Moreover, and more notably for its capacity to lead us astray when we operate within a conceptual box defined by anagenetic flux rather than variation in numbers of taxa, an additional 25-28 percent of genera fall into a third category of generally and symmetrically increasing variation through the sequence — that is, the final range for all species within the genus includes species both smaller and larger than the extremes of the ancestral spread. I strongly suspect that a previous inclusion of these genera as affirmations of Cope's Law engendered the false result of dominant relative frequency for phyletic size increase. Older treatments of the topic usually considered extreme values only, and affirmed Cope's Law if any later species exceeded the common ancestor in size — thus repeating, in miniature, the same error generally committed for life's totality by ignoring the continuing domination of bacteria, and using the motley sequence of trilobite to dinosaur to human as evidence for a central and defining thrust. Obviously, from a variational and [Page 905] speciational perspective, successful genera with substantially increasing numbers of species through time will probably expand their range at both extremes of size, thus undergoing a speciational trend in variation, not an anagenetic march to larger sizes!
Particular cases. This lowest macroevolutionary level of individual monophyletic clades has defined the soul of paleontological discourse through the centuries. Only the histories of particular groups can capture the details that all vivid story telling requires; the “why” of horses and humans certainly elicits more passion than the explanation (or denial) of Cope's Law, or the pattern of increasing mean species longevity in marine invertebrates through time. Yet, even here on such familiar ground, our explanations remain so near and yet so far — for these “closer” stories of particular histories must also be reexamined in a speciational light. Consider just two “classics” and their potential revisions.
1. HORSES AS THE EXEMPLAR OF “LIFE'S LITTLE JOKE.” As noted above (Fig. 7-3, and 580-581), the line of horses, proceeding via three major trends of size, toes and teeth from dog-sized, many-toed, “eohippus” with low-crowned molars to one-toed Equus with high-crowned molars (see Fig. 9-33 for W. D. Matthew's classic icon, linearly ordered by stratigraphy) still marches through our textbooks and museums as the standard-bearer for adaptive trending towards bigger and better.
I do not deny that, even in a refomulated speciational context, several aspects of the traditional story continue to hold. MacFadden (1986), for example, has documented a clear cladal bias towards the punctuational origin of new species at larger sizes than their immediate ancestors, so both the iconic transition from ladders to bushes, and the recognition that several speciational events lead to smaller sizes, even to dwarfed species, throughout the range of the lineage, does not threaten (but rather reinterprets in interesting ways) the conventional conclusion that horses have generally increased in size through the Cenozoic. Moreover, I do not doubt the usual adaptational scenario that a transition from browsing in soft-turfed woodlands to grazing on newly-evolved grassy plains (grasses did not evolve until mid-Tertiary times) largely explains the adaptive context for both the general reduction in numbers of toes from splayed feet on soft ground to hoofs on harder substrates, and the increasing height of cheek teeth to prevent premature wear from eating grasses of high silica content.
Nonetheless, a speciational reformulation in terms of changing diversity as well as anatomical trending tells a strikingly different, and mostly opposite, story for the clade as a whole. Modern perissodactyls represent but a shade of their former glory. This clade once dominated the guild of large-bodied mammalian herbivores, with speciose and successful groups, especially the titanotheres, that soon became extinct, and with diversity in existing groups far exceeding modern levels. (The rhinoceros clade once included agile running forms, the hyrachiids, wallowing hippo-like species, and the indricotheres, the largest land mammals that ever lived.) Modern perissodactyls exist as three small clades of threatened species: horses, rhinos, and tapirs.
Horses have declined precipitously from their maximal mid-Tertiary abundance
[Page 906]
9-33. The classic icon, drawn by W. D. Matthew early in the 20th century, of the linear evolution of horses depicted as a ladder of progress towards larger size, fewer toes, and higher crowned teeth.
[Page 907]
to modern marginality in both place and number. As O.C. Marsh proved to T.H. Huxley in a famous incident in the history of paleontology (see Gould, 1996a, pp. 57-58), horses evolved in America only to peter out and disappear completely in their native land, surviving only as a few lines of Old World migrants. In North America alone, from 8 to 15 million years ago, an average of 16 species lived contemporaneously. As recently as 5 million years ago, at least six sympatric species of horses lived in Florida alone (MacFadden et al., 1999). Only 6-8 species of horses inhabit the entire earth today.
Moreover, if one wished to argue that a particular affection for Equus permits a restricted focus on this pathway through the bush alone, with the pruning of all other pathways regarded as irrelevant; and if one then claimed that this sole surviving pathway illustrates the predictable excellence of increasing adaptation by expressing a pervasive cladal trend (so that any surviving lineage would still tell the same adaptive tale for its own particular sequence); then the speciational view must debunk this last potential version of the old triumphalist attitude as well. If all paths through the equid bush led to the same Rome of modern Equus, or even if all major and prosperous paths — as measured by species range, diversity, or any conventional attribute of phyletic success — moved in the same general direction, then one might separate issues of species numbers, where the decline of equids through time cannot be denied, from the question of cladal direction, where the classic trend could still be asserted as predictable and progressive, even while the clade declined in diversity.
