Darwin’s Dangerous Idea
Evolution and the Meanings of Life (1995)
by Daniel C. Dennett (1942-)
Dennett’s purpose in this book is to formally refute the many objections to Darwinism expressed by those who have given the subject less thought than he has. In this sense it is an unfair battle, and he naturally has the best of it. Even as a sympathetic reader, I feel bad for his targets (though to be fair, they fired first). Along the way, he treats the Darwinian aspects of memetics, though he seems skeptical of a real theory of memetics.
Dennett introduced an interesting convention: each chapter ends with a one-paragraph summary of its own content and a one-paragraph preview of the next.
The primary theme is that evolution (of any kind) is a tree or graph in a design space, and each move can be seen as a step across a fitness landscape, with varying height at each (multi-dimensional) position. He casts most objections to Darwinism as wishing for a “skyhook” (or the requirement of a skyhook) that would lift organisms up to some height of the fitness landscape, in a way impossible to reach by ordinary small steps. In all such cases, he prefers to identify (or expects that someone else will identify) a “crane” to perform the required lifting.
Part I opens with a quote from W. V. O. Quine, relevant to my own ambitions:
Analyze theory-building how we will, we all must start in the middle. Our conceptual firsts are middle-sized, middle-distanced objects, and our introduction to them and to everything comes midway in the cultural evolution of the race. In assimilating this cultural fare we are little more aware of a distinction between report and invention, substance and style, cues and conceptualization, than we are of a distinction between the proteins and the carbohydrates of our material intake. Retrospectively we may distinguish the components of theory-building, as we distinguish the proteins and carbohydrates while subsisting on them.
In chapter 7, he discusses Conway’s Life as an example of a cellular automata simulation. In a note, he recommends The Planiverse by A.K. Dewdney.
In chapter 8 he discusses the work of Stuart Kauffman, who said
Adaptive evolution is a search process – driven by mutation, recombination, and selection – on fixed or deforming fitness landscapes. An adapting population flows over the landscape under these forces. The structure of such landscapes, smooth or rugged, governs both the evolvability of populations and the sustained fitness of their members. The structure of fitness landscapes inevitably imposes limitations on adaptive search.
The populations and the landscape both evolve: once a population includes a feature, that feature becomes part of the landscape for all other populations that interact with the first, both descendants and those evolving on parallel branches. For instance, when the early molecular “macros” (Dennett’s term for pre-living complexes of mutually replicating molecules) hit on the use of amino-acid chains and nucleic-acid chains, the region of the fitness landscape that included sources of these molecules (or molecules compatible with them) became much “lower” than other regions; thus it was easier for populations to “flow” onto these parts of the fitness landscape. Similarly, during the period prior to the Cambrian explosion (just before body parts became hard enough to make good fossil impressions) it seems likely that a small number of relatively small innovations could have created a drastically lowered region in the fitness landscape that would be manifested in the explosion itself.
Also in chapter 8, Dennett quotes Stanislaw Ulam on the influence of constraints, this time in poetry.
When I was a boy I felt that the role of rhyme in poetry was to compel one to find the unobvious because of the necessity of finding a word which rhymes. This forces novel associations and almost guarantees deviations from routine chains or trains of thought. It becomes paradoxically a sort of automatic mechanism of originality.
Kauffman has proposed the principle that “local rules generate global order” in the context of biological engineering/evolution. Dennett points out that this principle is not found in human engineering (for instance the pyramids were built by harnessing a hierarchical executive/management structure), and so is not typically recognized as a principle of engineering at all. He proposes a reformulation: “Until you manage to evolve communicating organisms that can form large engineering organizations, you are bound by the following Preliminary Design Principle: all global order must be generated by local rules. So all the early products of design, up to the creation of something with some of the organizational talents of Homo sapiens, must obey whatever constraints follow from the “management decision” that all order must be accomplished by local rules.”
In chapter 10, Dennett takes on Stephen Jay Gould. Dennett finds that Gould’s criticisms of certain aspects of modern Darwinism have been grasped at by critics of all of Darwinism, and used to dismiss out of hand any attempt to use evolutionary arguments. Therefore he takes a lot of time to identify Gould’s actual criticisms, and to refute them. In the process he proposes a plausible reason for Gould’s attitude, which is a desire for skyhooks.
