A Biologist Reviews an Evolution Textbook from the ID Camp
By John Timmer
http://arstechnica.com/reviews/other/discovery-textbook-review.ars/

The Politics of Exploration

Ars book reviews typically focus on works for the general public that we consider significant and insightful. But today we're making an exception: the work in question is meant for school children, and it's an atrociously bad book. So why review it? Because, unfortunately, it may well turn out to be very significant. The leading lights of the Intelligent Design Movement, the Discovery Institute, have written this textbook on evolution, and they are doing everything they can to make sure it gets into schools.

In response to a column I wrote a few months back, one of the authors of the text, called Explore Evolution, kindly sent me a copy. What follows is not a comprehensive examination of the information contained in the text (which would require more text than EE itself), but rather a summary of the history and politics that make the book significant, and my own perspective as a biologist on how that context produced a text that's wildly inappropriate for use in a science classroom.

This doesn't mean that I won't mention the scientific evidence that's referenced in EE, but I generally won't delve into it in extensive detail. For the most part, extensive details aren't actually essential to understanding the problems with EE. So, with that caveat, we'll do as the Discovery Institute has, and start with the politics.

The Strategy for Exploration

The Discovery Institute, as indicated by its wedge document, wishes to eliminate science's focus on natural causes. The group views this focus as the source of society's increasing materialism, which makes it anathema in the belief system of Discovery's members. Stephen C. Meyer, the lead author of EE, heads the Discovery Institute and is mentioned by name in the wedge document, as is coauthor Paul Nelson.

Evolution has been singled out for special ire by Discovery, as it provides an explanation for the origin of humanity based solely on natural processes. Although the ID movement has not developed a research program or even proposed a scientific formulation of its ideas, it has gotten a surprising amount of traction with its attack on the science of evolution. Tapping into a rich vein of American thought that dates back roughly a century, the group's members have used popular books and appearances in the press to argue that the scientific theory of evolution is on the verge of abandonment, having been pushed to its most recent "inevitable" collapse by new molecular evidence.

More significantly, however, Discovery Institute fellows have been attempting to have their arguments against evolution incorporated into the US public school system. They testified in favor of education standards in Ohio and Kansas that targeted evolution for special criticism—Kansas' standards went even further and eliminated reference to science's search for natural causes. In the wake of the Dover case, however, both states have reversed these policies, leaving Discovery without a foot in the door of the US education system.

EE appears to be part of a strategy to change that. In June, Louisiana became the first state to enact a law specifically enabling the use of supplemental materials for the critical evaluation of evolution; similar legislation has been introduced in several other states. EE appears to have been intelligently designed to be the sort of supplemental text that's appropriate under the Louisiana legislation, and so it's likely to be making an appearance in classrooms there. But EE may appear in other states, as the approval process for supplementary material is often far less strict than that governing textbooks.

Inquiry-Based Nonlearning

That said, Discovery faces at least one very significant challenge in its anti-evolution campaign: evolution is overwhelmingly accepted by the scientific community because the evidence for it is extensive and comprehensive. Taking on that evidence runs the risk of simply emphasizing its significance, so EE maneuvers its way around this roadblock by using (and abusing) an approach to teaching called inquiry-based learning (IBL).

IBL avoids the rote memorization endemic in past science classes by having a teacher guide students through a limited version of the scientific process. Students are given a question or problem, provided with the opportunity to obtain information and data relevant to that problem, and then guided through the process of analyzing that information and reaching conclusions based on it. This isn't an "anything goes" approach to science education, though—teachers and lesson plans play an extremely important role in ensuring the students obtain accurate and relevant information and adhere to the rules of logic when drawing conclusions based on it. After all, it's not a good educational method if students come out of it deciding that the force of gravity is random or unmeasurable.

EE gives its authors the chance to determine what information is relevant for students in order to apply IBL to evolution, taking the teachers and professional educators out of the equation. It neatly dodges the issue of the vast evidence that has led to the acceptance of evolution by the scientific community; the book's introduction says that the students will see that in their normal textbooks anyway, so EE's authors can simply present an abbreviated version of mainstream science. 

