Why do senior academics ramble on?

Last month I was at the annual conference of the Canadian Society for Ecology and Evolution. It was a great meeting, as usual, and contributing to that were a whole lot of great talks.

I’ve seen a lot of conference talks over the years, and I’ve noticed a pattern: surprisingly often, grad students finish early, while senior scientists run long*. What I realized this year is that I’ve become part of that pattern: this year, like the last couple, my talks have run longer than planned (see the nifty little figure above). Nobody had to haul me away from the podium; but my talk, which ran a nice tidy 12 minutes in practice, was 14 minutes delivered (using up most of my 3 minutes for questions). As a grad student, I often had a couple of extra minutes to spare. So why is this?

I can think of four possibilities – four things that have changed for me over the years, as I suspect they’ve changed for other senior scientists who run long like me. I think two of these aren’t behind my own pattern, and two are – and I’d be interested to hear whether they’re involved for you too (whether you’re junior or senior). Here they are:

I don’t practice as much. As a grad student, I practiced each talk dozens of times, until I could deliver it upside-down in a barrel of water. (In fact, my talks were probably so over-practiced that the barrel would have improved them). Nowadays, two or three run-throughs are about all I do. So, do I run long because I don’t take prep as seriously now? This hypothesis fails, I think, on two predictions. First, grad-school overpreparation should have left my talks bang on their planned length, but in fact, they ran short. Second, underpreparation now should increase variance, leaving some of my talks short and some long – but that’s not my pattern; I haven’t run short for years.

The stakes are lower. As a grad student, every talk I gave seemed like an audition, with an audience full of potential employers. (I was probably right to think that way.) With the stakes high, I cared a lot about giving a good talk, and part of that was keeping to time. Now I’m a tenured professor with little to lose! Very subjectively, though, I think this hypothesis fails too. I’m far from the world’s most awesome speaker, but I don’t think my talks are terrible, and I think I’ve improved since my grad-student days. In addition, the fact that I’m writing this post suggests that I do care about running long.

I don’t get nervous anymore. I’m quite sure that my early talks usually ran short because I was nervous (well, maybe not “nervous” so much as “quaking in my boots”). Nervous people talk faster: what took 12 minutes in the privacy of my office could pour out in 9 minutes as I stood trembling at the podium. By now, I’ve given dozens of conference talks and departmental seminars and outreach talks, and (I’m not exaggerating) hundreds of undergraduate lectures. Pretty much everything that can go wrong has gone wrong**, and none of them ended my career. As a result, I don’t get nervous anymore (with one interesting exception I’ll save for a future post), and I don’t speed up in front of an audience. This can’t be the whole story, though: it explains why I ran short then, but not why I run long now.

I’ve learned to improvise. Part of my overpracticing early in my career was fear that if I strayed from my planned talk, things would go badly. Now, though, I improvise – in details, I mean, not the overall structure of the talk. I’m enough better at giving talks that I can improvise the precise sequence of points or set of details I cover without too much danger. Now, I don’t think this kind of improvisation explains my pattern. What does is another kind of improv: I’ve learned how to connect my talk, on the fly, to the talks before mine in the session or earlier in the meeting. Sometimes I refer explicitly to a previous talk, and sometimes I just adjust mine to make a connection implicitly. I like drawing these connections, and I think they add value to a session and make it more than just the sum of its talks. The problem is, they take a little bit of time, and it’s time I don’t budget for! I noticed myself doing this in my talk last month, realized I’ve been doing it for a while, and I think it accounts for a big part of my pattern.

What does all this mean? Well, maybe not a lot, except that I feel better having figured out (I think) the perfectly sensible reasons behind the pattern in my nifty figure. I promise not to think less of early-career scientists who talk too fast and finish early, because I understand why that happens. And I think I’ll be less irritated at my senior colleagues who run a little bit long***, like me, because I’ve identified some value they’re adding when they do.

Now, if I could just get those coffee break lines to move faster…

© Stephen Heard (sheard@unb.ca) June 25, 2015

*It’s also true that most of the best talks are given by grad students, but that’s not my topic today – perhaps because I don’t want to think about it too much.

**Well, I haven’t yet thrown up during a talk, although I know two people who have.

***But I’m still plenty irritated with ones who run way long, and skip over 25 of the 45 slides they had somehow planned to cover in a 12-minute talk. Go ahead, share your stories about disasters like that in the Comments (but disguise the names to protect the guilty).


