Darwin's Dangerous Idea (54 page)

presenting a true alternative to the rampant ramp-view of orthodoxy). This As Gould and Eldredge have themselves pointed out, there was an obvious makes it look as if each of the design revisions illustrated takes place in a problem of scale in such diagrams as figure 10.6. What if we zoomed way in twinkling, in no time at all. But that is just a misleading artifact of the huge on the orthodox picture and found that, once we enlarged it sufficiently, it vertical scale adopted, which shows millions of years to the inch. The looked like this:

sideways motion is not really instantaneous. It is only "geologically instantaneous."

An isolated population may take a thousand years to speciate, and its transformation would therefore appear glacially slow if measured by the irrelevant scale of our personal lives. But a thousand years, appropriately recorded in geological time, is only an unresolvable moment, usually preserved on a single bedding plane [in fossil-bearing rock], in a lifetime of species that often live for several million years in stasis. [Gould 1992a, pp.

12-14.]

So suppose we zoom in on one of these thousand-year instants, changing the vertical scale of the time dimension by a few orders of magnitude to see what might actually be going on (figure 10.8). The horizontal step taken between time t and time t' will have to be stretched out somehow, and we must turn it into relatively big steps or little steps or tiny steps, or some combination thereof.

Were any of the possibilities revolutionary? What exactly were Eldredge FIGURE 10.7

and Gould maintaining? Here their respective views diverged somewhat, at least for a while. The view was revolutionary, Gould claimed, because it At
some
level of magnification, any evolutionary ramp must look like a maintained that the punctuations were not just business-as-usual evolution, staircase. Is figure 10.7 a picture of punctuated equilibrium? If it is, then not just
gradual
changes. Remember the old joke about the drunk who falls orthodox Darwinism was already a theory of punctuated equilibrium. Even down the elevator shaft and says, on rising, "Look out for the first step—it's a doozy!"? For a while, Gould was proposing that the first step in the establishment of any new species was a doozy—a non-Darwinian
saltation
paleontological data. If new species arise very rapidly in small, peripherally isolated local ("somersault" and "saute" come from the same Latin root): populations, then the great expectation of insensibly graded fossil sequences is a chi-mera" (Eldredge and Gould 1972, p. 82).

Punctuated Equilibrium: A Hopeful Monster
287

286 BULLY FOR BRONTOSAURUS

It is this "creative role" of something other than selection that caught the skeptical attention of Gould's colleagues. To get clear about what caused the furor, we need to note that our diagram in figure 10.8 is really unable to distinguish several crucially different hypotheses. The trouble with the diagram is that it needs more dimensions, so we can compare the steps in
genotype space
(the typographical steps in the Library of Mendel) to the steps
in phenotype space
(the design innovations in Design Space ) and then evaluate these differences on a
fitness landscape.
As we have seen, the relations between recipe and result are complex, and many possibilities might be illustrated. We saw in chapter 5 that a small typographical change in the genome could in principle have a large effect on the phenotype expressed.

We also saw, in chapter 8, that some typographical changes in the genome can have no effect at all on the phenotype—there are over a hundred different ways of "spelling" lysozyme, for instance, and hence more than a hundred equivalent ways of spelling the order for lysozyme in DNA codons. We know, then, that at one extreme there can be organisms so similar in design as to be indistinguishable that nevertheless have large differences in their Gould's leap

DNA—for instance, you and whoever that person is for whom you are often FIGURE 10.8

mistaken (your
Doppelgänger
—no philosophy book would be complete with mentioning doppelgängers ). At the other extreme, there can be organisms Speciation is not always an extension of gradual, adaptive allelic substitu-that are bizarrely different in appearance, but almost identical genetically. A tion to greater effect, but may represent, as Goldschmidt argued, a different single mutation in just the wrong place can produce a monster—the medical style of genetic change—rapid reorganization of the genome, perhaps non-term for such deformed offspring is
terata,
Greek (and Latin) for "monsters."

adaptive. [Gould 1980b, p. 119.]

And there can also be organisms that are almost identical in appearance and structure, and almost identical in DNA, but dramatically different in fitness—

Speciation itself, in this view, is not an effect of accumulated adaptations for instance, fraternal twins one of whom happens to have a gene that gives it gradually driving populations apart but, rather, a cause with its own, non-either immunity or susceptibility to some disease.

Darwinian explanation:

A large leap in any of these three spaces, or a saltation, may also be called a
macromutation
(meaning a big mutation, not just a mutation in what I have But in saltational, chromosomal speciation, reproductive isolation comes called a macro—a macromolecular subsystem ).5 As Ernst Mayr (1960 ) has first and cannot be considered as an adaptation at all. . . . We can, in fact, observed, there are three different reasons we could call a mutation big: it is a reverse the conventional view and argue that speciation, by forming new big step in the Library of Mendel; it produces a radical difference in entities stochastically, provides raw material for selection. [Gould 1980b, phenotype (a monster); it produces (one way or another) a big increase in p. 124.]

fitness—a lot of
lifting,
in our metaphor for
good work done
by design changes.

