Scientific conferences, the too-slow movement of ideas, and giving an engaging talk


I went to a bunch of scientific conferences this summer. Four of ’em. I have a smorgasbord of reflections on the whole experience to share with you.

Yes, this is a lot of travel.

I spent about fifteen days padding around convention centers this summer. In early June there was Goldschmidt in Sacramento, then in July were both Social Insects and then Tropical Biology in Cairns, Australia. I recently got back from the Ecological Society of America meeting. If time served is a qualification, then I’m highly qualified to comment on the state of affairs in scientific conferences in my favorite disciplines (ecology, social insects, tropical biology).

The first thing I noticed — about all of these meetings — is that they’re full of white people like myself! So much whiteness. Europeans, and the more recent descendants of European colonists, numerically dominate scientific conferences.

Three of the conferences were crazy super-expensive. The only one that showed any attempt to keep costs down for students was the Ecology meeting. I’ve already complained about this so I’ll let that earlier post speak for itself.

Related to cost, most meetings were mostly comprised of older and more established scientists. Many students, postdocs and junior faculty members — whose research careers would get a bigger boost from being at the conference — were not there. I was one of very few who had undergrad members of the lab join at the meetings. The ecologists seem to the be only ones who are serious about both supporting students and growing diversity in the discipline. Some of lip service, but the Ecological Society of America is investing in the future more than other societies.

Another thing that I noticed about the meetings that, at each one, so many talks were about the same. exact. damn. idea. For example, I’m not joking when I estimate that about every other talk at the Tropical Biology meeting was about biodiversity along elevational gradients. The other meetings also had a huge amount of convergence on certain research ideas. Ecology was a bigger meeting, and it had more Big Ideas floating around on which people were hanging their work.

When there is a new Big Idea out there, it makes sense that people are working on it and presenting that work. But here’s what caught my eye about most of these Big Ideas at the meetings. These are ideas that have already played out. The biggest discoveries tied to those Big Ideas have already happened. Science is incremental, I realize (and embrace) — but what surprised me is the relatively scant ambition that most scientists have for selling new ideas. I’ve known this for a while, so this is not a revelation, but my season of conferences reinforced this fact. The enterprise of science is conservative. While there is supposedly a big market for transformational ideas, very few people are out there attempting to sell ideas like that. People are far more happy to present work about a theoretical bandwagon that already has a set destination.

It’s not that we don’t have big ideas. But, at a conference, most people have the opportunity to give one talk. In that talk, to keep our careers proceeding well we need to give a presentation sharing that we are actually productive and doing good science on a sizable scale. If you get one talk, practical considerations lead us to market the work in our lab, and market ourselves as quality scientists. That means we need to communicate ideas, but the ideas are only a piece of the Scientist Marketing Plan rather than the ultimate goal of the talk.

I suspect this is self-evident to anybody that’s been in science for a few years. We might be studying animals or plants or microbes or rocks, but science is still a human enterprise. The indicator of a good talk — especially for a junior scientist who needs to get and stay employed — is the fact that the talk was impressive. If you had to choose between giving people the impression that: A) you’re an awesome scientist, or B) a particular scientific idea is awesome, it’s only reasonable to pick A over B. Of course, if someone communicates a scientific idea that is awesome, that should make the scientist look awesome. But there is a big distinction between working towards A and working towards B. The most awesome scientific ideas are not yet wholly convincing, and have weak evidence. They’re new. If you spent too much time talking about awesome science, then it’s easy to give people the impression that you spend more time thinking about cool science then actually doing good science.

This is why science is conservative. (At least, that’s how I feel about it this morning.) Because scientists have a professional need to sell themselves as excellent practitioners of science even more than being transformative. Yes, NSF says that they only want to fund transformative ideas. Well, that might be true, but they only want to do that after the transformative nature has been super-well documented. When everybody already agrees that those ideas are transformative. When the trajectory of research is already pretty obvious. That makes the science in conferences a bit lackluster. You may not know the results, but you know the experiments that are going to happen. I want to be wowed not just by findings, but by the ideas for experiments. I don’t experience that too often.

For quite a long while, I’ve caved in to the social pressure to use conference presentations to demonstrate my quality as a scientist, rather than communicate an interesting and cool scientific idea. A few years ago — I guess right around the time I got tenure — I decided that I wasn’t going to give that kind of talk. And I think it’s worked out well. I gave 4 entirely different talks at conferences this summer. And I think 3.5 of them were mighty damn good, based on the feedback I received.

