Wednesday, August 12, 2015

David M. Raup (1933-2015)

David Raup died last month. He was 82 years old. Raup was a paleontologist at the University of Chicago, where he studied the big picture of the history of life, concentrating on mass extinctions. Here's an excerpt from the University of Chicago website [David Raup, paleontologist who transformed his discipline, 1933-2015].
Raup’s former students and colleagues uniformly praised his unique creativity along with his astute capabilities as an academic adviser, senior colleague and paleontological statesman. They remember him for the sweeping scope of the questions he asked, his analytical and quantitative rigor, and his skepticism and humility.

“David Raup ushered in a renaissance in paleontology,” said Raup’s former student and colleague Charles Marshall, SM’86, PhD’89, director of the University of California’s Museum of Paleontology and professor of integrative biology at UC Berkeley. “Before Dave, much of the discipline was centered on describing what was. Dave taught the discipline to think about the processes that might have generated the past record.”

Raup introduced statistical concepts to paleontology that treated the fossil record as an outcome of yet-to-be-discovered processes. Raup was widely known for the new approaches he brought repeatedly to paleontology, such as extensive computation, modern evolutionary biology, theoretical ecology and mathematical modeling.

As Marshall put it, Raup created new intellectual space for paleontology. “That was, in my opinion, his greatest contribution. It is not that Dave just transformed the discipline, but his students, and their students, continue to fill and expand that space,” Marshall said.

Another former student and colleague, Michael Foote, expressed similar sentiments.

“By any conception of what it means to be influential, Dave was one of the most influential paleontologists active during the second half of the 20th century,” said Foote, SM’88, PhD’89, a professor in geophysical sciences at UChicago. “In the areas he chose to touch, nobody, in my view, surpassed him.”
Raup hung out with the likes of John Sepkoski, Steven Stanley, and Stephen Jay Gould and they shared many of the same views on evolution. He was very good at describing those views in books for the general public and that's why I rate him as one of the best scientists who are also science writers [Good Science Writers: David Raup]. His book, Extinction: Bad Genes or Bad Luck? (1991), is one of my top five books on evolution [Top Five Books on Evolution]. Here's a quotation from the introduction to that book ...
I have taken the title of this book from a research article I published in Spain some years ago. I was concerned then with the failure of trilobites in the Paleozoic era. Starting about 570 million years ago, these complex, crab-like organisms dominated life on ocean bottoms—at least they dominated the fossil assemblages of that age. But through the 325 million years of the Paleozoic era, trilobites dwindled in numbers and variety, finally disappearing completely in the mass extinction that ended the era, about 245 million years ago.

My question in Spain is the one I still ask: Why? Did the trilobites do something wrong? Were they fundamentally inferior organisms? Were they stupid? Or did they just have the bad luck to be in the wrong place at the wrong time? The first alternative, bad genes, could be manifested by things like susceptibility to disease, lack of good sensory perception, or poor reproductive capacity. The second, bad luck, could be a freak catastrophe that eliminated all life in areas where tilobites happened to be living. The question is basically one of nature versus nurture. Is proneness to extinction an inherent property of a species—a weakness—or does it depend on vagaries of chance in a risk-ridden world?

Of course, the problem is more complex than I have presented it, just as the nature-nurture question in human behavior is complex. But in both situations, nature (Genetics) and nurture (environment) operate to some degree, and the challenge is to find out which process dominates and whether the imbalance varies in time and space. (pp. 5-6)
If you don't already own a copy of that book, you should buy one right now and read it. They may not be easy to get in the future and your life will be poorer if you don't learn about the difference between bad genes and bad luck.

Raup is famous for the Field of Bullets analogy to explain why extinction is as much bad luck as bad genes. He made a strong case for his belief that chance plays an important role in the history of life. He was not alone in making this claim but it doesn't seem to be popular these days for reasons that confound me. It's part of what I call Evolution by Accident. It means that if you replay the tape of life, things will come out very differently and there's no guarantee that sentient beings like ourselves will evolve [see Café Scientifique: Replaying the tape of life]. You don't have to agree with Raup, Gould, and other experts but if you want to participate in discussions of evolution you have to be familiar with this important concept.


