Biocurious is a weblog about biology, quantified.

Tiktaalik roseae legitimizes creation science [Updated]

by Andre on 7 April 2006

That’s right, paleontologists have discovered a “missing link” between fish and land animals! Unlike silly PZ Meyers, I can see the true significance of this result: rather than showing that evolution makes verifiable predictions even when operating in historical mode, it actually shows that creationists are scientists too since you can no longer deny that they make falsifiable predictions. You see, they assume that Earth was created in basically its present form and so there is no need for transitional forms with the obvious prediction that you won’t find any! Now that someone has found one,* creationists can happily join the ranks of respected scientists with their spectacularly falsified predictions.

And the best part? Just like real scientists, the more they learn, the more questions they have, i.e. Where are the missing links between normal fish and Tiktaalik roseae and Tiktaalik roseae and land animals? The corresponding prediction: “Bet you won’t find them!”

You can’t argue with that.
I saw Edward Daeschler at the Academy of Natural Sciences in Philadlephia last year where he gave an informal presentation on this work and about the expedition. One thing I noticed in the talk is that, although they knew what they were looking for and knew roughly where to find it, the fraction of candidate sites that they could actually excavate is minuscule. They were either unbelievably lucky or these fossils are actually pretty common in the area. So, here is your homework: develop a large scale ultrasound device that can detect the difference between fossilized animals and rock and then scan the whole ridgeline where they were looking, discover incredible fossils, get many Nature covers as you let the results trickle out. Repeat in other fossil rich areas. Enjoy fame and fortune.

Another thing I noticed is that only the supervisors are listed as authors on the two papers they published. I know there were many grad students also involved in the project because Daeschler showed a picture of someone fighting off a dust storm trying to get gear from a landing helicopter and he joked that that’s what they’re for. At least, I thought he was joking… Maybe it’s a palaeontology thing, but if this was say, a particle physics paper, all the significant contributors would be there. Get a load of this from the acknowledgments:

The illustrations are the work of K. Monoyios. Specimen preparation was performed by C. F. Mullison and B. Masek. NUFV 108 was discovered and quarried by S. Gatesy.

Poor S. Gatesy. (S)he doesn’t even have a first name.

So the authors didn’t discover or quarry the fossil or take care of the specimens, and they didn’t even prepare the figures! But they’re supervisors, they must have at least named it… oh wait, they also acknowledge

M. Shuvinai [who] coordinated the naming effort.

Life imitates PhD Comics ** yet again.

*Not that this is the first of course…

**The first comic in that series is here.

Update: It turns out that S. Gatesy does have a first name—Stephen—and that it’s unlikely that he was cheated or mistreated in this case since he’s already a tenured professor at Brown and apparently has known at least two of the papers’ authors for something like 20 years. The reader that brought this to my attention, who would like to remain anonymous, also suggested that there is a significant cultural difference between paleontology and lab sciences that could go a long way towards explaining the apparent differences in authorship practices.

  1. PhilipJ    3973 days ago    #

    When it comes to particle physics papers, I often feel like everyone who has ever taken an undergrad particle physics course must be in the list! (If only this were true, I’d have quite the publication record by now…)

    And those acknowledgments are brutal. If I ever get stuck in a sand storm while trying to do my research, I had better be on that paper.

  2. nate    3970 days ago    #


    We had a discussion about this a while ago (at Chaucer’s?...) but it’s worth mentioning again: your post illustrates two of the real problems with Popper’s account of science.

    First, if we take negative evidence as the sole means by which a theory is refuted, what do we scientists actually do when we observe phenomena that contradicts our current working theory? We might prefer a radically different theory, but we might equally well just replace one of the hypotheses of the now-defunct theory with a premise that is consistent with our observations. According to Popper, the radically different theory and the slightly-tweaked-version of the refuted theory are on equal grounds. And, here’s the critique, because there are so many ways of choosing a new theory, we need an explanation for (a) why we do what we actually do and (b) why it seems to work (if we’re realists: why we believe that our theories are ever-better approximations of reality). Popper’s account just doesn’t explain how or why we choose between these two equi-compatible (wrt our observations) theories.

    Another version of this critique (since we’re talking abut acknowledgements too, I read this idea in a paper by Peter Lipton): if you think of a theory as a giant logical implication P1 & P2 & ... & Pn implies Q, then when we observe not Q, all we know is that one of the premises Pi does not hold. There may be many ways to replace one or many Pis with something else so that not Q is entailed, but Popper doesn’t explain how or why we actually do this.

    Second, if we believe that negative evidence is the only real evidence, then we really have to be skeptics about the ‘truth’ of scientific theories. Because a theory is only as good as its last refutation, we seeminly never know anything, except that it is consistent with all our observations up until now. As far as correspondence with the ‘real world’, or some notion of teleology, well, good luck. Followed to its natural conclusion, this leads to a radical skepticism about pretty much all scientific knowledge… which I assume most of us do not want to buy into.

