In his 1995 book Darwin’s Dangerous Idea: Evolution and the Meanings of Life, the philosopher Daniel C. Dennett explores the numerous intellectual implications of Darwinism, the concept of evolution by natural selection. I use the word intellectual intentionally, because Dennett emphasizes that Darwinism impacts numerous fields outside its native habitat of biology, and even outside of science. It is, he argues, also essential to progress in the studies of culture, language, meaning, and morality, and closely linked to progress in philosophy, computer science, and cognitive studies.
This is a bold claim, but it is not Dennett’s boldest. That place is reserved for his reason why evolution must be central to a proper understanding of the world. Evolution, argues Dennett, is a set of algorithms that, like all algorithms, share three important traits: “guaranteed results,” “underlying mindlessness,” and “substrate neutrality”—indifference to the media used to implement the algorithm. The idea that life is not only the result of algorithms but can actually be best understood algorithmically is thus the dangerous idea of Dennett’s title. The danger stems from evolution’s ability to destroy (or, to use a less loaded term, replace) traditional ideas much like a hypothetical but unstoppable “universal acid” would “eat through anything.”
At the beginning of his book Dennett expresses some apprehension about the demise of so many time-honored ideas: “I, too,” he writes, “cherish many of the ideas and ideals that it [Darwinism] seems to challenge, and want to protect them.” As the book progresses, though, this intellectual conservationism fades far into the background, and Dennett’s enthusiasm for the revolutionary project of Darwinism grows, but conservationism returns suddenly in the last chapter and closes the book with a rather muddled ending. I will return to the subject of Dennett’s poor ending after exploring the bulk of his book, which sets much higher standards.
Part I: Starting in the Middle
The first third of Darwin’s Dangerous Idea introduces the language and some of the metaphors that Dennett will use in the rest of the book. Even if read independently of the rest of the book, it demonstrates how one might go about creating a cohesive philosophy of evolution. It also introduces the central concept of evolution as a design process. Dennett’s emphasis on—one might say obsession with—the concept of design is demonstrated by his early summary of Darwinism. “Give me Order,” he has Darwin say, “and time, and I will give you Design.”
This statement is what Dennett terms the “Principle of Accumulation of Design”: design could be the result of an unintelligent cause “that distributed that work over huge amounts of time.” This conception of evolutionary design sounds strange but is in fact sound. It possesses several notable strengths, which deserve elaboration beyond that given by Dennett.
First, this Principle removes a widening gap between our intuition and circumstantial evidence that the world is designed on one hand—the “design inference,” as William Dembski calls it—and our empirical and theoretical knowledge of evolutionary biology on the other. This requires us to give up the idea that only minds can be designers—Dennett spends some time exploring the history of this philosophical tradition in order to reject it—but this is necessary in order to reconcile our intuition that the world is designed with our understanding of biology.
Second, the idea that things other than minds can design things fits with our knowledge of the power of computers, a subject on which Dennett spends time later in his book. Whether or not some computers are or will be sophisticated enough to be considered minds, there are clearly computers that seem able to design things. With sufficiently complex programs computers at least mimic originality, and there is at least one situation in which it is appropriate to use the word design to describe the products of such electronic processes: if a computer program itself is self-modifying, the humans who programmed the original version become unable to understand their own software, and cannot fairly be credited with all of the design of something that clearly remains a designed object. In this situation, as with biological evolution, it makes sense to state that the process itself is a designer.
Darwin’s Dangerous Idea is at times a work of apologetics for what Dennett terms “orthodox neo-Darwinism,” and his section on reductionism is a good example of this. Dennett argues that there are three types of reductionism and that only two of these types are bad. The good type he terms the “bland sense” of reductionism: the concept that various sciences describe the same phenomena at different levels of detail. There is also a “preposterous sense” of reductionism, the claim that the most complex phenomena such as human politics are nonetheless best explained by “lower-level” sciences such as physics. In between is “greedy reductionism,” a cavalier way of doing science caused by the “zeal to explain too much too fast.”
Part II: Darwinian Thinking in Biology
After some insightful and relatively thorough descriptions of Darwinian concepts such as the Tree of Life and philosophical concepts such as kinds of possibility, Darwin’s Dangerous Idea undergoes a major but gradual transition into its second part. It begins with a dense but understandable explanation of molecular evolution and a less dense explanation of the subtleties of John Conway’s Game of Life, the best-known and arguably most interesting of many “cellular automata.” Life is itself a simple algorithm for changing the binary states of a grid of square cells, but Life grids can develop into very complex systems. Indeed, Life is interesting in part because it is possible to build any logical operator in its two-dimensional grid, and to connect these operators to make a universal Turing machine. Life, in other words, can not only act as a computer, it can emulate any computer.