Perhaps more significantly, it omits the entire process of assisting students in reaching a conclusion. It divides evolution into a series of topics that are discussed separately. Each topic includes a case for standard science, a reply to it, and then a further discussion area, where it switches back and forth between the two. The text assiduously avoids suggesting that any conclusion can be reached at all. At best, it could be described as a partial implementation of IBL, if it weren't for the atrocious presentation of scientific information it contains (more, much more, on that below).

Divide and Conquer

There are two obvious tactical reasons for the book's omission of any explicit conclusions about the "debate." The first reason is simply that the authors know precisely the sort of conclusions they'd like everyone to reach: some variation of the creationism that has been deemed legally intolerable by the courts. But they're also undoubtedly aware of survey results that indicate that well over 10 percent of US teachers "teach creationism as a 'valid scientific alternative to Darwinian explanations for the origin of species.'" By avoiding any conclusions, EE will allow those teachers to lead their students into precisely those conclusions that the book's authors desire, all while providing them, and the Discovery Institute, with plausible deniability.

But the "divide and conquer" approach has a third feature that may also have been precisely what the authors intended. As the book switches back and forth between perspectives and quickly changes topics, what gets completely lost is the internal consistency—the consilience—among the different lines of evidence. The molecular, cladistic, and fossil evidence agree on the derivation of whales, birds, tetrapods, and other evolutionary transitions. But EE presents them separately, attacking each in isolation.

One of science's great strengths is the ability to create a coherent understanding from disparate lines of evidence. One of the hardest tasks faced by science teachers is to weave the lines of evidence together for students, so that they can appreciate just how many phenomena a major theory ties together. By attacking these lines of evidence separately, EE's authors make it easier to claim that the fact that any one of them supports evolution was just a lucky fluke (and they do precisely that). This makes the hardest job of a teacher that much harder, and it makes a mockery of the use of IBL in the sciences.

Shaking and Breaking the Tree of Life
Unscientific from the Subtitle On

The scientific community has very valid reasons for accepting evolution as an accurate description of the history and current development of life on earth, reasons that are so so compelling that aspects of the theory can be safely treated as fact. Even presenting the "nobody knows for sure" conclusion as scientific would seem to be an insurmountable challenge, so authors have responded the only way they apparently knew how: skip the science.

This avoidance of science plays out in countless ways, some of which we'll discuss in detail, but it's worth pointing out that it starts with the book's subtitle, "The arguments for and against neo-Darwinism." During the roughly 20 years I was directly involved in biology research, I'd never come across the term "Darwinism." To confirm that this wasn't my imagination, however, I turned to PubMed, which archives information about, and abstracts from, over 18 million publications. 

Searching for "neo-darwinism" netted 30 references; "neodarwinism" another five. Trying "neodarwinian" and "neo-darwinian" pulled out a whopping 96 references. The term appears to have no significant presence in scientific communications. In contrast, searching for "evolution" pulled out 226,476 papers, while the more specific "selective pressure" 21,553. If this book is all about science, why not use the terminology actual scientists do? Presumably, because the institute producing the book promotes the idea that evolution isn't science, but an ideology, one that their fellows have pulled a Godwin on and attempted to tie to Nazism. Most of the issues with this book are more subtle than that, but the treatment of science doesn't get much better.

Building the Tree of Life

Darwin's Origin of Species proposed a mechanism by which a selective pressure, acting on inherited variations, could transform a single species or bifurcate it into two distinct species. Reasoning that there was no inherent limitation to this branching process, Darwin's single illustration in the book was a tree, with existent species being derived from a single trunk. Darwin concluded that life had been initially breathed, "into a few forms or into one," and all current species were derived from that event.