But reasonable, and very different, alternative scenarios for cladal direction remained viable until the recent restriction of the clade to the single genus Equus. MacFadden's (1988) study of all identifiable ancestral-descendant pairs showed 5 of 24, or more than 20 percent, leading towards decreased size. The general bias remains clear, but alternative scenarios also remain numerous and entirely reasonable (albeit, obviously, unrealized in the unpredictable history of life).
For example, the dwarfed genus Nannippus lived as a highly successful equid subclade for eight million years (far longer than Equus so far) at substantial species diversity (at least four named taxa) in North America, becoming extinct only 2 million years ago. Suppose that, in the reduction of the clade to one genus, Nannippus had survived and the forebears of Equus died? What, with this small and plausible alteration by “counterfactual history,” would we then make of the vaunted and canonical trend in the evolution of horses? Nannippus did not exceed ancestral eohippus in size, and still grew three toes on each foot. Nannippus molars were as relatively high-crowned as those of modern Equus, but I doubt that any biologist would challenge my noncynical scenario based on psychological salience: If Equus had died, and Nannippus survived as the only genus of modern horses, this “insignificant” clade would now be passing beneath our pedagogical notice as just one more “unsuccessful” mammalian line — like aardvarks, pangolins, hyraxes, dugongs, and several others — equally reduced to a shadow of former success. [Page 908]
The situation becomes even more paradoxical, and also entirely general, when we take an honest look at our iconographic prejudices in the light of speciational reformulations for macroevolution. Consider the true “success stories” of mammalian evolution — the luxurious clades of rodents, chiropterans (bats), and, among large-bodied forms, the antelopes of the artiodactyl clade that expanded so vigorously as perissodactyls constantly declined within the guild of large-bodied hoofed herbivores. Has anyone ever seen, either in a textbook or a museum hall, a chart or a picture of these truly dominating clades among modern ma
mmals? We don't depict these stories because we don't know how to draw them under our restrictive anagenetic conventions.
If we define evolution as anagenetic trending to a “better” place, how can we depict a successful group with copious modern branches extending in all directions within the cladal morphospace? Instead, and entirely unconsciously of course, for we would laugh at ourselves if we recognized the fallacy, our conventions lead us to search out the histories of highly unsuccessful clades — those now reduced to a single surviving lineage — as exemplars of triumphant evolution. We take this only extant and labyrinthine path through the phyletic bush, use the steamroller of our preconceptions to linearize such a tortuous route as a main highway, and then depict this straggling last gasp as the progressive thrust of a pervasive trend. I refer to this pattern of “life's little joke” (see Gould, 1996a).
2. rethinking human evolution. Ever since Protagoras proclaimed that “man [meaning all of us] is the measure of all things,” Western intellectual traditions, bolstering our often unconscious emotional needs, have invariably applied the general biases of our analytic procedures, with special energy and focussed intensity, to our own particular history. As a cardinal example, concepts of human evolution long labored under the restrictive purview (now known to be empirically false) of the so-called “single species hypothesis” (see Brace, 1977) — the explicit claim that a maximal niche breadth, implied by the origin of hominid consciousness, made the coexistence of more than one species impossible, since no two species can share the same exact habitat under the “competitive exclusion” principle, and since consciousness must have expanded the effective habitat of hominid species into the full ecological range of conceivable living space. Human evolution could therefore be viewed as a single progressive series gradually trending towards our current pinnacle.
Needless to say, such a scheme precluded any speciational account of human history because only one taxon could exist at any one time, and trends had to record the anagenetic transformation of the only existing entity. (I do not think that such a relentlessly limiting scenario has ever been presented a priori, or so stoutly defended in principle even by a special name of its own, for any other lineage in the history of life.) This theoretical stricture enforced several episodes of special pleading for apparently contrary data. For example, early evidence for two distinct and contemporaneous australopithecine lineages (the so-called gracile and robust forms, now universally regarded as [Page 909] separate taxa) inspired dubious proposals about sexual dimorphism. The same theoretical constraint led many researchers to regard European neanderthals as necessarily transitional between Homo erectus and modern humans, even though the empirical record indicated a punctuational replacement in Europe about 40,000 years ago, with no evidence for any anatomical intermediacy.
I freely confess my partisan attitude, but I do think that reforms of the past two decades have centered upon a rethinking of this phylogeny in speciational terms, with punctuated equilibrium acting both as a spur for reformulation and a hypothesis with growing empirical support. Claims for stasis have been advanced, and much debated, for the only two species with appreciable longevity of a million years or more — Australopithecus afarensis, aka “Lucy,” and Homo erectus (Rightmire, 1981, 1986, for support; Wolpoff, 1984, for denial). Lucy's earlier case seems well founded, while claims for our immediate ancestor, Homo erectus, have inspired more controversy, with arguments for gradual trending towards Homo sapiens, at least for Asian populations, generating substantial support within the profession. (But if the reported date — see Swisher et al., 1996 — of 30,000 to 50,000 years for the Solo specimens holds, then the geologically youngest Asian H. erectus does not stand at the apex of a supposed trend.)