  1. In the picture, Raup is holding an ammonite in apparent reference to his classic paper that introduced an ammonoid morphospace based on two simple parameters. Everyone should read that too.

    Raup, David M. (1966). "Geometric analysis of shell coiling: general problems". Journal of Paleontology 40: 1178–1190.

  2. Well, really everybody should read most of Raups bibliography. The MBL papers to me are the most important papers in paleobiology. And his work with rarefaction which Larry references as well as the morphospace concept which John mentions aren't far behind. Now, arguably David Raup was the most important paleontologist of the 20th century, his Textbook - cowritten with Stephen Stanley - taught paleontology not as it was practiced, but as they thought it should be practiced. It got a generation of scientists interested in the type of research that is now generally done - paleontology was mostly concerned with biostratigraphy before the 1960s, it was mostly there to aid geology, with little interest in biological questions. Raup was one of a few people more interested in contributing to biology and he contributed both through his research, but possibly even more so through the textbook.

    Now, I've mentioned this before, but I'm only doing what I do because of David Raup. Reading his other popular book ("The Nemesis Affair") when I was 13 turned me into a kid interested in Dinosaurs - which of course most 13 year old kids are to some degree - into a kid interested in the statistical analysis of fossil invertebrate data. There was a local newspaper that had a series looking for kids interested in Dinosaurs which ran for a couple of weeks. And every day there's be some kid with their collection of plastic dinosaurs or their dinosaur books. And one day there way me, sitting in front of a computer screen showing a paleobiodiversity curve (the article got my name wrong, listing me as Simone, and it's preserved by my parents, who of course cut it out and put it in a box with my baby pictures and other memorabilia).

    Nobody has had a bigger influence me as a scientist. Yes, my advisors undoubtedly had (and still have) an influence. And sure there are a lot of papers that are relevant to my work. But the questions I ask, the ways I try to get at them, the very basic notion of what my field is about. That's David Raup.

    1. his Textbook - cowritten with Stephen Stanley - taught paleontology not as it was practiced, but as they thought it should be practiced.

      It was translated into Polish in 1984 and had a lot of impact in these parts as well. The Polish edition had a preface written by Professor Zofia Kielan-Jaworowska, my personal hero(ine). She died five moths ago, aged 90.

    2. I don't know if I've mentioned before, but I took a course from Dave Raup, on models in paleontology. Fun stuff. Every week we had to do some kind of project: a morphospace analysis, a simulation, or something.

    3. Simon, you mention rarefaction as a tool Raup helped make popular in paleontology. You might like to know that this is a fundamentally flawed method when it is used to equalize sample sizes, because it systematically underestimates the relative differences in diversity between sites. The correct method is to equalize not sample size but sample completeness, as measured by Turing and Good's sample coverage. See Chao and Jost, Coverage-based rarefaction and extrapolation: standardizing samples by completeness rather than size. Ecology. 2012 Dec; 93(12):2533-47.
      John Alroy introduced almost the same idea (though without extrapolation) to paleontology.

      Lou Jost

    4. I am aware of Alroys work and also of the Chao and Jost paper. I disagree with their position, because the measure they introduce is not fully compensating for different sample sizes at different sites. If you take a dataset, and then use their method on random subsets of the dataset you get different results. That is a problem far worse than the one they have with rarefaction and it is not that hard to show that this property is unique to rarefaction. Whatever problems rarefaction has, it produces consistent results independent of the sample size for the same site.

      It is worth noting that Raup introduced rarefaction not only as an analytical tool, but also as a hypothesis on - at first - mass extinctions and later for extinction in general. It is useful as a null model for this purpose. In the same way, rarefaction as a tool to address paleobiodiversity makes for a good null model of both taphonomy and collection. We can use it to detect (and possibly correct for) biases in either. It's also worth noting that binomial rarefaction (classical rarefaction uses hypergeometric sampling, which here is replaced by binomial sampling) actually tells us something about how many alleles will be lost per generation, given a particular distribution of these alleles under a neutral Fisher-Wright model. Using rarefaction to detect biases and then using these results to correct the model is akin to introducing selection here.

