So if stasis could not be explained away as missing information, how could gradualism face this most prominent signal from the fossil record? The most negative of all strategies — a quite unconscious conspiracy of silence — dictated the canonical response of paleontologists to their observations of stasis. Again, a “culprit” may be identified in the ineluctable embedding of observation within theory. Facts have no independent existence in science, or in any human endeavor; theories grant differing weights, values, and descriptions, even to the most empirical and undeniable of observations. Darwin's expectations defined evolution as gradual change. Generations of paleontologists learned to equate the potential documentation of evolution with the discovery of insensible intermediacy in a sequence of fossils. In this context, stasis can only record sorrow and disappointment.
Paleontologists therefore came to view stasis as just another failure to document evolution. Stasis existed in overwhelming abundance, as every paleontologist always knew. But this primary signal of the fossil record, defined as an absence of data for evolution, only highlighted our frustration — and certainly did not represent anything worth publishing. Paleontology therefore fell into a literally absurd vicious circle. No one ventured to document or [Page 760] quantify — indeed, hardly anyone even bothered to mention or publish at all — the most common pattern in the fossil record: the stasis of most morphospecies throughout their geological duration.
All paleontologists recognized the phenomenon, but few scientists write papers about failure to document a desired result. As a consequence, most nonpaleontologists never learned about the predominance of stasis, and simply assumed that gradualism must prevail, as illustrated by the exceedingly few cases that became textbook “classics”: the coiling of Gryphaea, the increasing body size of horses, etc. (Interestingly, nearly all these “classics” have since been disproved, thus providing another testimony for the temporary triumph of hope and expectation over evidence — see Gould, 1972.) Thus, when punctuated equilibrium finally granted theoretical space and importance to stasis, and this fundamental phenomenon finally emerged from the closet, nonpaleontologists were often astounded and incredulous. Mayr (1992, p. 32) wrote, for example: “Of all the claims made in the punctuationalist theory of Eldredge and Gould, the one that encountered the greatest opposition was the observation of 'pronounced stasis as the usual fate of most species,' after having completed the phase of origination ... I agree with Gould that the frequency of stasis in fossil species revealed by the recent analysis was unexpected by most evolutionary biologists.”
(To cite a personal incident that engraved this paradox upon my consciousness early in my career, John Imbrie served as one of my Ph.D. advisors at Columbia University. This distinguished paleoclimatologist began his career as an evolutionary paleontologist. He accepted the canonical equation of evolution with gradualism, but conjectured that our documentary failures had arisen from the subtlety of gradual change, and the consequent need for statistical analysis in a field still dominated by an “old-fashioned” style of verbal description. He schooled himself in quantitative methods and applied this apparatus, then so exciting and novel, to the classic sequence of Devonian brachiopods from the Michigan Basin — where rates of sedimentation had been sufficiently slow and continuous to record any hypothetical gradualism. He studied more than 30 species in this novel and rigorous way — and found that all but one had remained stable throughout the interval, while the single exception exhibited an ambiguous pattern. But Imbrie did not publish a triumphant paper documenting the important phenomenon of stasis. Instead, he just becomes disappointed at such “negative” results after so much effort. He buried his data in a technical taxonomic monograph that no working biologist would ever encounter (and that made no evolutionary claims at all) — and eventually left the profession for something more “productive.”)
Paradoxes of this sort can only be resolved by input from outside — for gradualism, having defined contrary data either as marks of imperfection or documents of disappointment, could not be refuted from within. Reassessment required a different theory that respected stasis as a potentially fascinating phenomenon worthy of rigorous documentation, not merely as a failure to find “evolution.” Eldredge and I proposed punctuated equilibrium in this explicit context — as a framework and different theory that, if true, could validate [Page 761] the primary signal of the fossil record as valuable information rather than frustrating failure. We therefore began our original article (Eldredge and Gould, 1972) with a philosophical discussion, based on work of Kuhn (1962) and Hanson (1961), on the necessary interbedding of fact and theory. We ended this introductory section by writing (1972, p. 86):
The inductivist view forces us into a vicious circle. A theory often compels us to see the world in its light and support. Yet we think we see objectively and therefore interpret each new datum as an independent confirmation of our theory. Although our theory may be wrong, we cannot confute it. To extract ourselves from this dilemma, we must bring in a more adequate theory; it will not arise from facts collected in the old way ... Science progresses more by the introduction of new world-views or “pictures” than by the steady accumulation of information . . . We believe that an inadequate picture has been guiding our thoughts on speciation for 100 years. We hold that its influence has been all the more tenacious because paleontologists, in claiming that they see objectively, have not recognized its guiding sway. We contend that a notion developed elsewhere, the theory of allopatric speciation, supplies a more satisfactory picture for the ordering of paleontological data.