13 thoughts on “Why do senior academics ramble on?

  1. Rob Johns (@robcjohns)

    Nice post Steve!

    My talks tend to vary between long and short, partly because I usually only prepare for a few key slides (hypothesis or data slides usually) in the talk and ad lib the rest. Whether I overfill the gaps between with extra detail or end up running short appears to depend on how much coffee I’ve had in the lead up to the talk…

    I should say that for myself at least I actually prefer the variability that comes with this style. I spend a lot of time preparing and organizing slides but my talks tend to suffer if I over prepare the delivery (I have dozens of grad school talks that attest to that).


  2. Joshua Martin

    I still hit my mark these days (postdoc) but I now have stuff to leave out. As a grad student, I was scraping all the bits of delicious data from the bottom of the jar, giving details about methods, background from other labs, etc., just to fill the time.

    Liked by 1 person

  3. aonghais

    Nice post. I’ve never really practiced my talks that much, save for a couple of run throughs on my own the night before. I try to plan my slides so that there are some where I can gloss over or go into more detail on depending on how I’m doing timewise – this strategy falls down when I’m not delivering a talk in a room where I can glance at a clock every so often


  4. Jeremy Fox

    “underpreparation now should increase variance, leaving some of my talks short and some long – but that’s not my pattern”

    That’s not true for me. If I’m underprepared, it increases the length of my talks, because I use more words than necessary to say what I want to say. It’s not a matter of improvising to add additional content–nods to others talks in the session or whatever. I just use more words rather than fewer.

    Re: your final footnote, at the ESA meeting many years ago I did see a theoretician give a 20 minute talk comprised of something like 50 overheads (this was back when theoreticians routinely used overheads at conferences). Each overhead just had equations on it. The speaker would put up an overhead full of equations, say “I don’t really have time to explain this”, or words to that effect, and would then go to the next overhead. So by the end of the talk, we’d seen all 50 overheads–and learned nothing. I was told later that this individual was known for this–always bringing a 50-minute talk’s worth of material and then chopping it down on the fly if the timeslot was shorter than that.

    Re: the best talks being given by grad students, I haven’t found any correlation between quality of talk and seniority of speaker, though of course I haven’t formally kept track so perhaps there is a correlation (positive or negative!) I’ve failed to notice. (And of course I’m not going to see a random sample of talks from any well-defined statistical population…) I do think that poor student talks tend to be poor in different ways than poor talks by senior faculty. Students are rarely underprepared–but lack of preparation isn’t the only cause of poor talks. And there are some sorts of good talk that are only given by senior faculty.

    More broadly, I don’t know of any reliable predictors of talk quality that a conference attendee could use to choose talks, besides “you know, or are reliably informed, that the speaker has given good talks in the past.” https://dynamicecology.wordpress.com/2011/05/21/are-there-any-reliable-predictors-of-talk-quality/

    As an aside, I’m interested that you, like me, asked commenters not to name names of people they’ve seen give bad talks. I have the impression (perhaps incorrect) that we’re more willing to publicly criticize bad papers than bad talks. I’m not sure why that is: https://dynamicecology.wordpress.com/2012/08/28/are-we-harder-on-bad-papers-than-bad-talks/


    1. ScientistSeesSquirrel Post author

      I don’t remember which ESA it was, but I saw that talk too – it was at least a dozen years ago but my memory of it is crystal-clear. I also learned nothing from it except not to attempt 50 slides of equations in a 12 minute talk… Theory talks are difficult to give well, I’ve found, but I hope none of mine have been that bad.


  5. Pingback: Link Round-Up: On Academic Conferences

  6. Pingback: Don’t fear falling at the edge of knowledge | Scientist Sees Squirrel

    1. ScientistSeesSquirrel Post author

      That might be true, but I’m not sure it fits the pattern. That should make me run long in practice, too, not just in delivery for real. But a slight tweak of your hypothesis: I have more to say now, and although I still package the right amount in my planned talk, now it’s harder for me to resist temptation to add more (since I have so much more to say!) when actually at the podium. This could well be true. Thanks for commenting!

      Liked by 1 person

  7. Pingback: “Two cultures” and the talk that terrified me | Scientist Sees Squirrel

  8. Pingback: Can we just put down the laser pointers please? | Scientist Sees Squirrel

  9. Pingback: Posts for conference season | Scientist Sees Squirrel

Comment on this post:

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.