  3. I'll have to read Bad Genes or Bad Luck - hope a nearby library has it, since it's not available on Kindle.

    Of the rest of Larry's five, "Wonderful Life" was indeed wonderful. I just recently gave it to one of my wife's relatives, a very smart young woman who I think will appreciate it. I've also read "Why Evolution is True." I'm in the middle of Eldredge's "Eternal Ephemera" and enjoying it, so looks like I'll probably be going for "Reinventing Darwin" when I'm done with that.

    I've read Dawkins' "The God Delusion," and "The Greatest Show on Earth." (I liked the writing in the former more.) Will think about "Watchmaker" one of these days....

    Another book not on the list that I'm in the middle of and am enjoying because it's written by someone whose research interests give him a different perspective, is Eugene V. Koonin's "The Logic of Chance." Despite everyone paying lip service to the evolutionary importance of microbes, it's rare and pleasing to find a book that concentrates on them to such a degree.

  4. "Logic of Chance" was briefly free on Kindle. I got it then, and it's still there. Not a temporary loan. I think it stumbles into some odd speculation toward the end, but from that book I discovered that the majority of genes (protein coding) were invented by microbes, and that the most important and difficult bits of evolution were done before the Cambrian. Is that true? I haven't read anything to the contrary.

    1. How would you determine which bits of evolution were most important and difficult? Why would you think that the invention of new protein coding genes would be the important bits?

  5. I assume that novel sequences having little or nothing in common would take a long time to evolve. Based on a layman's resding of Koonin, it seems to take one or two million years to invent a new motif. That's for all living things combined.

    By comparison, the diversity of mammals evolved in an eyeblink. It would seem modification of regulatory networks is easier than inventing new proteins.

    1. So, "important" and "difficult" might not be well correlated.

  6. One of Raup's students was creationist Kurt Wise whom he helped send to Stephen Gould. Raup was also part of the Pajaro Dunes Intelligent Design conference.

    1. I do wonder why he was there. There's also on the web a claim that Raup used the manuscript of Darwin on Trial in a graduate seminar, but I can't find any details. Both Mike Foote and Charles Marshall would have been students at around that time. I wonder if they know anything about it.

    2. I was there (at the Pajaro Dunes conference). Not only was Raup in attendance, he contributed a really interesting theoretical talk about how many 'transitional forms' we should expect to find in the fossil record, if common descent were true. Raup and UC-Berkeley professor Phil Johnson, the Pajaro organizer, had become good friends, and that led to the invitation.

      Raup did indeed use an early version of the MS of Darwin on Trial as material for a graduate seminar. I hope to write up the whole story next week, for the Discovery Institute blog Evolution News & Views. Raup was a wonderful scientist and interlocutor, and will be missed dearly.

    3. Hey Paul, isn't ID okay with common descent? That's what some IDiots say at UD and elsewhere. And why don't you allow comments at ENV? What are you afraid of? Don't you have any confidence in your 'position'?

    4. Paul,

      I'll be interested to see the story. Do you know who the students were, and do you know if Marshall and/or Foote was among them? What year was that?

      I presume Raup's paper concluded that we do indeed find what we would expect to find if common descent were true. Is that correct?

  7. John,

    I think Dan McShea might have been in the class (I'm going to check with him). The years in question would have been 1989-90, because that's when Johnson's MS was circulating among outside readers such as Raup.

    About Raup's Pajaro talk. The original plan for the meeting was no presentations, but simply open discussion, around 4 questions posed by Phil Johnson in a long letter sent to the participants beforehand. So no one actually prepared papers: what happened was more interesting. Participants just gave spontaneous short talks, illustrating their points on a whiteboard -- and Raup's was deliberately provocative, especially because Kurt Wise, whom Raup had mentored when the latter was an undergraduate at the U of C, was also present. Raup's answer to his own question (about the probability of finding ancestor-descendant pairs in the fossil record) was "Well, it depends." Mike Foote later wrote a paper, published in Paleobiology in 1996, where he notes at the end that if we fail to find ancestors, this might count as evidence against "evolution" (meaning common descent). Raup is credited as "personal communication."