The paradox of stymied practice
This second paradox cascades from the first. If a theory — geologically insensible gradualism as the anticipated expression of evolution in the fossil record, in this case — can insulate itself against disproof from within by defining contrary data as artifactual, then proper assessments of relative frequencies can never be achieved — for how many scientists will devote a large chunk of a limited career to documenting a phenomenon that they view as a cardinal restriction recording a poverty of available information?
Paleontological studies of evolution therefore became warped in a lamentable way that precluded any proper use of the fossil record, but seemed entirely honorable at the time. We practitioners of historical sciences, as emphasized throughout this book, work in fields that decide key issues by assessment of relative frequencies among numerous possible outcomes, and only rarely by the more “classical” technique of “crucial experiments” to validate universal phenomena. Therefore, any method that grossly distorts a relative frequency by excluding a common and genuine pattern from consideration must seriously stymie our work. When traditional paleontologists eliminated examples of abrupt appearance and stasis from the documentation of evolution, they only followed a conventional precept — for they believed that both patterns recorded an artifact of imperfect data, thus debarring such cases from consideration. The relative distributions of evolutionary rates would therefore emerge only from cases of gradualism — the sole examples judged as sufficiently data-rich to record the process of evolution in adequate empirical detail. But this project could not even succeed in its own terms, for gradualism occurs [Page 762] too rarely to generate enough cases for calculating a distribution of rates. Instead, paleontologists worked by the false method of exemplification: validation by a “textbook case” or two, provided that the chosen instances be sufficiently persuasive. And even here, at this utterly minimal level of documentation, the method failed. A few examples did enter the literature (see Fig. 9-2 for comparison of an original claim with a secondary textbook version) — where they replicated by endless republication in the time-honored fashion of textbook copying (see Gould, 1988a). But, in a final irony, almost all these famous exemplars turned out to be false on rigorous restudy — see Hallam, 1968, and Gould, 1972, for stasis rather than gradual increase in coiling in the Liassic oyster G
ryphaea; Prothero and Shubin, 1989, on stasis within all documented species of fossil horses, and with frequent overlap between ancestors and descendants, indicating branching by punctuational speciation rather than anagenetic gradualism; and Gould, 1974, on complete absence of data for the common impression that the enormous antlers of Megaloceros
9-2. Trueman's original claim for phyletic gradualism in the increased coiling of Gryphaea in Lower Jurassic rocks of England (left). To the right, a textbook smoothing and simplification of the same figure. Trueman's claim has been invalidated for two reasons: first, Gryphaea did not evolve from Ostrea; and, second, subsequent studies have not validated any increase of coiling within Gryphaea, despite Trueman's graphs. Nonetheless, once such figures become ensconced in textbooks, they tend to persist even when their empirical justification has long been refuted in professional literature.
[Page 763]
(the “Irish Elk”) increased gradually in phylogeny, with positive allometry as body size enlarged.
Traditional paleontology therefore placed itself into a straight jacket that made the practice of science effectively impossible: only a tiny percentage of cases passed muster for study at all, while the stories generated for this minuscule minority rested so precariously upon hope for finding a rare phenomenon — and received such limited definition by the primitive statistical methods then available (or, more commonly, remained unidentified by any statistical practice at all) — that even these textbook exemplars collapsed upon restudy with proper quantitative procedures. But consider what might have occurred, if only paleontologists had recognized that stasis is data (I will grant some validity to the standard rationale for regarding the second phenomenon of punctuation as an artifact of an imperfect record). As Hallam said to me many years ago, after he had disproved the classical story of gradualism in Gryphaea: more than 100 other species of mollusks, many with records as rich as Gryphaea's, occur in the same Liassic rocks, yet no one ever documented the stratigraphic history of even a single one in any study of evolution, for all demonstrate stasis. Scientists picked out the only species that seemed to illustrate gradualism, and even this case failed.
Despite the widespread use of proper quantitative methods today, and despite increasing attention to the validity of stasis as an evolutionary phenomenon, this bias still persists. I do not doubt that several species of Cenozoic planktonic Foraminifera display gradual transitions (see pp. 803–810), but I know that these examples have been extracted for study from a much larger potential sample of species never documented in detail because their apparent stasis seems “boring” to students of evolution. An eager young statistician goes to a lifelong expert and says: I want to devote my doctoral thesis to a statistical study of evolution in a species of foram (the most promising of major taxa, thanks to a hyper abundance of specimens and excellent stratigraphic data in oceanic cores); which species shall I choose? And the expert advises: why not study Graduloconoides gradualississima; I know that this species shows interesting changes during the upper Miocene in cores A through Z. Meanwhile, poor old boring Stasigerina punctiphora, just as abundant in the same cores, and just as worthy of study, gets bypassed in silence.
