Darwin's Dangerous Idea (35 page)

keeps on changing in response to new challenges! If my point were to crown Darwin's great insight was that all the designs in the biosphere could be the Darwin as author and hero, there would be merit to this suspicion, but of products of a process that was as patient as it was mindless, an "automatic"

course this is not primarily such an exercise of intellectual history. It doesn't and gradual lifter in Design Space. In retrospect, we can see that Darwin really matter to my main thesis whether Darwin himself even existed! He himself could hardly have imagined, let alone supported with evidence, the could be, like the Average Taxpayer, a sort of mythical Virtual Author, for refinements and extensions of his idea that have permitted later Darwinians all I care. (Some authorities place Homer in that category.) The actual to go beyond his own cautious agnosticism about the origins of life itself, and historical man does fascinate me; his curiosity, integrity, and stamina inspire even the "design" of the physical Order his idea presupposed. He was in no me; his personal fears and flaws make him lovable. But he is, in a way, beside better position to characterize that Order than he was to describe the the point. He had the good fortune to be the midwife for an idea that has a constraints and powers of the hereditary mechanism; he just knew there had life of its own, precisely because it
does
grow and change. Most ideas can't to be such a mechanism, and it had to exploit the Order, whatever it was, that do that.

made "descent with modification" not only possible but fruitful.

In fact, a great deal of rhetoric has been expended by partisans on both The century-plus of subsequent focusing and extending of Darwin's great sides of the controversies about whether Darwin himself—St. Charles, you idea has been punctuated by controversy, amply illustrating, by the way, the might call him—was a gradualist, an adaptationist, a catastrophist, a capital-ist, a feminist. The answers to these questions are of considerable historical 190 BIOLOGY IS ENGINEERING

Darwin Is Dead
—
Long L ive Darwin!
191

interest in their own right, and if they are carefully divorced from questions cremental
hill-climbing
(or "gradient ascent") has been deservedly popular. It of ultimate justification, they can actually help us see what the scientific has even been popular enough to motivate theorists to calculate its issues really are. What various thinkers
think
they are doing—saving the limitations, which are severe. For certain classes of problems—or, in other world from one ism or another, or finding room for God in science, or words, in certain types of landscape—simple hill-climbing is quite impotent, combating superstition—often turns out to be at right angles to the contri-for an intuitively obvious reason: the climbers get stuck on local second-rate bution their campaigns actually succeed in making. We have already seen summits instead of finding their way to the global summit, the Mount Ever-instances of this, and more are in the offing. Probably no area of scientific est of perfection. (The same limitations beset the method of simulated research is driven by more hidden agendas than evolutionary theory, and it annealing.) The Local Rule is fundamental to Darwinism; it is equivalent to certainly will help to expose them, but nothing follows directly from the fact the requirement that there cannot be any intelligent (or "far-seeing" ) fore-that some people are trying desperately—whether they realize it or not—to sight in the design process, but only ultimately stupid opportunistic exploi-protect something evil or destroy something evil. People sometimes get it tation of whatever lucky lifting happens your way.

right in spite of having been driven by the most unpresentable hankerings.

What Eigen has shown is that this simplest Darwinian model of steady Darwin was who he was, and thought what he thought, warts and all. And improvement up a single slope of fitness to the optimal peak of perfection now he is dead. Darwinism, on the other hand, has more than nine lives. It just doesn't work to describe what goes on in molecular or viral evolution.

bids fair to being immortal.

The rate of adaptation by viruses ( and also of bacteria and other pathogens) is measurably faster than the "classical" models predict—so fast that it seems to involve illicit "look-ahead" by the climbers. So does this mean that 2. D

Darwinism must be abandoned? Not at all, for what counts as local depends ARWIN IS DEAD—LONG LIVE DARWIN!

(not surprisingly) on the scale you use.

