). Yes, as Peter notes, papers advocating ratio-dependence get cited more often than critiques of the idea–but I suspect that’s because theoreticians interested in the mathematical properties of ratio-dependent models don’t see any reason to cite the critiques. And yes, ratio-dependence is mentioned in a few textbooks.** But on balance, rather than being a zombie idea, ratio-dependent predation looks to me like an idea that persists thanks to the efforts of a few dedicated champions.*** Who aren’t likely to be swayed by repetition of familiar criticisms of ratio-dependent predation (any more than critics of ratio-dependent predation are likely to be swayed by familiar arguments for it).

I also worry that the use of “zombie ideas” rhetoric runs the risk of making scientific debates overheated and personal. Criticism of ideas is hard to do without it getting personal, and rhetoric probably makes it even harder to keep things from getting personal. Not that I think hurt feelings are to be avoided or minimized at all costs–I don’t; criticism of ideas is too important. And so even though I’m sure rhetoric like “zombie ideas” has and will upset some people, I think it has its place, as an entertaining way to spark debate and call attention to an important “failure mode” of science as a whole. I’m reassured that others agree. But I don’t think its place is as an aid to criticism of individual scientists. Looking back, I wasn’t sufficiently clear about this in my early zombie ideas posts. I’m sure that to some readers it looked like I was personally criticizing individual scientists who hold “zombie ideas”, which wasn’t my intent at all.**** So while I’d like to see more debate about zombie ideas in ecology, I want that debate to be productive rather than unproductive, and debates that get personal tend to become unproductive.*****

I also worry a little that peer-reviewed papers aren’t the most natural home for zombie ideas rhetoric. Not long after writing my original zombie ideas post, I submitted it a lightly edited version of it to a peer-reviewed journal, zombie jokes and all. It got rejected; the referees and the editor didn’t think the rhetoric was appropriate for a peer-reviewed paper. At the time, I disagreed with them on this, while appreciating where they were coming from. But looking back, I now kind of agree. I think the version of that paper that I eventually published is a much stronger paper, and remains provocative without (hopefully!) being so provocative as to offend anyone or cause them to dismiss my substantive points out of hand. One role of blogs as a form of scholarly communication is to give a home to rhetoric and other stylistic flourishes that might be out of place in a scientific paper. Readers who might balk at certain writing styles in peer-reviewed papers are fine with them on a blog. Much as we don’t expect, say, scientific talks to necessarily be delivered in the same dry, formal style as a peer-reviewed paper. Formal Oxford-style debates can serve the same role–they’re a venue in which people are supposed to disagree as strongly and entertainingly as possible. But then again, I’m not sure of this. You can’t discover–or change–what presentation styles are acceptable to the audience except by trying out non-standard styles. Some of my favorite peer-reviewed papers use humor and rhetoric (e.g., Ellstrand 1983), and that humor and rhetoric is effective precisely because the reader isn’t expecting it.****** And there’s no pleasing everyone when it comes to presentation style, no matter what the venue (e.g., see the comments on this post), so there are always going to be tough stylistic judgement calls to be made.

In an old post, I suggested some rules of thumb about when rhetoric like “zombie ideas” is appropriate in scientific writing. I don’t have much to add to that post, except the point that rhetoric goes down easier with the audience in certain venues. But my thoughts are tentative (as I’m sure you can tell from the wishy-washy, but-on-the-other-hand nature of this post). The appropriate use of rhetoric and humor in scientific writing is something on which scientists have disagreed for centuries. It’s hard to give advice on the right thing to do when there’s disagreement on the right thing to do.

*I don’t think anyone has anything new to say on that; I certainly don’t. For instance, see Arditi and Ginzburg 2014, where they respond to Barraquand 2014 by referring readers to their book, and to Abrams and Ginzburg 2000, without elaboration.

**None of this “prevalence data” shows that ratio-dependence is flawed, of course. Or shows that it’s not flawed. Science isn’t a popularity contest.

***Ironically, I could imagine Roger Arditi and Lev Ginzburg referring to predator-prey models with prey-dependent functional responses as a zombie idea. I personally wouldn’t agree if they did so, since I think there are good empirical and theoretical reasons to study prey-dependent models. But if they did so, they’d at least be attaching the “zombie” label to an idea that’s widely-held. That both sides in the ratio-dependence debate can make a case that the other side is in the grip of a zombie idea illustrates the slipperiness of rhetoric, and the inability of rhetoric to function on its own as a substitute for substantive argument.

****It probably didn’t help that I stole the term “zombie ideas” from economists who are known for no-holds-barred rhetoric that includes attacks on the competence of their opponents, deployed in the hopes of causing others to stop taking their opponents seriously.

*****I’ve made this point before. Those who don’t care at all whether scientific debates get personal, or who think they should get personal, are making the mistake of writing for the audience they wish they had rather than the audience they actually have.

******Humor is really useful. Ellstrand (1983) basically makes the same point as Gould and Lewontin’s “Spandrels of San Marco”, but while I love Ellstrand’s paper I find “Spandrels” really annoying. I think that’s because Ellstrand’s style is funny and playful, while Gould and Lewontin’s is dead-serious. Ellstrand’s style is unexpected, but in a way that gets the reader’s attention without putting the reader off. Similarly, I think you can write about “zombie ideas” in a funny way, or a dead-serious way. Much as zombie movies can be funny and include themselves in the joke rather than being dead serious. Think Shaun of the Dead or Army of Darkness. I try to be funny about zombie ideas, but I leave it others to judge how successful I’ve been.