Moving to a metazoan group generally regarded as relatively “simple” in form, and especially prominent in the fossil record, particularly in Paleozoic strata, Roberts (1981, p. 123) concluded from many years of studying Australian Carboniferous brachiopods: “There is no evidence of 'gradualistic' evolutionary processes affecting brachiopod species either within or between zones, and the succession of faunas can be regarded as 'punctuated.'”
Johnson (1975), inspired by Ziegler's (1966) documentation of one putatively gradualistic sequence in the brachiopod Eocoelia, decided to search for others — and found only examples of punctuation and stasis throughout the Paleozoic record. He wrote (1975, p. 657):
After completion of Ziegler's paper we talked a number of times about the possibilities of duplicating his efforts with other fossils and in other times. It was a heady prospect ... In subsequent years many workers have attempted to seek out and define lineages of brachiopod species and other megafossils in the lower and middle Paleozoic with little success. My conclusion, subjective in many ways, is that speciation of brachiopods in the mid-Paleozoic via a phyletic mode has been rare. Rather, it is probable that most new brachiopod species of this age originated by allopatric speciation.
Derek Ager, the world leader in studying later Mesozoic brachiopods, summed up his lifelong effort in several papers towards the end of his career. He wrote (1973, p. 20): “In twenty years work on the Mesozoic Brachiopods, I have found plenty of relationships, but few if any evolving lineages ... What it seems to mean is that evolution did not normally proceed by a process of gradual change of one species into another over long periods of time.” Ten years later (1983, p. 563), Ager reiterated: “The general picture seems to fit in with the Gouldian doctrine of 'hardly ever' [that is, documentation of gradualism only very rarely]. Certainly there is no evidence in the group as a whole [Page 754] of phyletic gradualism happening throughout a species at any one moment in time. Species A never changes into species B everywhere simultaneously and gradually.”
When we consider trilobites, the exemplars of Paleozoic invertebrate “complexity,” Robison (1975, p. 220) concluded from extensive study of Middle Cambrian agnostid trilobites in Western North America: “I have found a conspicuous lack of intergradation in species-specific characters, and I have also found little or no change in these characters throughout the observed stratigraphic ranges of most species.”
Fortey (1985) spent many years studying a particularly favorable sequence for fine-scale temporal resolution from the early Ordovician of Spitzbergen. He examined 111 trilobite and 56 graptoloid species, finding a predominance of punctuated equilibrium in both groups — with gradualism in “less than 10 percent of the total” for trilobites, and, for graptoloid species, with punctuational origins “at least four times as important as gradualistic ones” (1985, p. 27). Fortey's case becomes particularly convincing because he could calibrate punctuational sequences against rarer cases of gradualism in the same strata — and therefore be confident that punctuations do not merely represent the missing strata of conventional gradualistic rates. In a later paper, Fortey, who is, by the way, no partisan of punctuated equilibrium, reaches the following general conclusion, and also affirms our point about respect for the age-old knowledge of biostratigraphic practitioners: “Many invertebrate paleontologists would agree that the fossil record of species of their groups is dominated by lack of change — by stasis — and that where phylogenies have been worked out then the evidence direct from the rocks shows punctuated lineages in a majority of cases. For reasons I have explained, it is likely that stratigraphic paleontologists would always have maintained such a view, but the difference is that now this would be accepted by paleobiologists as well” (1988, p. 13).
Moving to a different arthropod group from another time, Coope's famous studies of Late Cenozoic fossil beetles (summarized in Coope, 1979) provide one of our best cases for dominance of the punctuational mode. Unusually good preservation greatly increases the power of this example. Coope discusses his best case (for beetles extracted from the carcasses of woolly rhinos in the western Ukraine), but then extends his argument to most examples:
Here the complete beetles were preserved down to the tarsal and antennal joints; when the elytra were raised, the wings could be unfolded and mounted; and parasitic mites, both larvae and adults, were found underneath the wings. Although this was quite exceptional preservation, it is common to find intact abdomens from which the genitalia can be dissected; the frequently transparent integument often reveals detailed structures of the internal sclerites. Preservation is frequently adequate to enable details of the microstructure of the surface of the hairs and scales to be examined with scanning electron microscopy (1979, p. 248). [Page 755]
Coope concluded that most species showed extensive stasis, even with such detail available for observation: “The early Pleistocene fossils, probably dating from over a million years ago, are referable to living species and some existing species extend well back into the late Tertiary” (1979, p. 250).