Darwin's conclusion has been spectacularly confirmed in the 150 years since. The basic biochemistry of the cell is shared by all known organisms, a fact that supports a common origin, while everything from fossil evidence to modern genomic sequencing has supported the tree-like pattern of common descent within the animal kingdom. There are some scientific debates remaining—some argue that horizontal gene transfer has created a web of life at the microbial level, rather than a tree—but scientists don't debate the general outlines of limited origins and organisms related through descent from a common ancestor.

But EE, in seeking to present a case against evolution, argues that there are viable alternative models of the history of life on earth. It favors what it calls an "orchard model," one in which there are many origins of life. In the orchard, current species are the product of severely restricted variation from an undefined number of origin events. Any time a problem with evolution is discussed, a separate origin is the implicit or explicit alternative, and that undefined number of separate origins appears to be very, very large. If that sounds familiar, it should—it's essentially biblical special creation of kinds.

So has creationism's orchard model achieved a sudden surge in scientific attention? Again, turning to PubMed, the answer is no. There is a lot of agricultural literature on orchards, so I searched for orchard occurring in the same paper as the competing theory, evolution. That netted 22 references on the evolution of crop plants and their pests, and three papers on evolution originating at institutions located on roads named after orchards. The more technical term for multiple origins, polyphyletic, pulls out a significant number of papers, but they seem to focus on specific features that have multiple origins, such as flightlessness in birds, rather than on the proposition that individual species do. It's worth noting that "orchard model" was easy to find if Google was set loose on the entire web, however—at sites such as "Answers in Genesis."

Leaving aside the whole choice of "orchard" and its implications of intelligent action, how can biblical creation possibly be presented as a viable scientific model, or the tree of life as so controversial that the orchard might actually be preferred to it? Very carefully, as it turns out.

What if We Had a Controversy and No One Showed Up?

EE's authors manage the feat of making well-studied science appear muddled using a variety of approaches. Part of the argument relies on what I call the "find a PhD" approach: basically, if you look hard enough, you can find someone with a PhD who's willing to say anything.

One such individual is Malcolm Gordon of UCLA. One of the triumphs of recent years has been the elaboration of fossil evidence that reveals the transitional species that existed as tetrapods evolved from lobe-finned lungfish. But Malcolm Gordon of UCLA (at least in the past) has noted that the fossils were found at sites that were widely separated, and concludes that there may have been a couple of distinct origins of tetrapods.

There are a number of serious problems with citing Gordon's work as evidence for the orchard model. The first is that Malcolm appears to be essentially the only person in the biological community that holds this opinion, meaning that this isn't actually a scientific controversy of any significance. Perhaps more significant, however, is the fact that Gordon recognizes that the origin of tetrapods occurred over 25 million years through the standard scientific explanation: evolutionary descent from a common ancestor. Yet somehow, his argument over a single detail has been portrayed as if it raised questions about evolution in general.

Another PhD the authors found is Christian Schwabe, who apparently has established a career studying a protein called reflexin, along with its relatives. But every couple of years he publishes a paper in which he argues in favor of his belief that the genomes of all modern and extinct species originated during the formation of life billions of years ago. According to Schwabe, those genomes have continued to exist, hidden underground as stem cell-like entities. Whenever these cells sense a favorable environment above ground, they head for the surface and self-organize into a fully formed, multicellular animal. No, I am not making this up.

This isn't simply evidence-free (although it is); it's borderline deranged. And yet, in the hands of Discovery's authors, it becomes a serious scientific controversy about the existence of the tree of life. And, if there's any controversy, then students should apparently think twice before accepting that science actually knows anything about the evolution of life on earth.

These are not scientific controversies; they're not part of a coherent scientific case that can be made against evolution. They're actually opinions that have barely registered within the wider scientific community.

Spurious Arguments and Logical Flaws

Seeing an Inch, Taking a Mile

Malcolm Gordon's situation is a good example of another facet of EE's case against evolution, one that ensnares a number of other scientists. Carl Woese is famous for having identified the Archaea, but he also appears in EE's footnotes as part of the case against a tree of life. Woese appears to have argued that the division between nonliving chemistry and actual life should be drawn somewhat later than others do, and he places this division after the Archaea, Bacteria, and Eukaryotes had diverged. As a result, he posits three origins of cellular life, rather than one, although his argument partly hinges on definitions, rather than some objectively apparent biological property.

