Authors: Daniel C. Dennett

Darwin's Dangerous Idea (60 page)

(3.5 billion instead of 2.5 billion) than orthodoxy has recently supposed. If confirmed, The reason orthodoxy prefers to assume a birthplace on Earth is that this is this would drastically revise a lot of standard assumptions about the intermediate dead-the simplest and most scientifically accessible hypothesis. That does not lines, giving more time for the evolution of advanced forms ("Whew!"), but only by reducing the time available for the process of molecular evolution to get all the way to make it true. Whatever happened, happened. If Hoyle is right, then (darn it) microbes ("Uh oh!").

we will find it much harder to confirm or disconfirm any detailed hypoth-A Clutch of Harmless Heresies 317

316 CONTROVERSIES CONTAINED

are the most powerful cranes we have yet discovered. Are there designs that process, and the rationale of the design aimed for, actually did play a role in simply could not be erected without the help of this particular crane? If there the process. In these cases, the goal was explicitly represented in the minds of are designs that cannot be approached by a gradual, stepwise redesign process the breeders who did the selecting. So the theory of evolution must allow for in which each step is at least no worse for the gene's survival chances than its the existence of such products, and such historical processes, as special predecessor, then the existence of such a design in nature would seem to cases—organisms designed with the help of supercranes. Now the question require, at some point in its ancestry, a helping hand from a foresightful arises: can such special cases be distinguished in retrospective analysis?

designer—either a gene-splicer, or a breeder who somehow preserved the Imagine a world in which
actual
hands from another galaxy supplemented necessary succession of intermediate backsliders until they could yield their the "hidden hand" of natural selection. Imagine that natural selection on this sought progeny. But could we ever conclusively establish that some design planet was aided and abetted over the eons by visitors: tinkering, farsighted, had this feature of
requiring
such a saltation in its ancestry? For over a reason-representing organism-designers, like the animal- and plant-breeders century, skeptics have hunted for such cases—thinking that, if they ever of our actual world, but not restricting themselves to "domesticated"

found one, it would conclusively refute Darwinism—but so far their efforts organisms designed for human use. (To make it vivid, we may suppose they have shown a systematic weakness.

treated Earth as their "theme park," creating whole phyla for educational or entertainment purposes.) These bioengineers would have actually formulated, Consider the most familiar example, the wing. Wings could not evolve in one and represented, and acted on, the rationales of their designs—just like fell swoop, runs the standard skeptical argument; and if we imagine—as we automobile engineers or our own contemporary gene-splicers. Then, let's Darwinians must—that wings evolved gradually, we must admit that partially suppose, they absconded. Now, would their handiwork be detectable by any completed wings would not only not have provided partial value but would imaginable analysis by biologists today?

have been a positive hindrance. We Darwinians need admit no such thing.

If we found that some organisms came with service manuals attached, this Wings that are good only for gliding (but not powered flight) have manifest net would be a dead giveaway. Most of the DNA in any genome is unexpressed—

benefits for many actual creatures, and still stubbier, less aerodynamically often called "junk DNA"—and NovaGene, a biotechnology company in Hous-effective protuberances could have evolved for some other reason, and then ton, has found a use for it. They have adopted the policy of "DNA branding": been exapted. Many versions of this story—and other stories—have been told writing the nearest codon rendering of their company trademark in the junk to fill in the gap. Wings are not an embarrassment to orthodox Darwinians, or if DNA of their products. According to the standard abbreviations for the they are, they are an embarrassment of riches. There are
too many
different amino-acid specifiers, asparagine, glutamine, valine, alanine, glycine, glu-plausible ways of telling the story of how functioning wings could have tamic acid, asparagine, glutamic acid = NQVAGENE ( reported in
Scientific
evolved by gradual increments! This shows how hard it would be for anyone to
American
June 1986, pp. 70-71). This suggests a new exercise in "radical devise an insurmountable argument to prove that a particular feature must have translation" (Quine I960) for philosophers: how, in principle or in practice, arisen by a saltation, but at the same time it shows that it would be just as hard could we confirm or disconfirm the hypothesis that trademarks—or service to prove that a feature must have arisen
without
a saltation, unaided by human manuals or other messages—were discernible in the junk DNA of any spe-or other intelligent hands.

cies? The presence of functionless DNA in the genome is no longer regarded Indeed, all the biologists I have queried on this point have agreed with me as a puzzle. Dawkins' ( 1976) selfish-gene theory predicts it, and elaborations that there are no sure marks of natural, as opposed to artificial, selection. In on the idea of "selfish DNA" were simultaneously developed by Doolittle and chapter 5, we traded in the concept of strict biological possibility and Sapienza (1980) and Orgel and Crick (1980) (see Dawkins 1982, ch. 9, for the impossibility for a graded notion of biological probability, but even in its details). That doesn't show that junk DNA
couldn't have
a more dramatic terms, it is not clear how one could grade organisms as "probably" or "very function, however, and hence it could have a meaning after all. Our imagined probably" or "extremely probably" the products of artificial selection. Should intergalactic interlopers could as readily have exapted the junk DNA for their this conclusion be viewed as a terrible embarrassment to the evolutionists in own purposes as the NovaGene engineers exapted it for theirs.