In what I regard as the most fascinating and revealing comment of all, George Gaylord Simpson, the greatest and most biologically astute paleontologist of the 20th century (and a strong opponent of punctuated equilibrium in his later years), acknowledged the literal appearance of stasis and geologically abrupt origin as the outstanding general fact of the fossil record, and as a pattern that would “pose one of the most important theoretical problems in the whole history of life” if Darwin's argument for artifactual status failed. Simpson stated at the 1959 Chicago centennial celebration for the Origin of Species (in Tax, 1960, p. 149):
It is a feature of the known fossil record that most taxa appear abruptly. They are not, as a rule, led up to by a sequence of almost imperceptibly changing forerunners such as Darwin believed should be usual in evolution. A great many sequences of two or a few temporally intergrading species are known, but even at this level most species appear without known intermediate ancestors, and really, perfectly complete sequences of numerous species are exceedingly rare . . . These peculiarities of the record pose one of the most important theoretical problems in the whole history of life: is the sudden appearance ... a phenomenon of evolution or of the record only, due to sampling bias and other inadequacies?
Such discordance between theoretical expectation and actual observation surely falls within the category of troubling “anomalies” that, in Kuhn's celebrated view of scientific change (1962), often spur a major reformulation.
DARWINIAN SOLUTIONS AND PARADOXES
Only one chapter of the Origin of Species bears an apologetic title — ironically, for the subject that should have provided the crown of direct evidence for evolution in the large: the archive of life's actual history as displayed in the fossil record. Yet Darwin entitled Chapter 9 “On the Imperfection of the Geological Record.”
In Chapter 2 (pp. 146–155), I discussed Darwin's convictions about gradualism, and the crucial link between his defense of natural selection and one of the three major and disparate claims subsumed within this complex concept: the insensibility of intermediacy. The theory of punctuated equilibrium does not engage this important meaning for two reasons: first, our theory does not question the operation of natural selection at its conventional organismic level; second, as a theory about the deployment of speciation events in macroevolutionary time, punctuated equilibrium explains how the insensible intermediacy of human timescales can yield a punctuational pattern in geological perspective — thus requiring the treatment of species as evolutionary individuals, [Page 756] and precluding the explanation of trends and other macroevolutionary patterns as extrapolations of anagenesis within populations.
Rather, punctuated equilibrium refutes the third and most general meaning of Darwinian gradualism, designated in Chapter 2 (see pp. 152–155) as “slowness and smoothness (but not constancy) of rate.” Natural selection does not require or imply this degree of geological sloth and smo
othness, though Darwin frequently, and falsely, linked the two concepts — as Huxley tried so forcefully to advise him, though in vain, with his famous warning: “you have loaded yourself with an unnecessary difficulty in adopting Natura non facit saltum so unreservedly.” The crucial error of Dawkins (1986) and several other critics’ lies in their failure to recognize the theoretical importance of this third meaning, the domain that punctuated equilibrium does challenge. Dawkins correctly notes that we do not question the second meaning of insensible intermediacy. But since his extrapolationist view leads him to regard only this second meaning as vital to the rule of natural selection, he dismisses the third meaning — which we do confute — as trivial. Since Dawkins rejects the hierarchical model of selection, he does not grant himself the conceptual space for weighing the claim that punctuated equilibrium's critique of the third meaning undermines the crucial Darwinian strategy for rendering all scales of evolution by smooth extrapolation from the organismic level. For this refutation of extrapolation by punctuated equilibrium validates the treatment of species as evolutionary individuals, and establishes the level of species selection as a potentially important contributor to macroevolutionary pattern.