I remember this entomology meeting in Atlanta in 1999. Most meetings from back then are a hazy blur at most, but this one was memorable. The talk I gave was the a small slice of my dissertation talk, which I had just given earlier in the year. I remember that meeting because every few hours, for the whole conference, people I didn’t know would walk past me in the hallway and say something like, “Hey, I don’t know you, but I just wanted to say that you gave a really cool talk!” A few years later, I had the realization that that was a remarkable one-off thing. I can’t routinely amaze people.

But this summer, I recaptured that mojo. By qualitative indicators, I kicked some serious butt. I think there are three contributing factors. First, in the last few years, I’ve been blessed with remarkable students who have worked with me to generate and evaluate some cool results and ideas. Second, after now that I’ve deemphasized lectures as a mode of teaching, I’ve figured out how to be engaging in a short time span. Third — and I think most important — I’ve decided that when I give my talk, I don’t need to try to impress anybody.

When I gave the talks that (at least I suspect) kicked butt, here’s what they all had in common. They were all short on data and they only had one idea. An idea that was interesting, but not with a mountain of data supporting them. I could have given the exact same talk, with three times as many data slides and with more macho statistics, but I chose to keep it simple, and to sell an idea, instead of selling myself. I was running the risk of looking like a small-time scientist at a small-time university doing cute little work. I realize that the people in the room could have suspected that I was in over my head and didn’t have a lot of data to present, and was talking about ideas rather than data because that’s all I had. But I put those thoughts aside, and decided to sell an idea as well as possible, instead of trying to sell myself as an impressive scientist. I intentionally pared back on the data and the overarching research agenda. I just wanted to speak to an idea and wasn’t too concerned about how it made it me look. And, it turned out, the result was that people thought I gave a really good talk. Were they sold on the ideas?  Maybe, I’ll save that for the reviewers, hopefully by the end of the year if I can get those results into manuscript form.

After going to all those meetings, I’m traveled out — I also did my normal field time in Costa Rica and helped teach a class in the Chiricahua mountains. I don’t want to go anywhere over the next couple months. But, if you happened to be one of those people at one of my talks, I also put on a helluva seminar too, just sayin’. I’m sure the travel bug will be opening its mandibles soon enough.

10 thoughts on “Scientific conferences, the too-slow movement of ideas, and giving an engaging talk

  1. I also think people got clarity from your talks and could grasp the ideas you were talking abotu with little data; people at conferences tend to be tired, distracted, and not fully engaged with talks a lot of the time, and so by keeping it simple, with good slides that have a clear message and one big idea to take away, I think you really did your audience a solid (if they want to know more about your work these days, with all the data, I imagine they can contact you/find you online.

    I wish I’d taken more risks in my own work, but being really conservative with ideas/projects tends to happen in tight economic times I think. Increasing diversity seems to slow down too I think in times when job markets aren’t expanding. And of course, just getting your voice out there, lots of data or not, is important these days…showing you’re thinking and doing things (ideally you have a good combination of new and established research, right?). Science doesn’t usually succeed at the level of individuals…only as a collective do we advance as findings are put to the test repeatedly with lots of variation. It’s the ultimate improv…yes, you found that…and I’m adding this!

  2. “…talking about ideas rather than data…”

    Funny thing is that everyone loves a good idea talk. But we are trained from grad school onward to only slip in the idea briefly – maybe in the conclusion slide – if there’s time. Perhaps because a not-so-good idea talk could be perceived as disasterous.

    I think it’s partly a matter of what’s “safe.” Decent data are always safe. An idea… not necessarily. As with any other performance art, the temptation is generally to go with the safe and reliable.

  3. Hey Terry, seeing this highlighted in Dynamic Ecology’s weekly link-fest reminded me I had meant to comment on this post when it first came out. I agree with a lot of what you wrote here (although for me the ATBC ok-here-we-go-again topic was mechanisms promoting seedling diversity/coexistence) but I’ll press you on the idea that students/postdocs were not there. 30% of ATBC 2014 registrants were students this year, which was a little lower than in Costa Rica last year but not too much (data in link below; not sure about postdocs, we don’t have that as a registration category). I would love to have more, and especially more undergrads, but for what number do you think we should be aiming?

    Out of curiosity, does anyone know what the number/% of students attending for ESA was? Would love to compare.

    Data on ATBC attendance by different category or registrant:

  4. I’d like to have meetings be at least 50% students.

    I’d also like meetings to be a place research-active undergrads with clear interests to be a place where they can shop for advisors and for advisors to shop for grad students. This does happen now, to a very small extent, but more of it would be great for everybody concerned.

  5. PS there is a strong tradition of undergrads going to meetings in Brazil, Argentina, and a few other Latin American countries. Many universities will even provide the bus for them to get there. I think it’s partly a product of being more advanced than undergrads in the US in terms of their major – no electives or gen ed classes, you dive into your major on day 1.

Leave a Reply