More importantly, hominid “bushiness” has sprouted on all major rungs of the previous ladder implied by the single-species hypothesis. The hominid tree may grow fewer branches than the comparable bush of horses, but multiple events of speciation now seem to operate as the primary drivers of human phylogeny (see Leakey et al., 2001, for a striking extension to the base of the known hominid bush in the fossil record), while humans also share with horses the interesting feature of present restriction to a single surviving lineage, albeit temporarily successful, of a once more copious array. (I would also suggest, on the theme of “life's little joke,” that the contingent happenstance of this current restriction has skewed our thinking about human phylogeny away from more productive scenarios based on differential speciation.)
Speciation has replaced linearity as the dominant theme for all three major phases of hominid evolution (see Tattersall and Schwartz, 2000; and Johanson and Edgar, 1996, for booklength retellings of hominid history centered upon this revisionary theme of bushiness vs. linearity). First, before the origin of the genus Homo, the australopithecine clade differentiated into several, often contemporaneous, species, including, at a minimum, three taxa of “robust” appearance (A. boisei, A. robustus, and A. ethiopicus), and at least two of more “gracile” form (A. afarensis and A. africanus). Some of these species, including at least two of “robust” form, survived as contemporaries of Homo. Historically speaking, the death of the single species hypothesis may be traced to Richard Leakey's discovery, in the mid 1970's, of two undeniably different species in the same strata: the most “hyper-robust” australopithecine and the most “advanced” Homo of the time (the African form of Homo erectus, often given separate status on cladistic criteria as Homo ergaster). [Page 910] (Needless to say, no true consensus exists in this most contentious of all scientific professions — an almost inevitable situation, given the high stakes of scientific importance and several well known propensities of human nature, in a field that features more minds at work than bones to study. Nonetheless, despite endless bickering about details, I don't think that any leading expert would now deny the theme of extensive hominid speciation as a central phenomenon of our phylogeny — see Johanson and Edgar, 1996.)
Second, the crucial period of 2-3 million years ago, spanning the origin of initial diversification of the genus Homo, also represents the time of maximal bushiness for the hominid clade, then living exclusively in Africa. The correspondence of a time of maximal speciation with anatomical change of greatest pith and moment in our eyes — that is, the origin of our own genus Homo, with an extensive expansion of cranial capacity — probably records a causal process of central importance in our evolution, not just an accident of coincidental correlation. Vrba's turnover-pulse hypothesis, an extension of punctuated equilibrium (see pp. 918–922), represents just one causal proposal for this linkage. As many as six hominid species may have coexisted in Africa during this interval, including three members of the genus Homo.
Third, the central theme of bushiness persisted far longer than previous conceptions of human evolution had ever allowed, right to the dawn of historical consciousness. Under earlier anagenetic views, European neanderthals marked a transitional stage in a global passage. But under speciational reformulations, and acknowledging extensive anatomical distance between neanderthals and moderns, Homo neandertalensis must be construed as a European offshoot of local Homo erectus populations, with Homo sapiens evolving in a separate episode of speciation, probably in Africa on strong genetic and more tenuous paleontological grounds, and then replacing neanderthals by migration. Similarly, Asian Homo erectus populations may not have passed anagenetically into Homo sapiens, but may also have been replaced by migrating stocks of Homo sapiens, originally from Africa. If the redating of the Solo Homo erectus specimens can be confirmed (as mentioned above), then this Asian replacement occurred at about the same time as the death of neanderthals in Europe.
This information implies the astonishing conclusion, at least with respect to previous certainties, that three human species still inhabited the globe as recently as 40,000 years ago — Homo neanderthalensis as the descendant of Homo erectus in Europe, persisting Homo er
ectus in Asia, and modern Homo sapiens, continuing its relentless spread across the habitable world. This contemporaneity of three species does not match the richness of an entirely African bush with some half a dozen species about 2 million years ago, but such recent coexistence of three human species does require a major reassessment of conventional thinking. The current status of our clade as a single species represents an oddity, not a generality. Only one human species now inhabits this planet, but most of hominid history featured a multiplicity, not a unity — and such multiplicities constitute the raw material of macroevolution. [Page 911]
This recasting of human evolution in speciational terms documents the extent of proffered revision, but the scope of reform gains even greater clarity when we recognize the pervasive nature of the speciational theme as a guide for resolving paradoxes, understanding puzzles of popular misconception, and offering new formulations to break impasses in almost every nook and cranny of discussion about the evolutionary history of our own lineage. When common claims seem askew or confused, I would venture to suggest that the first and best strategy for breakthrough will usually lie in a speciational reformulation for any puzzling issue. Consider just three bugbears of popular confusion in serious newspapers and magazines, and in books on general science for lay audiences, all finding a potentially simple and elegant resolution in speciational terms.
The Structure of Evolutionary Theory Page 144