In the hands of the authors of EE, those three origins of life become indistinguishable from the many, many origins required for the orchard model. Woese, in fact, is grouped in the footnotes with Schwabe, who is arguing for what's essentially an unrelated idea about the origin of life.
This sort of presentation is pervasive in EE, and it's worth commenting on three aspects of it. The first is EE's use of what could be called "borrowed credibility." Carl Woese is a serious and significant scientist who has made contributions to our understanding of the history of life on earth; Christian Schwabe is not. By lumping them together in a single footnote, the authors attempt to illegitimately transfer some of Woese's credibility to the evidence-free inanity that Schwabe publishes.

The second thing worth mentioning is that, like the universe, real biological controversies appear to undergo an inflationary stage within the book. There are a number of different ideas regarding the origin of the Archaea, from Woese's distinct origin of life to a proposal from Thamas Cavalier-Smith, who suggests that the Archaea are a relatively later offshoot of a bacterial lineage. All of the proposals, however, exist within an evolutionary framework where there are a limited number of origins-of-life, and organisms are related to their origin by common descent. Somehow, these arguments over the details are inflated to the point where they encompass controversies that don't exist in the scientific community, such as the plethora of origins required in the orchard model.

This presentation can also be considered a "bait and switch"—take a real scientific controversy, tell your readers that it exists, and then substitute in the controversy you'd like them to think exists without comment. This is obvious in the section on the fossil record, where the Reply section contains a long list of academic discussions of the limitations in our collections of fossils. That section wraps up by claiming these limitations, "have led some scientists to doubt that the fossil record supports the case for common descent."

Who are those scientists? Well, poor Malcolm Gordon (who actually wrote in favor of common descent) gets dragged out again, but the rest aren't actually scientists, nor are their publications peer-reviewed science. Instead, there's a book by an Italian creationist and another by Discovery Institute Fellows, including some of EE's authors. The bait of real issues has been switched to a statement that isn't actually supported by the footnote.

Pruning the Tree of Life

Not content to build up a case for the orchard through spurious reasoning, EE attempts to tear down the tree of life, or at least to prune it radically. The authors have a simple solution in mind: if you break the branches that connect species, then the tree of life looks a bit more like their favored orchard.

What this approach needs to evade, however, is the fact that biologists have very good reasons for drawing specific connections on phylogenetic trees. In Darwin's time, these connections were based on rough inferences derived from morphology and sporadic instances of fossil evidence of transitions. Modern biology has changed all of that. It has rigorously applied cladistic methods to construct more robust trees based on collections of shared characteristics. A description of cladistic methods doesn't appear at all in EE.
In recent decades, those cladistic trees have been reinforced and extended by molecular data. Again, EE's arguments follow an inflationary model in which problems with a study based on a single protein are used to suggest that no molecular trees can possibly be trusted. The fact that we now produce trees based on entire genomes isn't apparently deemed worthy of mention.

The book also suggests that previously unidentified genes argue for the special creation of the species that contain them, neglecting to point out that, as we sequence the genomes of more species, the fraction of unidentified proteins is actually shrinking. That shrinkage supports the conclusion that genomes are largely populated with proteins that are shared among organisms due to common descent. Paul Nelson, an EE author, has even been made aware of this, but appears to have chosen not to act on that information.

Finally, in an increasing number of cases, scientists have individual fossils that provide evidence of key evolutionary transitions. EE accepts that one of these, archaeopteryx, exists, but it essentially argues that its existence is meaningless—it could just be a lucky fluke. None of the other fossils on either side of the transition to flight are deemed worthy of mention. A few other examples of well-characterized transitional fossil series—land mammal to whale, fish to tetrapod, etc.—are mentioned in the book, but they're all treated in isolation so that they can be individually dismissed as exceptions. The fact that there are an increasing number of these series is handled simply: even if they're real, there still aren't enough of them.

