What sort of papers win the Mercer Award?

The deadline for submitting nominations for the Ecological Society of America’s various awards is Dec. 15. Details of the awards and how to submit nominations are here. Nominating someone for an award is a great way to honor deserving work and the people who did it, and to give a career boost to less-senior people. It’s also a way to shape the direction of the field, since the winning individuals and projects attract attention and influence. Anyone can submit a nomination and in my experience all nominations are taken seriously, even those from very junior people. For instance, back when I was a postdoc I nominated a Mercer Award winner.

Scientific awards also are interesting as a window into what sort of work is held in highest esteem by whoever is giving the award. So I thought it would be interesting to look back and see what sort of papers have won the ESA’s Mercer Award over the past 20 years.*

For reference, here’s a list of all the Mercer Award winners. You wouldn’t necessarily expect any common threads, since ecology a whole, the makeup of the awards committee, and the identities of those submitting nominations all change over time. But I do think many of the award winners over the last 20 years share some features. I emphasize that the following impressions are mostly based on memory or in a few cases a skim of the abstract, so take them with a grain of salt and please let me know of any mistakes in the comments.

• They’re really good papers! Which isn’t something we should take for granted. Think of how major awards in other fields (the Oscars, the Golden Glove awards in major league baseball) often go to bad picks. Or think of how you sometimes hear people say that everything published in leading journals (as Mercer Award winners invariably are) is oversold, trendy rubbish. Not so with the Mercer Award winners.
• Remember Meg’s post on the power of combining a diversity of approaches? Well, that appears to be the best way to win the Mercer award (and Meg should know since she won it!) That’s the common thread that jumps out at me: the winners are mostly papers that combine different approaches and lines of evidence. They often have extensive field observations documenting some pattern, plus experimental data testing alternative causal hypotheses about the processes generating that pattern. Many also have a mathematical model demonstrating that all the various kinds of data are in fact quantitatively consistent with one another, meaning that the “story” actually works as opposed to merely being plausible. Not uncommonly, those mathematical models are tailored to the system and partially or completely parameterized from independent data as opposed to being curve-fitted.
• There are a few exceptions from the last 20 years, papers that relied on one or two approaches. Jon Chase and Shahid Naeem won for mesocosm experiments. Lars Hedin won for a comparative observational study. Jean Richardson won for a phylogenetic comparative analysis. Dan Bolnick won for a review paper, which is kind of a category unto itself. And a couple of folks (Brian Enquist, Jordi Bascompte) won for papers developing a mathematical model and then comparing its predictions to observational data. In general, I think the list of Mercer Award winners reinforces the point that lots of approaches can lead to great ecology, but none are essential. Indeed, even “combining different approaches” isn’t absolutely essential.
• The winning papers usually have strong links to big, generally-applicable ideas, but also really nail what’s going on in some specific system. You don’t win the Mercer by developing some general theory and then waving your arms about how it kinda, sorta works if you squint at the data. Conversely, you don’t often win the Mercer merely by nailing what’s going on in one particular system.
• Winning papers usually cover all the bases–they’re mostly very convincing rather than merely being plausible or suggestive. For instance, they typically don’t just test some hypothesis or prediction. Usually, they also check whether the assumptions underpinning that hypothesis actually hold, thereby showing, rather than merely inferring, that their hypothesis holds for the right reasons. That’s why winning papers often combine different approaches: the kind of data you need to check an assumption often is totally different than the kind of data you need to test a prediction.
• You mostly don’t win the Mercer just for developing hypotheses. Nobody in the last 20 years has won with a pure modeling paper, for instance. Bolnick et al. is the closest anyone’s come as best I can tell–it was a review paper, but it won in large part because it suggested new ideas and productive lines of research for others to pursue. On the other hand, you don’t win the Mercer without having hypotheses either–nobody in the last 20 years has won for a purely descriptive study, or for discovering some intriguing phenomenon they can’t explain. You mostly win the Mercer for testing hypotheses, including hypotheses developed by others.**
• Nobody wins the Mercer by following a “recipe” that anyone can apply in their own system. Most of the Mercer Award winners are quite creative in terms of how they address the question asked. Nobody’s won the Mercer for, say, testing neutrality by fitting alternative models to species-abundance distributions, or for testing “habitat filtering” vs. “limiting similarity” by looking at how closely related co-occurring species are, or by plotting local vs. regional species richness to infer whether local communities are “saturated” with species (to pick three examples of popular “recipes” that many authors have followed in recent years). This is related to my old post on how techniques are less powerful than the people applying them, and to this old post on how “recipes” for inferring process from pattern hardly ever work.
• Most winners are single individuals, members of the same lab group, or small collaborations. Though I expect that to change at some point (at the risk of jinxing them, I think the NutNet project has a good shot at the Mercer one of these years).

*I also thought this would be an easy post to write, because it requires only slightly updating a comment I wrote on an old post.* You get the effort level you pay for on this blog. But I throw in all the laziness you want for free! 🙂

**Perhaps because when you test a hypothesis your paper reads as a complete “story”, even if a key part of the story–the hypothesis–was “written” by others. Whereas if you only develop a hypothesis, the “story” seems incomplete–it has a beginning, but not an ending. Of course, just because that’s the way things are doesn’t mean they should be that way. For instance, think of how the Nobel Prize in physics often is split between theoreticians who predicted something, and experimentalists who tested that prediction. Obviously, the Nobel Prize in physics is quite different than the Mercer Award in that it’s not usually given for a single paper. But even so, I think you can argue that the Mercer Award should sometimes go to papers that develop hypotheses but don’t test them, since it does sometimes go to papers that test hypotheses but don’t develop them.