Eigen draws our attention to the fact that when viruses evolve, they don't I have taken the section title from the title of the "Resume" with which go single-file; they travel in huge herds of almost identical variants, a fuzzy-Manfred Eigen ends his 1992 book. There is an unmistakable engineering edged cloud in the Library of Mendel that Eigen calls a "quasi-species." We flair to Eigen's thinking. His research is a sequence of biological construction problems posed and solved: how do the materials get amassed at the building already saw the unimaginably large cloud of
Moby Dick
variants in the site, and how does the design get determined, and in what order are the Library of Babel, but any actual library is likely to have more than one or various parts assembled so that they don't fall apart before the whole two variant editions of a book on its shelves, and in the case of a really structure is completed? His claim is that the ideas he presents are revolu-popular book like
Moby Dick
it is also likely to have multiple copies of the tionary, but that after the revolution, Darwinism is not only alive and well, same edition. Like actual
Moby Dick
collections, then, actual viral clouds but strengthened. I want to explore this theme in more detail, since we will include multiple identical copies but also multiple copies of minor typo-see other versions of it that are nowhere near as clearcut as Eigen's.

graphical variants, and this fact has some implications, according to Eigen, What is supposed to be revolutionary about Eigen's work? In chapter 3 we that have been ignored by "classical" Darwinians. It is the
shape
of the cloud looked at a fitness landscape with a single peak, and saw how the Baldwin of variants that holds the key to the speed of molecular evolution.

Effect could turn a well-nigh-invisible telephone pole on a plain into Mount A classical term among geneticists for the canonical version of a species Fuji, with a steadily rising surrounding slope, so that no matter where in the (analogous to the canonical text of
Moby Dick
) is the
wild type.
It was often space you started, you would eventually get to the summit if you simply supposed by biologists that among the many different genotypes in a pop-followed the Local Rule:

ulation, the pure wild type would predominate. Analogous would be the claim that in any library collection of copies
of Moby Dick,
most copies will Never step down; step up whenever possible.

be of the received or canonical edition—if there is one! But this doesn't have to be the case for organisms any more than for books in libraries. In fact, the The idea of a fitness landscape was introduced by Sewall Wright (1932), wild type is really just an abstraction, like the Average Taxpayer, and a and it has become a standard imagination prosthesis for evolutionary the-population may contain no individuals at all that have exactly "the" wild-type orists. It has proven its value in literally thousands of applications, including genome. (Of course, the same is true of books—scholars might debate for many outside of evolutionary theory. In Artificial Intelligence, economics, years over the purity of a particular word in a particular text, and until such and other problem-solving domains, the model of problem-solving by in-debates were resolved, nobody could say exactly what the ca-192 BIOLOGY IS ENGINEERING

Darwin Is Dead
—
Long Live Darwin!
193

nonical or wild-type text of that work was, but the identity of the work would leading—usually—from one temporarily stable problem-setting to another.

hardly be in jeopardy. James Joyce's
Ulysses
would be a good case in point.) No sooner do you climb a peak than the whole landscape pitches and billows Eigen points out that this distribution of the "essence" over a variety of into a new mountain range and you start climbing all over again. In fact, the nearly identical vehicles turns out to make that essence much more movable, landscape is constantly shifting under your feet (if you are a quasi-species of much more adaptable, especially in "rugged" fitness landscapes, with viruses ).

multiple peaks and few smooth slopes. It permits the essence to send out Now, this is really not as revolutionary as Eigen claims. Sewall Wright efficient scouting parties into the neighboring hills and ridges, ignoring himself, in his "shirting balance theory," tried to explain how multiple peaks wasteful exploration of the valleys, and thereby vastly (not Vastly, but and shifting landscapes would be traversable not by individual "wild-type"

enough to make a huge difference) enhancing its capacity to find higher exemplars, but by various-sized populations of variants, and Ernst Mayr has peaks, better optima, at some distance from its center, where the (virtual) stressed for many years that "population thinking" is at the heart of Dar-wild type sits.1

winism, something overlooked by geneticists at their peril. So Eigen has The reasons it works are summarized by Eigen as follows: really not revolutionized Darwinism but, rather—no small contribution—

created some theoretical innovations that clarify and strengthen underap-Functionally competent mutants, whose selection values come close to preciated and imperfectly formulated ideas that had been around for years.