Missing Reality

I've tried to give a sense of the major issues with the approach taken by the book, but questionable information, inaccuracies, and distortions are pervasive throughout the book. I'll give you a flavor of some of those in this section. 

Biogeography, which relates the distribution of species through evolution and migrations enabled by past geological configurations, provides one of the strongest lines of evidence for evolution, and that was what led Alfred Russel Wallace to independently formulate the theory. But EE gets basic facts wrong by suggesting that marsupials are pervasive in Australia because they originated in isolation there. Instead, the evidence suggests that placental mammals originated after Australia was isolated, and outcompeted most marsupials on the other continents. Once it gets the basics wrong, EE suggests its favorite conclusion—marsupials get their own trees in the orchard—based on a fact that's consistent with evolution: opossums exist in North America. 

In fact, EE dismisses all of biogeography this way. Sure, evolution could explain the distribution of species, the authors admit. But, since they don't mention the geological, fossil, morphological, and molecular evidence, they find it easy to claim that the orchard model works just as well. Why bother with the messy process of migration when you can just plant a new tree in the orchard?

Also not mentioned in this specific case is the completion of marsupial and monotreme genomes, which clearly retain features of their reptilian ancestors and contain a partial complement of mammalian novelties, exactly as would be predicted based on evolutionary considerations. Hey, if we can make a bad argument about one aspect of their evolutionary history, the authors seem to imply, you can forget all the rest of that evidence—marsupials must have been specially created.

Not Fit for the Classroom
Missing Reality (Continued)

EE frequently fixates on the sudden appearance of some species in the fossil record, as if such an event were a major hole in evolutionary theory. For example, it portrays the Cambrian explosion as an insurmountable problem for evolution and common descent. This ignores the fact that the standard, branching tree of evolution accounts for the appearance of major animal groups within the Cambrian. Its illustration of life in the Cambrian as a series of unconnected lines is simply false.

Elsewhere, the book argues that, "the first fossil bat appears suddenly." But this year, an early fossil bat species was discovered, one that has short wings and claws at the end of its digits adapted for climbing. The discovery of this primitive bat species doesn't simply point out problems with the book's argument; it highlights the problem with this entire class of arguments. Specifically, such arguments are essentially an attempt to rule out evolution by assuming that something (such as a bat ancestor) will never be discovered.

EE also spends time critiquing information found in textbooks and museum displays. In a discovery that will surprise no one, inaccuracies in graphical representations of science clearly exist. But those are no reason to assume that the science behind them is faulty. The equivalent logic would suggest that, because EE's supposed picture of the Galapagos Islands centers on the Yucatan Peninsula instead, we can conclude that geographers have things all wrong.

These sorts of logical flaws are pervasive. In one section of the book, the authors argue that scientists are questioning the dinosaur-bird transition, but once again its footnote references a book published by an author that was a Discovery Institute fellow. The text then focuses on the lungs and, unbelievably, notes that evidence suggests that the distinctive avian lung configuration had evolved in dinosaurian ancestors. This is not much of a surprise; feathers and other avian "innovations" are now known to predate birds by millions of years. But it is actually evidence that runs counter to the original contention: that the dinosaur-bird transition is the subject of scientific controversy.

Rehashing Old Favorites

The text of EE assiduously avoids any mention of intelligent design or creationism, but anyone familiar with the literature of these movements will recognize that their ideas pervade EE. These go beyond the obsessive focus on problems with the fossil record and the repackaging of special creation.

An entire section of the book is devoted to Discovery Fellow Michael Behe's contention that complex, multiprotein systems cannot evolve, a concept called "irreducible complexity." Again, PubMed reveals no significant presence of this concept in the scientific literature. There are 18 papers, only three of which address it directly; all of them conclude that "irreducibly complex" systems can evolve. Indeed, scientists have proposed at least three mechanisms by which irreducibly complex systems can evolve, any one of which would invalidate Behe's contention that they can't. EE mentions only one of these, and again concludes that nobody really knows what's going on.

