Friday links: math for human flourishing, smiley Charles Darwin, and more

Also this week: the scholarly literature as a mud moat, people named Neil vs. imposter syndrome, Joe Felsenstein was way ahead of you, compression algorithms vs. pop lyrics, induction vs. deduction vs. abduction, game theory vs. grade inflation, how to interview for a British PhD position, and MOAR.

From Jeremy:

Remember: I need your help identifying ecologists hired into tenure-track assistant professor faculty positions at N. American colleges and universities in 2016-17.

Francis Su, outgoing President of the American Mathematical Association, marked the end of his term with a speech on mathematics for human flourishing. I won’t try to summarize it, just go read it. Best thing you’ll read all week, I promise.

Markus Eichhorn on discovering that pioneering plant ecologist Heinz Ellenberg was a Nazi collaborator. Ellenberg was part of a scientific team assembled by enthusiastic fascist Otto Schultz-Kampfhenkel to produce maps for the German military. Markus suggests that scientists and their science ought to be evaluated separately. I agree.

“Vast literatures” as “mud moats” that prevent productive online discussion of controversial topics. Includes a very interesting suggestion, the “two paper rule”, that might improve matters.

Sticking with Noahpinion: he says that economists need to do a better job of linking theory and data. Read it, compare to how ecologists link theory and data.

Following on from the previous link: You probably think that the two modes of inference are deduction and induction. But there’s a third: abduction. Also known as “inference to the best explanation”. It’s a fascinating, understudied, and controversial topic in philosophy of science, since it’s not easy to say what distinguishes a good abductive inference from overfitting and researcher degrees of freedom. The philosophical literature on abduction is of direct practical scientific relevance since many everyday scientific inferences are best thought of as abductive rather than inductive or deductive. Here’s a great accessible primer on abduction in the context of economics, that applies straightforwardly to ecology as well. And here’s another accessible primer aimed at philosophers. (ht @Noahpinion)

Hoisted from the comments: In past posts I’ve asked readers to name neglected should-be classics in ecology and evolution. Writing in Am Nat, Pennell & O’Connor suggest a candidate: Felsenstein 1978 on the evolution of ecosystems. Joe Felsenstein’s name also comes up in discussions of everything from great acknowledgments to amusing co-authors to classic contrarian papers. In another old post that I can’t find just now, I asked readers to name current or past scientists who would be, or would’ve been, great bloggers. I’m embarrassed that Joe Felsenstein’s name didn’t come up–he should’ve been right at the top of the list!

The Trump administration is making it harder to access all sorts of government data. This seems to me like the sort of thing that scientists ought to push back against hard, and that scientists can push back against without being seen as left-wing partisans. (ht @dandrezner)

Grade inflation and compression is a thing, especially (but not exclusively) at elite US prep schools, colleges, and universities. In response, some elite institutions are proposing to get rid of grades entirely in favor of qualitative transcripts that would resemble reference letters. Catherine Rampell argues that this move is what game theory would predict–and that it’s not in the interests of less well-off students. (ht Crooked Timber, which has commentary)

Advice for interviewing for a PhD position in the UK. See also the previous post on preparing your PhD application.

Wait, cicada brood X is emerging 4 years early? Anybody know more about this? Is it always the case that there’s a ton of within-brood variance in emergence time?

Aaron Ellison on why he blogs, and some advice on how to get started. Remarks for a general public audience. Related: my advice for anyone thinking of starting a science blog.

Sticking with Aaron: advice for anyone editing a book of contributed chapters. The final piece of advice is “don’t do it!”

I’m a bit late to this, so slightly belated congratulations to David Tilman FRS! (ht@JeffOllerton)

Neil Gaiman on Neil Armstrong’s (and his own) imposter syndrome. (ht @jtlevy)

Pop lyrics are getting more repetitive. As demonstrated via compression algorithms. (ht Marginal Revolution)

And finally, a cheerful visual history of philosophy. 🙂 (ht @kjhealy) Which inspired me to do the same thing for evolutionary biology. Behold:

From Meghan:

NSF’s DEBrief blog posted the results of this year’s round of preproposal reviews in DEB. The cluster “invite” rates ranged from 24-29%.