Part III is titled Mind, Meaning, Mathematics, and Morality. Chapter 12 is titled The Cranes of Culture. In this chapter Dennett discusses memes as the (symbolic?) mental objects on which human minds construct culture. It seems that Dennett uses the word meme as most people do: as a convenient label for something not very well defined, but suggestive in a convenient way. However, he does give a few examples. For instance, we all filter the information flowing into our minds, and one filter (meme) could be expressed as
Ignore everything that appears in X
There are evidently people for whom X is Dennett, and some for whom X is National Geographic, Pravda, or The New York Review of Books. Another example is the term “Artificial Intelligence”. Section 3 of this chapter is titled Could There Be a Science of Memetics?, and Dennett seems to doubt it. In expressing his doubt that there can be a close analogue with genetics, he says “The obvious problem noted by all is that it is very unlikely – but not quite impossible – that there is a uniform “brain language” in which information is stored in different human brains, and this makes brains very different from chromosomes. … The meme is primarily a classification not a syntactic classification that might be directly observable in “brain language” or natural language.” Dennett mentions Dan Sperber as one “who has thought a great deal about cultural evolution”. Although Dennett doesn’t mention this directly, he seems to prefer the epidemiological conception of memetics, as opposed to the genetic conception of memetics.
He mentions Sperber’s observation that people remember stories better than they remember text. (He also refers to Albert Lord’s The Singer of Tales, study of the modern era bards of the Balkans. I intend to look into this work.)
Dennett’s target in chapter 15, The Emperor’s New Mind, and Other Fables, is Roger Penrose. Penrose tries to combine Gödel’s Theorem and some speculation about quantum gravity to postulate a skyhook for consciousness. Dennett refutes him, primarily on the basis that Penrose concentrated on a small subset of possible algorithms, and ignored the much larger set that might be cranes. In discussing this type of argument, Dennett capitalizes the word “Vanishingly” and uses “Vast” to mean “very-much-more-than-astronomical”. These usages emphasize the sizes of numbers that are very far from intuitive.
In chapter 16, On the Origin of Morality, he quotes E.O. Wilson: “The genes hold culture on a leash. The leash is very long, but inevitably values will be constrained in accordance with their effects on the human gene pool.” Dennett responds:
But all this means (unless it is just false) is that, in the long run, if we adopt cultural practices that have disastrous effects on the human gene pool, then the human gene pool will succumb. There is no reason to think, however, that evolutionary biology shows us that our genes are powerful enough, and insightful enough, to keep us from making policies quite antithetical to their interests. … That is our transcendence, our capacity to “rebel against the tyranny of the selfish replicators,” as Dawkins says, and there is nothing anti-darwinian or antiscientific about it.
He discusses briefly the effort by Tooby and Cosmides to replace the “Standard Social Science Model”, an effort in which Steven Pinker has also participated. He suggests that they have gone too far, and proposes his own “Only Slightly Nonstandard Social Science Model”. Here are the two statements of precepts:
Whereas animals are rigidly controlled by their biology, human behavior is determined by culture, an autonomous system of symbols and values. Free from biological constraints, cultures can vary from one another arbitrarily and without limit. . . . Learning is a general-purpose process, used in all domains of knowledge. [Pinker on the SSSM]
Whereas animals are rigidly controlled by their biology, human behavior is largely determined by culture, a largely autonomous system of symbols and values, growing from a biological base, but growing indefinitely away from it. Able to overpower or escape biological constrains in most regards, cultures can vary from one another enough so that important portions of the variance are thereby explained. . . . Learning is not a general-purpose process, but human beings have so many special-purpose gadgets, and learn to harness them with such versatility, that learning often can be treated as if it were an entirely medium-neutral and content-neutral gift of non-stupidity. [Dennett’s italics, my bolding highlighting the differences between the two statements]
I have not seen any references to a formulation of the SSSM expressed by those supposed to believe it.
In chapter 17, Redesigning Morality, he says:
Human culture, religion in particular, is a repository of ethical precepts, ranging from the Golden Rule, the Ten Commandments, and the Greeks’ “Know Thyself” to all manner of specific commands and prohibitions, taboos, and rituals. Philosophers since Plato have attempted to organize these imperatives into a single rationally defensible and universal system of ethics, so far without achieving anything approaching consensus. Mathematics and physics are the same everywhere, but ethics has not yet settled into a similar reflective equilibrium. Why not? Is the goal illusory? Is morality just a matter of subjective taste (and political power)? Are there no discoverable and confirmable ethical truths, no forced moves or Good Tricks?
He goes on to analyze various failings of the philosophy of ethics, which mostly boil down to a failure to recognize the satisficing (Herbert Simon’s term) nature of rational inquiry with limited resources. He has treated this subject further in a 1988 article The Moral First Aid Manual.
As usual, I find Dennett to be a persuasive writer, but sometimes dull. This is because his approach is meticulous and intended to stand up to the kind of scrutiny and criticism he applies to others. For a non-specialist, it can be tedious and sometimes impenetrable. Nonetheless, I found it worthwhile in the absence of a less-specialized treatment of the issues.