But EE returns to the arguments that have pervaded the creationist literature ever since scientists developed conclusive evidence that speciation occurs: there are things called micro- and macro-evolution, and the authors are willing to accept micro-evolution while contending that macro-evolution is impossible. The authors accept variation within a limited range (the branches of the orchard), but anything beyond that is no good.

But, in a book that's supposed to be about presenting evidence, there's a curious silence: nothing is said about how to identify what that limited range is, or what the biological basis for the limits are. The footnotes simply register the scientific debate over whether evolution ever occurs through anything other than the gradual accumulation of minor changes.

There is also a notable omission of any discussion of the details of the orchard model. IBL, remember, is supposed to involve an evaluation of evidence, but nothing is said about the biology behind the origin of new trees in the orchard—there's no how or when, or even any indication of whether the process is ongoing. Despite its pervasive appearance in the book, where it's suggested as an alternative whenever a problem with evolution is supposedly identified, the orchard is apparently not to be subjected to any inquiry.

These sorts of grand, evidence-free claims about biology are ubiquitous in the literature of the creationist/intelligent design movements, as is the failure to provide any indication that there is a biological basis for these claims. EE follows this line of attack against evolution precisely. If it never mentions creationism or intelligent design, the decision to avoid doing so appears to be strategic, rather than intellectual. Presumably, it's done for the same reason that many of the Discovery Fellows have chosen to downplay their personal beliefs about the identity of the designer: it would only get them in trouble when they try to get the book into the school system.

Ready for a Science Class?

I'd be remiss if I didn't recognize that there were some sections of the text that actually touch upon real science. There's an interesting debate going on within the scientific community regarding the degree to which evolution is constrained by functional and morphological limits. It's possible that only a limited set of configurations, such as potential enzyme structures, are compatible with function, and EE mentions this. But, once again, inflation sets in, and this very real debate is used to raise the possibility that structural similarities can never be used to infer common descent; nobody in the scientific community is making that argument.

This is pretty typical of all the scientific material in the book. Even when it has its facts right, they're embedded in interpretations that none of the actual scientists cited are likely to recognize. The mere presence of actual science does nothing to outweigh the general morass of errors, distortions, and faulty logic that comprise the bulk of the book. The book as a whole acts like a funhouse mirror, distorting and removing the context from the bits of science that do appear.

Collectively, these problems ensure that anyone using this as a source of information about science in the classroom will leave their students with a picture of modern biology that is essentially unrelated to the way that science is actually practiced within the biological science community. More generally, the logical inconsistencies will leave students bewildered about the nature of scientific reasoning.

In this sense, the book's claim that it represents an attempt at inquiry-based learning is a sham. The process of IBL requires both an accurate presentation of information and an effort to lead students through scientific reasoning based on it. EE not only skips the accuracy requirement, but it abdicates the responsibility for reasoning entirely.

As I was reading the text, I was repeatedly reminded of the testimony of Berkeley Professor Kevin Padian, who described the statement required by the Dover, PA school board as follows:

"I think it makes people stupid. I think essentially it makes them ignorant. It confuses them unnecessarily about things that are well understood in science, about which there is no controversy, about ideas that have existed since the 1700's, about a broad body of scientific knowledge that's been developed over centuries by people with religious backgrounds and all walks of life, from all countries and faiths, which everyone can understand."

Sadly, thanks to the actions of the Louisiana state government, that state's students are much more likely to be exposed to this sort of stupidity. 

But the book doesn't only promote stupidity, it demands it. In every way except its use of the actual term, this is a creationist book, but its authors are expecting that legislators and the courts will be too stupid to notice that, or to remember that the Supreme Court has declared teaching creationism an unconstitutional imposition of religion. As laws similar to Louisiana's resurface in other states next year, we can only hope that legislators choose not to live down to the low expectations of EE's authors.