that of the wild type (though remaining below it), reach far higher pop-When Eigen (1992, p. 125) says, "The (quantitative) acceleration of evolution ulation numbers than those that are functionally ineffective. An asymmetric that this brings about is so great that it appears to the biologist as a surprising spectrum of mutants builds up, in which mutants far removed from the new
quality,
an apparent ability of selection to 'see ahead', something that wild type arise successively from intermediates. The population in such a would be viewed by classical Darwinians as the purest heresy!" he is chain of mutants is influenced decisively by the structure of the value indulging in a familiar form of overdramatization, ignoring the many landscape. The value landscape consists of connected plains, hills, and biologists who at least anticipated, and perhaps even fomented, his "revo-mountain ranges. In the mountain ranges, the mutant spectrum is widely lution."

scattered, and along ridges even distant relatives of the wild type appear After all, when traditional Darwinian theorists postulate fitness landscapes with finite [that is, not infinitesimal] frequency. It is precisely in the moun-and then randomly sprinkle genotypes on them in order to calculate what tainous regions that further selectively superior mutants can be expected.

As soon as one of these turns up on the periphery of a mutation spectrum theory says would happen to them, they know that, in nature, genotypes don't the established ensemble collapses. A new ensemble builds up around the just get thrown randomly into pre-existing parts of the world. Every model of superior mutant, which thus takes over the role of the wild type___ This a time-consuming process has to start at some arbitrary "moment"; the curtain causal chain results in a kind of 'mass action',
by which the superior mu-rises and the model then plots what happens next. If we look at such a model
tants are tested with much higher probability
than inferior mutants, even and see that at the "outset" it shows a bunch of candidates down in the if the latter are an equal distance away from the wild type. [Eigen 1992, valleys, we can be pretty sure that the theorist recognizes that they weren't p. 25.]

"always" down there—whatever that would mean! Wherever on the fitness landscape there are candidates at one time, there were peaks before, or those So there is a tight interaction between the shape of the fitness landscape candidates wouldn't be there, so these must be relatively
new
valleys these and the population that occupies it, creating a series of feedback loops, candidates are occupying, a new predicament that evolution has placed before them. Only that assumption could justify locating the candidates in the valleys in the first place. Eigen's contribution reinforces the appreciation that we have to add these complications to the models if we want them actually to 1. The similarity between these themes and the themes I develop in
Consciousness
do the work that Darwinians have always supposed that their simpler models
Explained
(1991a) about the need to break up the Cartesian Theater, with its Central could do.

Meaner, and distribute its intelligence work around to a variety of peripheral agents, is of course no accident. It is, however, mainly a case of convergent evolution, so far as I can It is certainly no accident that our appreciation of the need for these much determine. I had not read any of Eigen's work at the time I was writing my book, though more complicated models coincides in time (almost down to the month, and it certainly would have inspired me if I had. A useful bridge between Eigen on molecules certainly to the year) with our capacity to build and explore such models on and me on consciousness is Schull 1990 on the intelligence of species, and my commen-existing computers. No sooner do more powerful computers become tary, Dennett 1990a.

available than we discover with their help that more complex 194 BIOLOGY IS ENGINEERING

Function and Specification
195

models of evolution are not only possible but positively required if we are viruses to mutate in deadly combat with modern medicine that spurs on and really to explain what Darwinism has always claimed it can explain. Darwin's funds much of this research. (The AIDS virus has undergone so much mu-idea that evolution is an algorithmic process is now becoming an ever more tation in the last decade that its history over that period exhibits more genetic enriched family of hypotheses, undergoing its own population explosion diversity—measured in codon revisions—than is to be found in the entire thanks to the opening up of new environments for it to live in.