    I'll say more about all this in my write-up for ENV next week.

    Hey, Whole Truth: I voted for open comments at the DI blogs, but was outvoted by the other contributors, and especially the editors when the blogs began. No one wanted the unbelievably tedious task of monitoring the trolls.

    1. Yes, Dan is very likely, and within that timeframe so are Mike and Charles. They all left right around the time I arrived, in 1990. You should check with all of them.

      Could you give the exact Foote citation?

    2. No one wanted the unbelievably tedious task of monitoring the trolls.

      Perhaps the principles of intelligent design could be used to generate an algorithm to detect and delete pertinent, critical questions and comments, a.k.a. posts from "trolls".

    3. The Foote citation is
      Foote, M. (1996) "On the Probability of Ancestors in the Fossil Record", Paleobiology, 22:141-151.

    4. "Hey, Whole Truth: I voted for open comments at the DI blogs, but was outvoted by the other contributors, and especially the editors when the blogs began. No one wanted the unbelievably tedious task of monitoring the trolls."

      So then, Paul, rather than growing a pair and distancing yourself from your fellow theocrats at the discotoot you chose to stay in that cult for lying cowards. I'm shocked, not.

      Trolls? Is everyone who questions or challenges you IDiot-creationists a troll? And since you snivelers are so worried about monitoring "trolls" at your ENV pulpit, why don't you at least regularly engage with questioners, challengers, and "trolls" at sites like this one, Pandas Thumb, AtBC, etc., where someone else does the monitoring? These sites get IDiot-creationist questioners, challengers, and very tedious "trolls" but don't shut off all comments. You discotooters rarely if ever even comment at UD, where you would be protected by your obedient mouthpiece banny arrington.

      I would think that you so-called 'intelligent design leaders' would be eager to regularly comment at UD so that you can garner the oohs and ahhs of groveling admiration from the IDiotic sheeple there that contribute to pushing your theocratic agenda. I suppose it's hard to find the time to preach your sermons from anywhere but the ENV pulpit when you're very busy at trying to cram your religious dogma into science, education, government, and everything else. Maybe you guys should hire a few more lawyers and philosophers who have no clue about evolution and evolutionary theory to help lighten your load.

      By the way, you IDiots really should stop all of the false accusations aimed at atheists, evolutionists, "Darwinists", materialists, etc., that you maliciously spew, and especially the ones where you accuse anyone who doesn't kiss your asses of being equivalent to Hitler/Nazis. After all, you're the thumpers who support genocide and lots of other atrocities.

    5. Hmmm...that Foote paper seems to have nothing to do with what Paul Nelson said. Either he's mistaken or you have the wrong citation.

    6. I looked at Foote's CV. The only other Paleobiology paper he published in 1996 was with Raup, so it seems unlikely that it would contain a personal communication from Raup as a citation.

      If anybody wants to look more thoroughly, Foote's CV is here:

    7. John and Dave,

      Mike Foote’s citation to “Raup personal communication” is here:

      "Given estimates of completeness, median species duration, the time required for evolutionary transitions, and the number of ordinal- or higher-level transitions, we could obtain an estimate of the number of major transitions we should expect to see in the fossil record. This would have important implications for whether the small number of documented major transitions provides strong evidence against evolution (Valentine 1986; D.M. Raup personal communication)."

      Mike Foote, “On the Probability of Ancestors in the Fossil Record,” Paleobiology 22 (1996):141-151; p. 148.

      Paper available here:

      This was the topic Raup addressed in his informal Pajaro Dunes talk in June 1993.

  8. I don't think "if we fail to find ancestors, this might count as evidence against 'evolution' (meaning common descent)" means the same thing as "This would have important implications for whether the small number of documented major transitions provides strong evidence against evolution". But maybe that's just me.

    Anyway, if we find ancestors, how would we know? That's one problem Foote raises but doesn't deal with.