This book could be called the philosophical manifesto of the metaphysical Darwinian fundamentalists.
It is packed with ideas, some of which are brilliant, and some not so brilliant. It is important to remember that this is a work of philosophy rather than science. It is an important work to read, even though it is likely to energize creationists as much as it would its own supporters. Dennett is an unabashed metaphyical naturalist, which distinguishes him from such critics of creationism as Robert Pennock, who believes that science is restricted to methodological naturalism.
The book is divided into three parts: philosophical Darwinism, Darwinian thinking within biology, and a Darwinian memetic theory of the mind. The first two parts have powerful and useful ideas; the memetics of the third part is weak, even though Dennett may care about that portion most strongly.
In part one, Dennett presents a pyramid of orderliness proposed by Locke, with God at top and chaos at bottom. He then describes natural selection as a Universal Acid that transcends biology and can operate from metaphysics and cosmology to thought. Natural selection allows the flow of order to move in reverse: from Chaos to Order to Design to consciousness and thoughts of the infinite. While I don't share Dennett's uncompromising metaphysical naturalism, I like many of the ideas he presents. Whether or not the pyramid is truely inverted, Darwin's demonstration that you can get organic Design from Order and Time proves Locke wrong that such an inverson is inconceivable. Whether it is True, it is nonetheless conceivable. This book has the clearest exposition of that thinking that I have read, and it shows why religious fundamentalists might view evolution with such loathing (beyond the issue of Genesis disagreeing with the geological record).
I liked his explanation of natural selection as an algorithm that is not guaranteed to produce a specific outcome. To me, it makes sense that substrate-neutrality allowed Darwin to get it right without knowing the mechanisms that produced it. I liked Dennett's ideas about Design, such as how you unscramble an egg. He does a nice job of showing where the scientific and philosophical connect and inform each other.
Dennett describes a Library of Babel, which contains all possible texts. It helps visualize the enormous size of design space. As a librarian, I appreciated how the Library of Babel pulls Library and Information Science into the continuum of other sciences and rational thought. I agree that you could come up with a DNA coding scheme for the Roman alphabet and code all of Babel into DNA sequences, but I disagree with Dennett's claim that the Library of Mendel contains the Library of Babel *because* it contains our DNA, and we are capable of generating items for the Library of Babel. In the same sense, I disagree that modern human culture is really part of the extended phenotype of homo sapiens. To me, that claim is too reductionistic: it seems to imply that we are hard-wired for written language and the wheel. We may be hardwired for spoken language, but details of culture are to me emergent phenomena.
Dennett well compares the detection of plagerism to the detection of biological homologies. He contrasts art and science, where a scientific discovery has the air of inevitability (Pasteur even told his students to make their results seem inevitable) while an art work is fundamentally not inevitable. The Unity of Design Space is a puzzler for me. As I have said, if it means that all human artifacts and ideas are coded in our DNA, then I disagree with it. But I think that it merely says that there is a geneology of designed things: in order to have intelligent design, you first have to evolve intelligent agents to design. And if that is his claim, then that was implied by the inverted pyramid, and he could have cut the long wind.
I like his argument that "biology is engineering" albeit a non-human kind of engineering. I liked the idea that biological structures develop by an interplay between historical accident and "the 'discovery' of important truths." I like the section on reverse engineering, artifact hermeneutics and biology, his revision of the Panglossian paradigm into the Leibnizian paradigm, and his idea for measuring design: "by comparing the cost of making predictions from the lowest-level physical stance ... with the cost of making predictions from the higher stances: the design stance and the intentional stance," (p. 237). Here is a great line for a cocktail party: "Biology is the hermeneutics of designedness."
As to his critique of Stephen Jay Gould, it centers on Dennett's point that if you don't start with the hypothesis that the design should be there for a reason, you can't analyze functions at all. Dennett's critique of Eldredge's example of horseshoe crabs comes to mind. Unless we hypothesize that the Jurassic species actually *used* their legs, and tried to get optimal performance out of their legs, we can't really hypothesize about why they were that long, and why they are now shorter.
I have a friend who is a biology professor specializing in the ecology of spiders. We are both Gould fans. When I summarized Dennett's critique, she readily agreed and said that while Gould in the end saves his criticism for shoddy adaptation stories, his rhetoric makes it sound as though all adaptational hypotheses are to be avoided, which would nearly grind the field to a halt. For example, Lynn Margualis once criticized my friend's research as being too adaptationist. But she would have no questions to test if she did not begin with adaptation assumptions about optimality. Again, Lynn manages to stay within the realm of science in her own writings, but her rhetorical flourishes encourage a group of new age Gaia fans to bash biological research in general.