Dennett has a philosophical interest in Conway’s work, and makes a clever point: the creation of the rules of Life was a combination of design and discovery. “It took more than a year for [Conway and his students] to find the simple Life Physics rule in the Vast space of possible rules,” but they found it as much as invented it. They did, in other words, the same thing that mindless evolutionary processes do as they design life: they tried many things, and found a few that worked well. Dennett’s point is that physical laws, whether those of the Life world or of our world, can be evolved through “a purely algorithmic Darwinian process of world-trying.” This is Dennett at his most speculative but also at his most fun, and even here he is careful to describe influences on his ideas, in this case David Hume and the physicist Lee Smolin.
The next chapter discusses the role of engineering in biology, a role that Dennett argues is dominant because “there is just one Design Space,” one vast (or Vast, to imitate Dennett’s special usage) set of all possible designs, biological and otherwise, through which both biological evolution and human engineers must navigate in order to design their products. It is here that Dennett’s defense of functionalism begins, a defense that runs through the rest of his book. The first deeply interesting argument for functionalism occurs 31 pages into the chapter, though, after Dennett has already defended his assertion that “reverse engineering” must be central to biology: “When you ask functional questions about anything—organism or artifact—you must remember that it has to come into its current or final form by a process that has its own requirements, and these are exactly as amenable to functional analysis as any features of the end state.” Functionalism properly done does not claim that all features have functions but merely that all features have reasons—often historical reasons, but reasons nonetheless.
Dennett also defends a related approach to biology, adaptationism—the idea that organisms adapt optimally to specific environmental challenges—, and it is here that he really makes the second part of his book fit his earlier description of it; “Part II,” said Dennett’s very first section, “examines the challenges to Darwin’s idea—to neo-Darwinism and the Modern Synthesis—that have arisen within biology itself.” This means both apologetics and polemics on Dennett’s part, for this is not only an examination of challenges to evolution as an algorithm but a defense of several related ideas and a potent assault on those who challenge them.
Those individuals include most notably Stephen Jay Gould, whom Dennett challenges for his opposition to adaptationism, sociobiology, and extrapolationism, as well as for his claim that the “hypothesis of punctuated equilibrium overthrew orthodox neo-Darwinism.” This attack on Gould has virtues, but also exhibits an unwarranted level of viciousness: “Anybody as prolific and energetic as Gould,” writes Dennett, “would surely have an agenda beyond that of simply education and delighting his fellow human beings about the Darwinian view of life.” Gould did have an agenda, of course, but it seems a substantial portion of it was that which Dennett dismisses. Furthermore, it is certainly possible—the odds against it are not Vast, one might say—that someone could be as prolific and energetic for purely altruistic reasons.
Dennett raises the truly fascinating question of Gould’s motivations again two pages later only to decide that it is too big a subject to tackle. He opts instead to demonstrate that Gould “is opposed to the very idea that evolution is, in the end, just an algorithmic process.” His most successful argument, though, is that Gould has repeatedly overstated the revolutionary nature of his research, a substantially more modest claim.
Nonetheless, Dennett’s critique of an influential 1979 paper by Gould and Richard Lewontin, “The Spandrels of San Marco and the Panglossian Paradigm: A Critique of the Adaptationist Programme,” is an enjoyable textual analysis and demonstrates Dennett’s ability to rival Gould at his own game of building bridges between evolutionary biology and the arts. It may well be, though, that both Dennett and those he critiques fail to build a strong bridge, for their analogies of spandrels demonstrate two different misunderstandings of the construction of large domes. In light of later criticism concerning the true “structural demands of support for a large dome over arches,” Dennett’s analysis seems more correct than Gould and Lewontin’s, and his defense of adaptationism is a good one, but rethinking the section to accommodate accurate architecture is a challenge.