20 thoughts on “Friday links: math for human flourishing, smiley Charles Darwin, and more

  1. Geoff West, a founding father of Metabolic Ecology, has just published his BIG book on scaling laws …power function rules applied to problems as diverse as metabolism, aging, and the functional structure of human cities. I have not gotten a copy yet but its bound to be great, and probably controversial . Geoff was a particle physicist, who learned to talk with biologists, particularly Jim Brown and Brian Enquist, and the rest is history. I worked with him for several yrs about 2000, and he is very,very good, and listens to others carefully.

      • The book is written as a non-technical introduction to Scaling; and Geoff says there is not one equation in it. More of a popular science book.

      • I’ll still read it with interest. I’ve yet to read a book-length treatment of what I think of as the “scaling law” mindset. Just individual papers describing or proposing explanations for this or that power law. Some of which strike me as deep and insightful, some of which strike me as fairly uninteresting (as when the “explanation” is *so* generic or abstract as to be unhelpful), and some of which strike me as unconvincing or wrong (e.g., as when the data don’t really exhibit a power law as distinct from a mixture of lognormals or something, or when one quite specific explanation is proposed for a power law that could’ve arisen for many other reasons).

    • Yes, Wild Life is one of those books on which your mileage may vary *a lot*. I think it helps to go in knowing what to expect (in my case, I’ve met Trivers and spoken to some people who know him well). I think it probably *doesn’t* help to be a huge admirer of Trivers based on having read his science, or having read and admired the people who blurbed the book (Richard Dawkins, Steven Pinker). Because Trivers doesn’t talk about his science, and because he isn’t the sort of person I bet many of his scientific admirers imagine him to be. He’s not anything like Dawkins or Pinker, for starters.

      I think Wild Life is best read in the spirit of ecumenical curiosity about people in all their glorious and not-so-glorious variety. I suspect that’s the spirit in which Tyler Cowen read it. He voraciously and deliberately reads about literally *everything* as a way of learning and experiencing new things and constantly pushing himself out of his comfort zone. But Cowen’s unusual in that way, and I’m sure that’s not the only spirit in which it could be read with some enjoyment.

  2. One interesting tidbit I learned about the early emergence of some brood X cicadas: it’s a 17 year brood, so the individuals emerging 4 years early are emerging after 13 years. There are of course also 13 year broods. Coincidence? Or does this observation suggest something about the proximate or ultimate determinants of brood length?

  3. I strongly suspect that the same type of analysis, applied to the musical phrasing, would give the same result, that songs with repetitive lyrical and musical phrasing would tend to be one and the same, and that both of these would correlate positively with tempo and negatively with the attention span and age of listeners. Personally, I prefer jackhammers and street noise, and the appreciation of e.g. Bob Dylan and Yes grows daily. And the graphics at that site are really nice, and unique BTW.

    The guy’s on the money with the “vast literature” comments. A really excessively large mass of literature indicates that this is the phrase of choice whenever “increasing body of literature” or “numerous studies” just won’t quite lay the hammer down. The idea of citing a study or two that clearly and definitely demonstrates one’s point, we want to steer well clear of that type of thing.

    Completely agree on the data disappearance issue–that’s a perfect example of where scientists–and possibly legal experts–should be making legitimate noise.

    • #parsimonygate? [clicks through] [reads] [makes this face:

      I read David Hull’s Science As A Process with great interest back in grad school. Hull uses the debate over cladistic methods for phylogeny reconstruction as his core example. Hull’s book came out in 1990. In other words, the 70s and 80s called, they want their debate over phylogenetic methods back. 😉

      Ok, eyerolling and obvious jokes aside, Joe’s comments sound very sensible to me. He’s surely right that Cladistics (the journal, and the method) is going to wither away eventually. But in the meantime, it’s not actually hurting anything or anybody, because it’s a ghost idea ( So anyone who doesn’t already buy into it is best off just ignoring it, unless you’re out to settle personal scores or just like being angry on the internet or something. Because nobody can do anything to make it wither away any faster.