This broadest third meaning of gradualism may not be required for natural selection at the organismic level, but gradualism as slowness and smoothness of rate (not just as insensible intermediacy between endpoints of a transition) forms the centerpiece of Darwin's larger worldview, indeed of his entire ontology — as illustrated (again, see Chapter 2) in the crucial role played by this style of gradualism throughout the corpus of his works — from his first book on the origin of coral atolls (1842) to his last on the formation of topsoil by the action of worms (1881).
Lest anyone doubt that Darwin strongly advocated this most inclusive form of gradualism as slowness and smoothness (in addition to the less comprehensive claim for insensible intermediacy of transitions), I shall cite a few examples from the full documentation of Chapter 2 — cases where Darwin clearly meant “slow and steady over geological scales,” not just “insensibly intermediate at whatever rate.”
For example Darwin argues that species may arise so slowly that the process generally takes longer than the entire duration of a geological formation (usually several million years) — thus explaining apparent stasis within a formation as gradual evolution over insufficient time to record visible change! Darwin writes (1859, p. 293): “Although each formation may mark a very long lapse of years, each perhaps is short compared with the period requisite to change one species into another.” Darwin even argued that the pace of evolutionary change might be sufficiently steady to serve as a rough geological [Page 757] clock: “The amount of organic change in the fossils of consecutive formations probably serves as a fair measure of the lapse of actual time” (1859, p. 488). I also show in Chapter 2 that Darwin's conviction about extreme slowness and steadiness of change can be grasped, perhaps best of all, as the common source of his major errors — particularly his fivefold overestimate for the denudation of the Weald, and his conjecture that complex metazoan life of modern form must have undergone an unrecorded Precambrian history as long, or longer, than its known Phanerozoic duration.
Despite this strong belief in geological gradualism, Darwin knew perfectly well — as all paleontologists always have — that stasis and abrupt appearance represent a norm for the observed history of most species. I needn't rehearse Darwin's solution to this dilemma, for his familiar argument represents more than a twice-told tale. Following the lead of his mentor, Charles Lyell, Darwin attributed this striking discordance between theoretical expectation and actual observation to the extreme imperfection of the fossil record.
(As discussed more fully on pages 479–484, this argument served as the centerpiece for LyelPs system, and for the entire uniformitarian school. But then, what alternative could they embrace? The literal appearance of the geological record so often suggested catastrophe, or at least “moments” of substantial change, especially in faunal turnover. To assert a gradualism of geological rate against this sensory evidence, one had to declare the evidence illusory by advancing the general claim — quite legitimate as a philosophical proposition — that science must often work by probing “behind appearance” to impose the expectations of a valid theory upon an empirical record that, for one reason or another, cannot directly express the actual mechanisms of nature. Moreover, the “argument from imperfection” holds substantial merit and cannot be dismissed as “special pleading.” Like most chronicles of history, and far more so than many others, the geological record is extremely spotty. To cite Lyell's famous metaphor once again, if Vesuvius erupted again and buried a modern Italian city atop Pompeii, later stratigraphers might find a sequence of Roman ruins capped by layers of volcanic ash and followed by the debris of modern Italy. Taken literally, this sequence would suggest a catastrophic end to Rome followed by a saltation, linguistically and technologically, to the industrial age — an artifact of nearly 2000 years of missing data that would have recorded the evolution of Italian from Latin and a gradual passage from walled cities to traffic jams.)
To quote the two most famous statements on this subject from the Origin of Species, Darwin summarizes his entire argument by closing Chapter 9 with Lyell's metaphor of the book (1859, pp. 310-311):
For my part, following out Lyell's metaphor, I look at the natural geological record, as a history of the world imperfectly kept, and written in a changing dialect; of this history we possess the last volume alone, relating only to two or three countries. Of this volume, only here and there a short chapter has been preserved; and of each page, only here and there a few lines. Each word of the slowly-changing language, in which the history [Page 758] is supposed to be written, being more or less different in the interrupted succession of chapters, may represent the apparently abruptly changed forms of life, entombed in our consecutive, but widely separated, formations.