It is Dennett’s criticism of Gould’s views on contingency, though, that seems truly flawed. He deals here with Gould’s claim “that ‘the most common misunderstanding of evolution, at least in lay culture,’ is the idea that ‘our eventual appearance’ is ‘somehow intrinsically inevitable and predictable within the confines of the theory.’” Dennett describes this as a cryptic statement and is uncertain who “us” is in Gould’s framework. There are, Dennett writes, a few possibilities: “us” could be specific individuals alive today, in which case Gould is right, or “something very general, such as ‘air-breathing, land-inhabiting vertebrates,’” in which case Gould is probably wrong, or perhaps “something intermediate, such as ‘intelligent, language-using, technology-inventing, culture-creating beings.’” Dennett concludes that Gould’s meaning was the last of these and soon forms the “tentative conclusion” that “Gould thinks his thesis of radical contingency would refute the core Darwinian idea that evolution is an algorithmic process.” Gould may well have thought this, but it doesn’t follow from his statement about “the most common misunderstanding of evolution,” which I strongly suspect describes the belief that the evolution of humans—Homo sapiens—was inevitable.
Part III: Mind, Meaning, Mathematics, and Morality
The third part of Darwin’s Dangerous Idea, which discusses memetics, linguistics, sociobiology, and other relationships between mind and evolution, is as mixed in content as the second. It begins with a strong argument for the importance of memes to a valid evolutionary understanding of human culture. Much of this argument builds up to the claim that humans are uniquely cultural beings—that our ability to use language distinguishes us from other animals. Dennett presents a good model for understanding a hierarchy of ways of thinking, but his arguments again seem less solid when he goes on the offensive. The extreme of this is his apparent proof that consciousness and free will can be explained—it hinges on the idea that “the best statement… of the solution of the problem of free will” is a genuinely good statement, ignoring the possibility that free will is a true mystery, the very possibility Dennett is trying to disprove.
Dennett’s arguments against Noam Chomsky’s anti-adaptationism in linguistics echo his earlier advocacy of an engineering perspective in biology, but his originality returns with a chapter on evolution and meaning. Much of this chapter is devoted to a rather successful philosophical argument that machines can understand meaning, that “human meanings are just as derived” as those of machines, and that meaning can thus emerge from meaninglessness in a Darwinian fashion. This argument involves a handful of thought experiments that amount to well-told, if very simple, stories about the concept of meaning.
Dennett’s last few chapters play out the ideas of reductionism in the realm of the human mind, and are usually insightful but do not demand further commentary. There is one exception, though: Dennett’s last chapter takes his poor understanding of the human phenomenon of religion, which he has only briefly displayed before, and sets it in the forefront of his writing. Most of this chapter consists of a section entitled “In Praise of Biodiversity,” which draws a parallel between biodiversity and cultural diversity and focuses on the latter. Dennett asserts that “the Design Space perspective… helps to explain our intuition that uniqueness or individuality is ‘intrinsically’ valuable.” Unique objects indeed contain information that cannot be found elsewhere, but Dennett states that this is valuable without explaining to whom. Some information—most information—is simply not useful to humans, and it is sadly impossible to keep everything. “Among the precious artifacts worth preserving are whole cultures themselves,” writes Dennett, but without argument. Dennett seems incapable of saying much about cultural diversity except in the realm of religion, though, and seem incapable of saying much about religion without polemics.
Dennett earlier expressed opposition to Gould’s views on science and religion (which were then less clear than they later became; Gould had not yet published Rocks of Ages), and held that “evolutionists who see no conflict between evolution and their religious beliefs have been careful not to look as closely as we have been looking.” To say that Dennett has little respect for religion would be an understatement: he believes that religion is dying (except for fundamentalism, which is experiencing at “tremendous rebirth”), that this end of religion is largely a good thing, and that the religiosity that remains ought to be kept at arms length so that we can avoid its danger. His rhetoric here is sufficiently strong that Phillip Johnson takes literally his threat to put religions “in cages,” and it’s genuinely hard to tell how metaphorical this statement is meant to be. This is a strange ending for a book entitled Darwin’s Dangerous Idea; the conclusion says nothing about the dangers of Darwinism for tradition and everything about the dangers of tradition for humanity. Dennett demonstrates that his breed of Darwinian thinking, for all the rigorous argument and brilliant science behind it, is indeed a universal acid, and that he is himself part of this acid, eating away at the traditional ideas that surround him.
- Daniel C. Dennett, Darwin’s Dangerous Idea: Evolution and the Meanings of Life (1995).
- Dennett, “Errors in Darwin’s Dangerous Idea,” (1997), <http://ase.tufts.edu/cogstud/papers/errors.html>.
- Phillip E. Johnson, “Daniel Dennett’s Dangerous Idea,” The New Criterion, October 1995, and Access Research Network, 1996, <http://arn.org/docs/johnson/dennett.htm>.