Let it be noted that I did not know what I was getting into.

I got this book because it was described as an introduction to cladistics -- that is, the scientific (mathematical!) discipline of classifying living things (or other things, like manuscripts) into stemmata (family trees) based on shared traits. This is a very hot new scientific field, and I wanted to learn how it is done.

From that standpoint, you can forget it. No mathematics here. No doubt the vast majority of readers will cheer. But it's also a problem, because author Gee never even says how cladistics works.

Let's give an analogy. To plan the path of a satellite into orbit, you need Newton's Laws and calculus. But you don't have to know calculus to know how Newton's Laws work. You just need to know that pushing makes things move, and the more you push, the more they move, and by using calculus, you can add up all the pushes over a long period of time to figure out the total movement. So you can explain the idea of orbital mechanics without doing calculus. It won't tell you enough to launch a satellite, but it will give you the idea.

Gee never does that. He just handwaves at cladistics and presents it as the alternative to the alleged older version of evolution that sees evolution as a path of continuous progress.

But that's a black-and-white choice with a vengeance. Most people who really studied evolution know that it isn't a form of progress; it's merely a form of change -- of adapting more closely to the environment. Monarch butterflies, for instance, only lay their eggs on milkweed. Is that more "advanced" than a form of butterfly that can lay its eggs on several kinds of plants? Of course not; it's merely more specialized -- and if milkweed dies out, so will monarchs. Often evolution makes a creature less fit for every environment except the one it lives in.

Gee's obsession that evolution is not progress leads him down some rabbit holes -- he argues that, because we weren't around to see how (e.g.) triceratops used its legs and horns, we cannot know how it used the legs and horns, and because we cannot know, we can't even hypothesize. It's true, we can't know. But we can still potentially learn a lot -- "if a horn does this for creatures now, couldn't it have done so in the past, and if it did so in the past, might we not find some traces in creatures alive today?" All those things Gee says we don't know can still be a fertile field for useful experiments.

And Gee is self-contradictory. He argues, for instance, that we can't really be sure which ancient hominid fossils belong to which species. Absolutely true, and fossil-hunters have a dreadful tendency, whenever they find a new fossil, to try to make it into a new species so that they get credit for it. But Gee then starts making arguments about various genus of ancient hominids, such as genuses Homo and Australopithecus and Paranthropus -- and then goes on to use the genus difference as a basis for discussions about behavior (doing exactly what he said above that you couldn't do about a fossil!). I'm not entitled to an opinion, but personally, I've never believed Paranthropus is distinct from Australopithecus, and since Homo is almost certainly descended from Australopithecus, I can't see why that distinction should be drawn, either. Gee would, in fact, very possibly agree with me -- but he can't then use the genus distinction to argue about behavior! He can't have it both ways!

This is a book about an important subject (cladistics) that makes an important point (that evolution is not progressive), but it is so all-or-nothing that I fear it loses almost its entire point. ( )

Stephen Jay Gould, this author is not. (To which he would likely say, "Duh," but then he does not seem to understand the purpose of figurative language.) It does seem like he tries to be, at least in this book (the only one of his I've read as of this writing), but he fails.

His main argument is that the fossil record is too sparsely distributed through too much time to support narrative interpretations, and therefore narrative histories of evolution & adaptation should be replaced by diagrams of similarity.

In making this argument he makes several good points: e.g., current narrative interpretations often make far too much use, unjustifiably, of the myth of progress; and adaptation, while indisputably an important factor in evolution, is too random and too temporally fine-grained for actual historical instances of it to be traced through the fragmentary fossil record.

In support of all of this he gives several examples, each getting its own chapter, such as the evolution of non-fish vertebrates from fish, the evolution of birds from reptiles, and the evolution of humans from apes. In doing so he critiques the particular form taken by the myth of progress in each case.

But he takes far too many pages to do all of these things. The text is repetitive, within and between chapters. It reads slightly polemically. It seemed to me that in a couple of places he would employ a form of argument that he had critiqued earlier in the book. Theoretically, he seems to have a bit too much faith in the explanatory power of parsimony. I found his conceptualization of Deep Time to be simplistic -- he didn't discuss when the conceptualization of the past should shift from historical "ordinary time" to Deep Time, and it seems to me that the nearer one draws to the present, the more supportive the past becomes of historical narrative. Because of this, I found his arguments about the recent paleontological past to be less convincing than those about the more distant past. (For example, he critiqued the myth of bipedalism as the driving force in human evolution, in the same way that he'd critiqued the myth of flight as the driving force in bird evolution. But the former is based on skeletal structure, whereas the latter was based on the presence of feathers. These two things are not equivalent in my mind: a feather is not like a bone, let alone an entire limb or limb-complex. And, structurally and energetically, bipedalism is a shitty mode of locomotion, so I find it really, really hard to conceive of the humanlike pelvis and legs having epiphenomenal origins.)

The biggest problem, though, is what it would mean for paleontology if this guy got his way with respect to its public presentation. He wrote on one page that paleontology already has enough trouble justifying its existence to the budget-meisters. OK, so he thinks that would change if all paleontology produced was a bunch of diagrams? Cladistic relationships, as a work product, are f'ing BORING to most non-specialists (remember Al Gore's and the cranky old 1996 third-party candidate's charts & graphs? Yeah, I thought not).

This is because there's a well-established sociological difference between specialists and the general public that Gee seems to have ignored completely. People like stories. They're the non-specialist way of making complex, unfamiliar information meaningful. So stories aren't really testable hypotheses; big deal. Narratives actually can be critiqued (literarily, wouldn't you know), and they can even change. It's just that these things happen very slowly. So really, it isn't that cladistics is more correct in the long term than narrative, so much as that it allows for greater efficiency in incorporating new information into the paleontological consensus. And this is one context in which efficiency matters far less to the public bean-counters than do interesting stories. ( )

This seems to have been put together from a collection of essays. As a result, the key ideas are approached several times, but never definitively. One of the key notions in the book is cladistics - but I was never able to find out what it was! ( )

I read this on the strength of the author's book Jacob's Ladder, but it's nothing close to that great book.

The subject matter is supposedly cladistics, but while it does discuss that, it does so poorly. It's another of these books written to further one side of an obscure feud, a feud that's vitally important to paleontologists and of little interest to everyone else.
As such it's preachy and repetitive without being especially convincing. ( )

Cladistic analysis/Paleontology > Statistical methods