In epitomizing both geological chapters, Darwin begins with a long list of reasons for such an imperfect record, and then concludes with his characteristic honesty (1859, p. 342): “All these causes taken conjointly, must have tended to make the geological record extremely imperfect, and will to a large extent explain why we do not find interminable varieties, connecting together all the extinct and existing forms of life by the finest graduated steps. He who rejects these views on the nature of the geological record will rightly reject my whole theory.” (Huxley must have been thinking of this line when he issued his warning that Darwin's unswerving support of natura non facit saltum represented “an unnecessary difficulty.” Darwin's “whole theory” — the mechanism of natural selection — does not require, as Huxley pointed out, this geological style of gradualism in rate.)
The paradoxes set by Darwin's solution for the current practice of paleontology and macroevolutionary theory receive their clearest expression in another remarkable statement from the Origin of Species (1859, p. 302), a testimony to Darwin's sophisticated understanding that nature's “facts” do not stand before us in pristine objectivity, but must be embedded within theories to make any sense, or even to be “seen” at all. Darwin acknowledges that he only understood the extreme imperfection of the geological record when paleontological evidence of stasis and abrupt appearance threatened to confute the gradualism that he “knew” to be true: “But I do not pretend that I should ever have suspected how poor a record of the mutations of life, the best preserved geological section presented, had not the difficulty of our not discovering innumerable transitional links between the species which appeared at the commencement and close of each formation, pressed so hardly on my theory.”
The paradox of insulation from disproof
The “argument from imperfection” (with its preposition purposefully chosen by analogy to the “argument from design”) works adequately as a device to save gradualism in the face of an empirical signal of quite stunning contrariness when read at face val
ue. But if we adopt openness to empirical falsification as a criterion for strong and active theories in science, consider the empty protection awarded to gradualism by Darwin's strategy. For the data that should, prima facie, rank as the most basic empirical counterweight to gradualism — namely the catalog of cases, and the resulting relative frequency, for observed stasis and geologically abrupt appearances of fossil morphospecies — receive a priori interpretation as signs of an inadequate empirical record. How then could gradualism be refuted from within?
The situation became even more insidious in subtle practice than a bald statement of the dilemma might suggest. Abrupt appearance (the punctuations [Page 759] of punctuated equilibrium) might well be attributed to the admittedly gross imperfection of our geological archives. The argument makes logical sense, must certainly be true in many instances, and can be tested in a variety of ways on a case-by-case basis (particularly when we can obtain independent evidence about rates of sedimentation).
But how can imperfection possibly explain away stasis (the equilibrium of punctuated equilibrium)? Abrupt appearance may record an absence of information, but stasis is data. Eldredge and I became so frustrated by the failure of many colleagues to grasp this evident point — though a quarter century of subsequent debate has finally propelled our claim to general acceptance (while much else about punctuated equilibrium remains controversial) — that we urged the incorporation of this little phrase as a mantra or motto. Say it ten times before breakfast every day for a week, and the argument will surely seep in by osmosis: “stasis is data; stasis is data...”
The fossil record may, after all, be 99 percent imperfect, but if you can, nonetheless, sample a species at a large number of horizons well spread over several million years, and if these samples record no net change, with beginning and end points substantially the same, and with only mild and errant fluctuation among the numerous collections in between, then a conclusion of stasis rests on the presence of data, not on absence! In such cases, we must limit our lament about imperfection to a wry observation that nature, rather than human design, has established a sampling scheme by providing only occasional snapshots over a full interval. We might have preferred a more even temporal spacing of these snapshots, but so long as our samples span the temporal range of a species, with reasonable representation throughout, why grouse at nature's failure to match optimal experimental design — when she has, in fact, been very kind to us in supplying abundant information. Stasis is data.
The Structure of Evolutionary Theory Page 121