Last month, I spent a couple of days in Oxford with a group of paleoecologists of many nationalities, timescales, and taxonomic foci, as we frantically narrowed down a list of more than nine hundred crowd-sourced questions to fifty. Our mission: to determine the most pressing, five-year-horizon-scanning questions in the field of paleoecology (or palaeoecology, as my UK colleagues spell it). Many cups of tea and pints of beer were sacrificed to the gods of Paleo in our efforts.
How did it work?
The process was fascinating, often brutal, always challenging, and — I think– really, really revealing. It was also pretty fun. Palaeo50, led by a team at Oxford including Ambroise Bakker, Anson Mackay, and Alistair Seddon, was inspired somewhat by a couple of recent efforts led by William Sutherland on the identification of 100 ecological questions of high policy relevance in the UK, and 100 fundamental ecology questions. The Palaeo50 organizers began by soliciting questions from the broader paleoecological community (here’s my earlier post on that process), and folks interested in participating in the actual workshop were able to apply as part of that process. I was the only member of my graduate lab at UW Madison able to attend, but as you can see from the Williams Lab website, the lab group was heavily involved in coming up with questions.
Conference attendees (of which I believe were around 60) were emailed the questions in advance, and asked to narrow them down to each of our own individual top fifty, as well as rank which subgroups we were most interested in– I ended up in Biodiversity Through Time. Every subgroup had a scribe (to record information about which questions were particularly contentious, or when concerns or points were raised), a chair, and a co-chair (for organizational and time-keeping purposes). Each subgroup was given dozens of questions, organized into loose themes, that we had to narrow down to twenty in the first day. This process was much more complex that it initially sounds– after an initial round of voting, there was a considerable amount of discussion, word-smithing, and merging of questions. Overnight, the 20 questions from each group were sent to everyone who submitted a question but couldn’t attend, to ensure that folks from other regions were able to participate as much as possible given the constraints of time and labor. On the second day, we then reduced those initial 20 questions to 7, and submitted 5 “contenders” from each subgroup for plenary voting to determine the final few contended questions as a group.
I really liked the organizational structure of the workshop; the process was very efficient, if you consider that we actually did go from >900 questions to 50 in 48 hours. Every effort was made to be as democratic as possible, and this was truly a crowd-sourced effort in that any paleoecologist with internet access was able to participate if they so chose. There was also quite a lot of tweeting, as folks recorded particular questions of interest or noted interested debates (check out this awesome Storify of the tweets here). I was the Scribe for my group, which was both really challenging (not only to keep up and actively participate, but to make sure my notes were as neutral as possible, recording important conversations and not just my own pet battles), but also very rewarding.
Some thoughts on the process
There were a few interesting outcomes of the workshop process that are worth noting. First, there were a lot of paleoecologists from a range of different timescales represented, which is a great thing– as a Quaternarist, I had been thinking of these questions mostly in terms of my own field, and certainly a lot of the questions submitted reflect that particular background as well. However, there were a number of deep-time paleoecologists in my group (think millions of years), and one thing we occasionally struggled with were questions that focused on a particular timescale. For example, one question — Are plant extinctions during the Quaternary as rare as they appear? — that was very highly ranked in the initial voting ended up not making it into the final fifty, in large part because I think participants were reluctant to elevate any particular time frame as important (remember, each subgroup only gets 7 questions, ultimately!). Interestingly, folks who worked on evolutionary timescales didn’t necessarily think processes of extinction during the Quaternary were particularly useful frameworks, because the Quaternary is only 2.5 million years long, and essentially represents one datapoint– not very useful in comparing rates of extinction to a background rate. Partly as a consequence of this and some of the other issues I get into below, no extinction questions made it in to the final 50. I found these kinds of differences really revealing about our respective fields, especially in terms of making sure that the why of our science is effectively communicated even to folks that are doing similar kinds of work.