SIDEBAR: However, I still love Gould. Gould delights in pointing out that the adaptive stories we concoct often come from our cultural wishful thinking, and that other explanations are available--that a feature might be a non-functional by-product of some other adaptation, that the range of possible adaptations is often limited by the developmental channel that the species is in, that the feature was adapted to a previous mode of existence and is now useless, etc. I find Gould a healthy corrective to the knee-jerk idea that a feature must be optimally adapted for the current organism. Also, I like his keen (well, to me keen) sense that what is appropriate for DNA is not okay for humans--his antipathy toward Social Darwinism. Finally, I love his habit of showing that older thinkers out of style were doing sensible things, William Jennings Bryan for example.
However, I think that Dennett's critique of the Spandrels is on the money, as is the critique of the attitude that the Spandrels article has created. Rampant adaptation stories are bad. But they take place against a background of adaptation thinking, some of which is necessary, testable, and well-reasoned.
Part three of Dennett's book is the worst, IMHO. In it, he articulates a theory of memetics, where ideas, or "memes," compete against each other in the environment of the human mind and culture to produce our thoughts and consciousness. This very radical view creates problems for his entire project.
To see the problem, compare his views here with Steven Pinker's ideas. To Pinker, the human mind is a product of evolutionary design whose function is the processing of ideas. Allowing ideas to bump around and evolve in your brain and social circle would take a long time, time that you don't have. One thing good about Pinker's position is that he can argue that human minds do strive toward Truth, even if it is a Herbert Simon type of bounded rationality. You can really care about the veracity of ideas, and not just their utility, in a competition with other ideas. He can then defend his own project of seeking the true condition of the mind and human nature. He can say, hey look, I am just seeking the Truth like other thinkers before me.
Dennett's position is fundamentally awkward. He sees all ideas as memes competing in meme design space for more mental territory. So Truth is peripheral to the success of an idea. Our minds are a mafia of computer viruses that either let in other ideas that they like, or are infiltrated by unfriendly ones they cannot stop. So his claim that biological evolution is True and that all Design is from the bottom up slips around on philosophical relativism. If all ideas are selfish memes, then there is no Truth, even limited truth.
So, within Dennett's theory of the mind, Dennett's own ideas about natural selection may or may not be near the truth of the situation. It doesn't matter. Their reproductive success is all that counts. So one could argue that according to Dennett, there is no reason to listen to Dennett. If it gets you anywhere, it won't necessarily be because it successfully explains reality. But paradoxically, that would only be the case if his ideas were true. If his ideas are true, then his ideas would be very complicated mental organisms that want to squat in your mind like all the other ideas of the world, but they would, by happenstance, be true ones. For a philosopher, he has painted himself into quite a philosophical corner.
Dennett would argue that he has found meaning rising up from below: no taking without mistaking, no easy point to divide intentional design from non-intentional order, etc. That's fine. He assumes a constant truth in the world in that the physical laws stay the same in any one universe for it's duration and provide a context for evolution of order, design, meaning and thought. That's fine. But by claiming that all ideas are selected for reasons other than their truth value, he calls his whole programme into question. His whole book could be just another pick-up line in the single's bar of thought.
SIDEBAR: One tee-shirt that I find very amusing has an underwater scene. One creature was labelled "Darwinism." A virile shark had the creature in its teeth. That shark was labelled "Christianity." The image was entitled: "Survival of the fittest." The upshot is that the wearer was claiming that Christianity (his version) would triumph over Darwinism, not because Darwinism was false, but precisely because a memetic version of Darwinism was true. The unintended irony was palpable.
Just as Pinker dispatched the fear of Newspeak in his "Language Instinct," he dispatches this radically relativistic memology in "How the Mind Works" with a ray of hope that, even if truth for him does not come from either the God of Israel or the God of Aristotle, there at least is a Truth for one to seek. He brings it back to Darwin's original observation: there is a pre-existence of the soul, but for "pre-existence," read "monkeys."
Dennett had read Pinker's earlier work, and toward the end of "Darwin's Dangerous Idea," Dennett specifically challenges Pinker's theory of the mind, without naming him directly. When you look at the scientific work that Pinker draws on for his theory, versus Dennett's more vague handwaving, I think that Pinker carries the day. And the implications of Pinker's work strongly limit Dennett's fond wishes for memetics. Dennett knows this, and disagrees with Pinker as to the extent to which structures in the brain constrain how and what we think. For an author of books about consciousness, Dennett's concept of the mind is quite simplistic.
In the review in the journal *Evolution*, Allen Orr points out that Dennett really wants to make memetics a science like genetics, but that the nature of memes, or information in the mental/cultural sphere effectively prevents that sort of science of geneology. Dennett admits as much at the end of his memetics chapter. This gives his own work, where he himself has invested so much of his effort, an anticlimax.
Dennett's extreme suggestions for religion at the end of the book stem as much from his metaphysical naturalism as from his memetics. It only helps fan extremism for both supporters and opponents. One has to keep in mind that he is a philosopher more than a scientist. It only makes the third section more unfortunate, even though that seems to be where his heart is.
All in all: an important though flawed and very human book.