      In retrospect, I’m kind of embarrassed that I forgot to bring up cladistics in my old post on lost causes in science (

      • “He’s surely right that Cladistics (the journal, and the method) is going to wither away eventually. But in the meantime, it’s not actually hurting anything or anybody, because it’s a ghost idea”

        I only skimmed the linked Felsenstein piece and don’t have time to explore how a “ghost” idea differs from a “zombie”, but the idea that cladistics will “surely” fade away as a “lost cause” is directly contradicted by the powerful and ongoing data revolution, one capable of addressing literally all kinds of phylogenetic arguments.

        So I’d like to hear your reasoning for this claim Jeremy.

      • @Jim:

        “Cladistics” here is not the same thing as phylogeny reconstruction. “Cladistics” in this context means only reconstructing phylogenies via maximum parsimony, for philosophical reasons that are independent of whether maximum parsimony actually gives you the right answer. If you follow the links in the post look again at Felsenstein’s comments, you’ll see that the controversy is flaring up because the journal Cladistics routinely *rejects* phylogenetic reconstructions based on modern, computationally-intensive approaches like maximum likelihood and Bayesian approaches. That is, Cladistics (the journal, and the approach to which it is dedicated) is actually *opposed* to the “powerful and ongoing data revolution” in phylogenetics. That’s why Cladistics (the journal, and the approach) is dying.

        All of which is what Joe Felsenstein said in his comments.

      • It’s surely a strange twist to find myself defending a field as impractical and argumentative as is phylogenetics. On re-reading the linked JF piece I realize I did read it the first time. I also read the original editorial. That’s not what Felsenstein says.

        I only partially see his case, especially since he does not characterize the Cladistics editorial fairly (or correctly). The latter is clear that the journal is open to non-parsimony-based phylogenies, if they differ from parsimony-based, as long as a defense thereof is presented. What they appear to be against is the presentation of different trees that may hardly differ from a parsimonious one, or if they do, it’s apparently not clear even to the authors just which method, and therefore result, should be favored. A shotgun presentation of a bunch of trees based on different methods, some to many of which are opaque, without an analysis of which is more likely correct, is not helpful. It’s climate modeling all over again: lots of results with no idea of why they differ or which is best.

        The editorial also notes that any phylogeny is a hypothesis–they’re not saying parsimony is The Right Method To Use. Whether it accurately describe the actual state of affairs for Cladistics manuscipts, I have no idea. But neither does Felsenstein–he openly states he’s just guessing as to what’s going on. Lastly, so what? If Cladistics wants to follow parsimony for reasons they state they’ve hashed out at length in the past and an author doesn’t like it–well they can submit their ms to another evolution or phylogenetics journal.

      • @Jim:

        With respect, I think you’re reading the Cladistics editorial, and Felsenstein’s comments, without enough historical context. Joe Felsenstein isn’t infallible (no one is), but he knows a lot more of the context than you (I know some, from having read David Hull’s book).

        As I said above, I agree with you and Felsenstein that it’s fine for the journal to adopt whatever publication policies it wants and let authors vote with their feet, and that it’s best if these policies are out in the open rather than unwritten.

        I suspect that we’ll have to agree to disagree. I won’t comment further on this topic as I don’t think it would be productive. As always, thank you for taking the time to comment.

  4. A “vast literature” on a topic might mean that everyone has been completely baffled by it for many years. Solvable problems don’t require a “vast literature” – people solve them and move on to new frontiers.

    All the more reason *not* to get a vastlitectomy. Read two papers even if they *are* reviews. I’m sure I could work something in here about seeking more fertile intellectual grounds if I wasnt so sleepy

Leave a Comment

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

You are commenting using your 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.