    So if we’re going to be Popperians, we need to remember that the evidence supplied by this ‘missing link’ fossil doesn’t actually tell us anything about theories that it is consistent with, only those that it is not. While I (not a fan of Popper) agree that the missing link fossil poses serious problems for literal-seven-day-the-earth-is-6K-years-old-and-the-species-we-know-today-were-created-as-such-by-God-Creationists (sorry for all the dashes); on his view it doesn’t really tell us anything else. In particular, it does not lend any support to the theory that the universe was not created by anyone and that all life evolved via random processes, and it does not refute the theory that there was a Creator of the universe, but that species also evolve à la Darwin.

  3. Andre    3969 days ago    #

    Hi Nate,

    Thanks for the insightful comment. My opinions on the philosophy of science are in their infancy but just for the fun of it, I’ll try to outline what I think about how science “actually” works (as if I know!).

    Science progresses by maximizing falsifiability (theories that make the widest variety of furthest reaching predictions are preferred), maximizing parsimony (theories that use the smallest number of premises are preferred), and by a sort of necessary—if not desired—conservatism (it’s not that a slightly new theory is preferred to a radically new one, it’s just that it’s nearly impossible for someone to come up with a whole new framework that is consistent with everything else we’ve observed without keeping the vast majority of the premises of other theories).

    In principle, any of your n premises could be modified to account for “not Q,” but I think in practice it’s more difficult since theories tend not to be a linear chain of premises leading to predictions, but something more like a web of interdependent premises. For example, if P1 is conservation of energy, there are many “not Q” that could be accounted for by simply scrapping P1, but since so many other theories whose predictions have so far turned out to be consistent with observation also have P1 as a premise, it is not so easy to get rid of it. Real world case: it was easier for Pauli to postulate neutrinos than to reinvent a new kind of physics that didn’t assume energy is conserved. You could argue that it is more parsimonious to say “physics stays the same, but energy is not conserved in this particular particle decay” but this idea fails utterly in the maximizing falsifiability department since it makes no new predictions.

    As you said, this does lead to a kind of radical scepticism about scientific truth for two reasons. Where we disagree is your assumption that I don’t want to buy into that. :) First, I think it’s entirely possible that someone will discover a new framework that meets the first two conditions for progress, but is based on entirely different premises that also account for everything we observe. It might look totally different from the ground up and destroy the idea that what we have now is even close to the “truth” because the new theory seems better in every way except for its unfamiliarity. Maybe that means I’m not even a realist, but I shouldn’t go that far since I don’t know everything that connotes! Second, the very presence of parsimony as a fundamental principle takes a lot of the steam out of a claim to truth since it’s such a pragmatic principle: after all, it could be the Truth that we are all in the Matrix. We just don’t have access to that kind of truth, so I leave it out of science.

    In this sense, I wholeheartedly agree with your last point. For me, there’s no argument on that since it’s equally possible that the universe was created a microsecond ago to look as if it was created billions of years ago.

    Here’s an analogy to give you a flavour of my feelings for this type of point. I used to get frustrated when a math teacher in high school accidentally gave us a trigonometric equation to solve that had no explicit solution. Eventually, I realized that it was impossible (not just really hard) to rearrange things to get such a solution and I happily gave up. Similarly, it’s impossible (not just really hard) to “lend any support to the theory that the universe was not created by anyone” so I don’t try. Similarly, it’s impossible to “lend any support to the theory that [...] that all life evolved via random processes” but it becomes harder and harder to come up with an alternative with each new discovery that is consistent with it.

    That was a little bit rambly, so here’s a summary: Popper + science converging to Truth does lead to a dilemma. I choose to give up on Truth since, whatever you think you observe, you could always just be experiencing a convincing virtual reality and so I take a pragmatic view that what we have works pretty well and that something radically different is possible, and may prove necessary (consider current efforts in high energy physics), but is very difficult to achieve.

    Finally, since positive evidence has such intuitive appeal, here’s my (weakened) version. I consider something contingent positive evidence for a statement if the only way you can think of to counter the statement is with some flavour of the argument that we’re in the Matrix. So, I consider the fact that I can see a desk in front of me (and could do innumerable other experiments to support that assertion) to be positive evidence for there actually being a desk in front of me, contingent on the assumption that I’m not being elaborately deceived.

  4. agm    3969 days ago    #

    Nate, basically, Occam’s razor (or what is commonly understood to be Occam’s razor as applied to science, i.e., pick the explanation that requires the least-amount of work to make consistent with everything else). If your options are radical new explanation or tweaked current theory, in the absence of a compelling reason to go with radical new explanation, you pick the one that gets just as much done but has much less likelihood to contradict the rest of the framework.

    Note that this lies soundly in physicists trying to emulate the lazy nature of, well, nature (as Brian Greene puts it).

  5. nate    3969 days ago    #

    agm and André.

    I’ll grant that one could use Ockham’s razor / parsimony to describe what Thomas Kuhn calls “normal science”, the period of incremental refinements / extensions of existing theories, but doesn’t it fail to explain why truly revolutionary theories are ever accepted?

  6. Jason    2617 days ago    #

    Since biblical creationists hold that all living things were not necessarily created in the form they now hold, but are rather descended from families (baramin) that contained the potential for the forms now extant you are guilty of burning a strawman.

    Tiktaalik is not a transition between fish and land animals. Fish and “land animals” are suprespecific groups. You have not identified a lineage, you have just created the illusion of one. In a world that includes the platypus finding a mosaic animal like tiktaalik cannot be considered particularly odd.

  Textile help