This suggestion, which I call Gould's leap, is represented in the right-hand It is possible for the molecular replicating machinery to take large steps in graph in figure 10.8. Only part of the punctuation process, the gradual, the Library of Mendel—there are cases in which whole chunks of text get cleaning-up process at the end, is "Darwinian," Gould claimed: transposed, inverted, or deleted in a single copying "mistake." It is also If new
Baupläne
often arise in an adaptive cascade following the saltational origin of a key feature, then part of the process is sequential and adaptive, and therefore Darwinian; but the initial step is not, since selec-5. For an introduction to the term, see Dietrich's essay "Macromutation" in the excellent new sourcebook,
Keywords in Evolutionary Biology,
edited by Keller and Lloyd (1992 ).

tion does not play a creative role in building the key feature. [Gould 1982a, p. 383.]

288 BULLY FOR BRONTOSAURUS

Punctuated Equilibrium: A Hopeful Monster
289

possible for typographical differences to accumulate slowly (and, in general, had ever endorsed—Goldschmidt's saltationism, they scoffed all the more randomly) over a long time in the large portion of DNA that never gets derisively. They knew what he'd said.

expressed, and if these accumulated changes suddenly got expressed, thanks But did they? I must admit that I thought they did until Steve Gould to some transposing error, a huge phenotypic effect would be expected. But it insisted to me that I should check
all
his various publications, and see for is only when we turn to the third sense of macromutation—large differences myself that his opponents were foisting a caricature on him. He struck a in fitness—that we can get clear about what seemed to be radical in Gould's nerve; no one knows better than I how frustrating it is to have the skeptics proposal. The terms "saltation" and "macromutation" have tended to be used hang a crude but convenient label on one's subtle view. (I'm the guy who to describe a successful move, a
creative
move, in which offspring in a single reputedly denies that people experience colors or pains, and thinks that generation shift from one region of Design Space to another
and prosper as a
thermostats think—just ask my critics.) So I checked. He chose to dub his
result.
The idea had been promoted by Richard Goldschmidt (1933, 1940), denial of gradualism "the Goldschmidt break" (Gould 1980b), and recom-and made unforgettable by his catchphrase: "hopeful monsters." What made mended for serious consideration—without endorsing—some radical Gold-his work notorious was that he claimed that such leaps were necessary for schmidtian views, but in the same paper he was careful to say, "We do not speciation to occur.

now accept all his arguments about the nature of variation." In 1982, he made This suggestion had been roundly rejected by neo-Darwinian orthodoxy, it clear that the only feature of Goldschmidt's view he was endorsing was the for the reasons we have already considered. Even before Darwin, the received idea of "small genetic changes producing large effects by altering rates of wisdom of biologists was, as Linnaeus said in his classic work of taxonomy development" (Gould 1982d, p. 338), and in his introduction to the reprinting (1751),
"Natura non facit saltus"
—nature does not make leaps—and this was of Goldschmidt's notorious book, he expanded on this point: one maxim that Darwin didn't just leave untouched; he provided enormous support for it. Large leaps sideways
in a fitness landscape
will almost never Darwinians, with their traditional preferences for gradualism and continu-be
to
your benefit; wherever you currently find yourself, you are where you ity, might not shout hosannas for large phenotypic shifts induced rapidly are because this has been a good region of Design Space for your ancestors—

by small genetic changes that affect early development; but nothing in you are near the top of some peak or other in the space—so, the bigger the Darwinian theory precludes such events, for the underlying continuity of step you take (jumping randomly, of course), the more likely you are to jump small genetic changes remains. [Gould 1982b, p. xix.]

off a cliff—into the low country, in any case (Dawkins 1986a, ch. 9).

According to this standard reasoning, it is no accident that monsters are Nothing revolutionary, in other words:

virtually always hopeless. That is what made Goldschmidt's views so heretical; he knew and accepted that this was true in general, but proposed One may be excused for retorting: "so what else is new?" Has any biologist that nevertheless the extremely rare exceptions to this rule were the main ever denied it? But ... progress in science often demands the recovery of lifters of evolution.

ancient truths and their rendering in novel ways. [Gould 1982d, pp. 343-Gould is a famous defender of underdogs and outcasts, and he deplored the 44.]

"ritualistic ridicule" (Gould 1982b, p. xv) to which Goldschmidt had been subjected by the orthodox. Was Gould going to try to rehabilitate Still, he could not resist the urge to describe this possibly underappreciated Goldschmidt? Yes and no. In "Return of the Hopeful Monster" (in Gould fact about development as a non-Darwinian creative force in evolution, 1980a, p. 188), Gould complained that "defenders of the synthetic theory

"because the constraints that it imposes upon the nature of phenotypic change made a caricature of Goldschmidt's ideas in establishing their whipping boy."

guarantee that small and continuous Darwinian variation is not the raw So it seemed to many biologists that Gould was arguing that punctuated material of all evolution," for it "relegates selection to a negative role equilibrium was a theory of Goldschmidtian speciation through mac-

(eliminating the unfit) and assigns the major creative aspect of evolution to romutation. To them it seemed that Gould was trying to wave his wonderful variation itself' (Gould 1982d, p. 340).

historian's wand over the tarnished reputation of Goldschmidt, and bring his It is still not clear how much importance to assign to this possibility in ideas back into favor. Here the mythic Gould, Refuter of Orthodoxy, principle, but, in any event, Gould has not pursued it further: "Punctuated seriously got in the way of the real Gould, so that even his colleagues equilibrium is not a theory of macromutation" (Gould 1982c, p. 88). Con-succumbed to the temptation to read what he wrote with a broad brush. They fusion on this score still abounds, however, and Gould has had to keep scoffed in disbelief, and then, when he denied that he was endorsing—

issuing his disclaimers: "Our theory entails no new or violent mechanism,