I find this situation particularly frustrating as paleontology's primary example of an insidious phenomenon in science that simply has not been recognized for the serious and distorting results perpetrated under its aegis. Most scientists do not even recognize the problem — though some do, particularly in the medical and social sciences, where the error has been named “publication bias,” and has inspired a small but important literature (Begg and Berlin, 1988). In publication bias, prejudices arising from hope, cultural expectation, or the definitions of a particular scientific theory dictate that only certain kinds of data will be viewed as worthy of publication, or even of documentation at all. Publication bias bears no relationship whatever with the simply immoral practice of fraud; but, paradoxically, publication bias may exert a far more serious effect (largely because the phenomenon must be so much [Page 764] more common than fraud) — for scientists affected by publication bias do not recognize their errors (and their bias may be widely shared among colleagues), while a perpetrator of fraud operates with conscious intent, and the wrath of a colleague will be tremendous upon any discovery.
Begg and Berlin (1988) cite several documented cases of publication bias. We can hardly doubt, for example, that a correlation exists between socioeconomic status and academic achievement, but the strength and nature of this association can provide important information, for both political practice and social theory. White (1982, cited in Begg and Berlin) found a progressively increasing intensity of correlation with prestige and permanence of published source. Studies published in books reported an average correlation coefficient of 0.51 between academic achievement and socioeconomic status; articles in journals gave an average of 0.34, while unpublished studies yielded a value of 0.24. Similarly, Coursol and Wagner (1986, cited in Begg and Berlin) found publication bias both in the decision to submit an article at all, and in the probability for acceptance. In a survey of outcomes in psychotherapy, they noted that 82 percent of studies with positive outcomes led to submission of papers to a journal, while only 43 percent of negative outcomes provoked an attempt at publication. Of papers submitted, 80 percent that report positive outcomes were accepted for publication, but the figure fell to 50 percent for papers claiming negative results.
In my favorite study of publication bias, Fausto-Sterling (1985) tabulated claims in the literature for consistent differences in cognitive and emotional styles between men and women. She does not deny that genuine differences often exist, and in the direction conventionally reported. But she then, so to speak, surveys her colleagues' file drawers for studies not published, or for negative results published and then ignored, and often finds that a great majority report either a smaller and insignificant disparity between sexes, or no differences at all. When she collated all studies, rather than only those published, the much-vaunted differences often dissolved into statistical insignificance or triviality.
For example, a recent favorite theme of pop psychology attributed different cognitive styles in men and women to the less lateralized brains of women. Some studies have indeed reported a small effect of greater male lateralization; none has found more lateralized brains in women. But most experiments, as Fausto-Sterling shows, detected no measurable differences in lateralization at all and this dominant relative frequency (even in published literature) should be prominently reported in the press and in popular books, but tends to be ignored as “no story.”
Paleontology's primary example of publication bias — the nonreporting of stasis under the false belief that such stability represents “no data” for evolution — illustrates a particularly potent form of the general phenomenon, a category that I have called “Cordelia's dilemma” (Gould, 1995) to memorialize the plight of King Lear's honest but rejected daughter. When asked by Lear for a fulsome protestation of love in order to secure her inheritance, Cordelia, disgusted by the false and exaggerated speeches of her sisters Goneril and [Page 765] Regan, chose to say nothing, for she knew that “my love's more ponderous than my tongue.” But Lear mistook her silence for hatred or indifference, and cut her off entirely (with tragic consequences later manifested in his own madness, blindness, and death), in proclaiming that “nothing will come of nothing.”
Cordelia's dilemma arises in science when an important (and often predominant) signal from nature isn't seen or reported at all because scientists read the pattern as “no data,” literally as nothing at all. This odd status of “hidden in plain sight” had been the fate of stasis in fossil morphospecies until punctuated equilibrium gave this primary signal some theoretical space for existence. Apparent silence — the overt nothing that actually records the strongest something — can embody the deepest and most vital meaning of all. What, in western history, has been more eloquent than the silence of Jes
us before Pilate, or Saint Thomas More's date with the headsman because he acknowledged that fealty forbade criticism of Henry VIII's marriage to Anne Boleyn, but maintained, literally to the death, his right to remain silent, and not to approve?
In summary, the potentially reformative role of punctuated equilibrium resides in an unusual property among scientific innovations. Most new theories in science arise from fresh information that cannot be accommodated under an old explanatory rubric. But punctuated equilibrium merely honored the firmest and oldest of all paleontological observations — the documentable stasis of most fossil morphospecies — by promoting this pattern to central recognition as an expected result of evolution's proper expression at the scale of geological time. This reformulation cast a bright light upon stasis, a preeminent fact that had formerly been mired in Cordelia's dilemma as a grand disappointment, and therefore as “no data” at all, a pattern fit only for silence in a profession that accepted Darwin's argument for gradualism as the canonical expression of evolution in the fossil record.
The Structure of Evolutionary Theory Page 122