their struggle against creationists? One can imagine the headlines-. "Scientists Finding the high-tech version of "Kilroy was here" written in the genome Concede: Darwinian Theory Cannot Disprove Intelligent Design!" It would be of a cabbage or a king would be unnerving, but what if no such deliberate foolhardy, however, for any defender of neo-Darwinism to claim that clues were left around? Would a closer look at the organism designs them-contemporary evolution theory gives one the power to read history so finely selves—the phenotypes—reveal some telltale discontinuities? Gene-splicers from present data as to rule out the earlier historical pres-318 CONTROVERSIES CONTAINED

A Clutch of Harmless Heresies
319

ically coherent (however lame and
ad hoc)
for Darwinians to reply that what ence of rational designers—a wildly implausible fantasy, but a possibility they were being shown was telling evidence for the surprising hypothesis of after all.

intergalactic interlopers! The power of the theory of natural selection is not In our world today, there are organisms we
know
to be the result of the power to prove exactly how (pre)history was, but only the power to prove foresighted, goal-seeking redesign efforts, but that knowledge depends on our how it could have been, given what we know about how things are.

direct knowledge of recent historical events; we've actually watched the Before leaving this curious topic of unwelcome but nonfatal heresies, let's breeders at work. These special events would not be likely to cast any fossily consider one that is a bit more realistic. Did life on Earth arise just once, or shadows into the future. To take a simpler variation on our thought perhaps many times? Orthodoxy supposes it happened just once, but there is experiment, suppose we were to send "Martian" biologists a laying hen, a no skin off its back if in fact life arose twice or ten or a hundred times.

Pekingese dog, a barn swallow, and a cheetah and ask them to determine However improbable the initial bootstrapping event may have been, we must which designs bore the mark of intervention by artificial selectors. What not commit the Gambler's Fallacy of supposing that after it happened once, could they rely on? How would they argue? They might note that the hen did the odds rose against its happening again. Still, the question of how many not care "properly" for her eggs; some varieties of hen have had their instinct times life arose independently opens up some interesting prospects. If at least for broodiness bred right out of them, and would soon become extinct were it some of the assignments in the DNA are purely arbitrary, then might there not for the environment of artificial incubators human beings have provided not have been two
different
genetic languages coexisting side by side, like for them. They might note that the Pekingese was pathetically ill-equipped to French and English, only entirely unrelated? This has not been discovered—

fend for itself in any demanding environment they could imagine. But the DNA has clearly coevolved with its parent, RNA—but that does not yet show barn swallow's innate fondness for carpentered nest sites might fool them into that life
didn't
arise more than once, because we don't (yet) know how wide the view that it was some sort of pet, and whatever features of the cheetah the scope for variation in genetic code actually was.

convinced them that it was a creature of the wild might also be found in Suppose there were exactly two equally viable and constructible DNA greyhounds, and be features we know to have been patiently encouraged by languages, Mendelese (ours) and Zendelese. If life arose twice, there would breeders. Artificial environments are themselves a part of nature, after all, so be four equiprobable possibilities: both times Mendelese, both times Zen-it is unlikely that there are
any
clear signs of artificial selection that can be delese, Mendelese and then Zendelese, or Zendelese and then Mendelese. If read off an organism in the absence of insider information on the actual we ran the tape of life many times, and looked at the times in which life arose history that created the organism.

twice, we'd expect that half the time both languages would get created, but in Prehistoric fiddling by intergalactic visitors with the DNA of earthly spe-one quarter of those replays only Mendelese would appear. In those worlds, cies cannot be ruled out, except on the grounds that it is an entirely gratuitous the DNA language of all organisms would be the same, even though another fantasy. Nothing we have found (so far) on Earth so much as hints that such a language was just as possible. This shows that the "universality" (at least on hypothesis is worth further exploration. And remember—I hasten to add, lest our planet) of the DNA language does not permit a valid inference that all creationists take heart—even if we were to discover and translate such a organisms had arisen from a single progenitor, the ultimate Adam, since,
ex

"trademark message" in our spare DNA, or found some other uncontestable
hypothesi
in these cases, Adam could have had an entirely independent twin mark of early tampering, this would do nothing to rescind the claim of the of sorts, accidentally sharing the same DNA language. Of course, if life arose theory of natural selection to explain all design in nature without invocation many more times—say, a hundred times—under these same conditions, then of a foresighted Designer-Creator
outside the system.
If the theory of the likelihood of only one of the two equiprobable languages' appearing evolution by natural selection can account for the existence of the people at would plummet to Vanishing. And if in fact there are many more than two NovaGene who dreamt up DNA branding, it can also account for the equally usable genetic codes, this would similarly change the implications existence of any predecessors who may have left their signatures around for about probability. But until we know more about the range of genuine us to discover.

possibilities and their associated probabilities, we can't get any good leverage Now that we have seen this possibility, however unlikely it is, we also see to decide for sure that life arose just once. For the time being, it's the simplest that, if the skeptics had ever found their Holy Grail, the You-Couldn't-Get-hypothesis—life only
has
to have arisen once.

Here-from-There Organ or Organism, it would not have been
conclusive
against Darwinism after all. Darwin himself said that he would have to abandon his theory if such a phenomenon were discovered (see note 5 of chapter 2 ), but now we can see that it would always have been log-