Similarly, I noticed that few questions that focused on a particular taxonomic group or proxy made it through. More than 100 pollen questions were submitted, for example, but most were very specific, and none ultimately made the cut. Similarly, while there were a number of questions on the Pleistocene megafauna, none ultimately made it into the final seven in my group (though a fire/herbivory question did survive in the ecosystem group). You could argue that on the one hand, such a large proportion of the initial question submissions might warrant a mandate to include at least one question from a set of topics– the organizers could do some kind of pre-screening and determining that at least one question from key, high-interest fields was included. On the other hand, the process was ultimately democratic, and only fifty questions could remain at the end of the workshop. Painful cuts had to be made, and questions had to be merged and in some cases generalized. On some level, the really interesting questions were the specific ones, and so some information was lost in this process. Ultimately, though, getting bogged down in the specifics can blind you to the broader relevance of the process/system/method/region you’re interested in (certainly, this comes up again and again as I prepare my first grant proposal!).
There were also some surprising moments where it felt like group members were talking past one another a little; I’m used to this in other contexts, but wouldn’t have expected it in an all-paleo meeting. One interesting example came when the deep-time folks pushed back against the term “range shift,” which was advocated by the Quaternary paleoecologists, particularly because it’s commonly used in neo-ecology and the conservation biology literature. It turns out that in deep-time paleontology, where the fossil record can be even more incomplete than it is in my field, “range” and “distribution” have very specific meanings. It took the rest of us in my group a surprisingly long time to realize that this was the case, though, because in my fields, “range” and “distribution” are often used interchangeably!
One of the reasons I wanted to participate in the workshop was the fact that it was largely a European effort, and I was eager to interact with a new group of researchers. However, an unaviodable consequence of this was that there were a couple of questions that I think would have been elevated in importance in a North American workshop that didn’t make the final cut. Specifically, I think that the questions on 1) identifying beetle and pathogen outbreaks in lake sediment paleorecords, and 2) whether fire intensity (rather than just frequency) could be determined from charcoal records, are both of interest to North American paleoecologists, in large part because fire and beetle outbreaks are important ecological processes here in the US and Canada.
50 questions– and beyond
I mention the above not to be critical, but to capture interesting parts of the process (and questions!) that might otherwise be lost. I was only in one subgroup, so I can’t speak to what things were like in other subgroups, and I encourage folks from other groups to add their thoughts in the comments. I don’t honestly see a way around some of the points I brought up, and I don’t think the organizers couldn’t have done anything differently. I can’t share the 50 questions here, but I can tell you that I can’t imagine not including any of them, and ultimately many of the interesting questions that didn’t make the final cut are embedded within others that did (e.g., herbivory, fire, and beetles are all forms of disturbance). I did really like the fact that there was a specific methodological group, because it meant that some approach-based questions made it in– often with these exercises, it’s easy to focus on the big picture ideas, and ignore important advances in proxies and statistical methods. I also really liked the fact that we kept in some high-risk (but potentially high-reward) questions, like one related to neutral theory, another on ecosystem function and especially a question on trait-based methods (all areas I am keenly interested in, to reveal my particular biases!). These are questions that we might not yet have the tools to address, but will ultimately help advance the field.
The work that Alistair, Anson, Ambroise, and others at Oxford did was astounding– the workshop ran very smoothly, and was honestly a joy to participate in. I’ve already chatted with a couple of folks I met there about potential future project ideas. During the process of refining, winnowing, merging, and adjusting, I found myself writing down as many questions as we discarded. Many questions (like the one on plant extinctions during the Quaternary that I mentioned above) are still very cool and need answering, even though they didn’t make it into the final cut (stay tuned to see some in-progress work from me, in fact!). When a couple of very cool questions related to the Pleistocene megafauna extinctions were rejected*, I burst out in mock indignation that “I’ll study them anyway, and prove to you all that they’re interesting!” Joking aside, at times it was hard for me to let go and let the process work itself out in a democratic way– I sometimes felt the pressure of representing all of North American Quaternary paleoecology in a diverse room of scientists, and so at times advocated strongly for questions that ultimately didn’t make it. It can be hard to let go of your ego in those moments, and not feel as though people think your work isn’t meaningful merely because a given question doesn’t end up as one of the final fifty. At the end of the day, I’m proud and excited about the work we did, I stand behind our fifty questions, and I eagerly look forward to the final paper and the future work it inspires.
*How important were the Pleistocene megafauna as agents of dispersal? Did the loss of megafauna affect the abundance and distribution of other species via top-down trophic effects or via bottom-up changes in vegetation composition and structure?