* Examples of this sort illustrate the important point that drives of directional speciation do not necessarily require a differential number of speciation events along the route of the trend. A directional bias may also arise if numbers of speciation events occur with equal frequency in either direction, but the average phenotypic magnitude of the trending half exceeds the amount of change in the half oriented away from the trend. Such cases may be common when a founding lineage lies near a boundary, and amounts of change become severely constrained in one direction. Thus, for the bacterial mode of life, for example, we may easily imagine (data for an adequate test do not exist, so far as I know) that as many speciation events yield a less complex as a more complex descendant. But so little room exists between the mode and the lower limit that changes to reduced complexity cannot depart far from the ancestral state, while an open range to the right of the mode permits a far greater magnitude of change in the direction of greater complexity. For an actual example, Wagner (1996) documented a general trend to increasing spire height in Paleozoic gastropods, but found an equal frequency of speciation events towards lower-spired and higher-spired daughters. The trend, however, records a bias in amounts of change. For some reason, gastropods that become high spired also experience a marked reduction in the amount of change per speciation event, even though they continue to produce equal numbers of daughters in both directions — whereas low-spired ancestors generate much higher average change per speciation event. The mean spire height of the entire clade therefore increases.
* I may be an arrogant man, but I would never be so pompous as to use the 'royal' we. I cannot separate my views on punctuated equilibrium from those of my colleague and partner in this venture from the start, Niles Eldredge. When I write 'we' in this section, I mean 'Eldredge and Gould.'
* All professions maintain their parochialisms, and I trust that nonpaleontological readers will forgive our major manifestation. We are paleontologists, so we need a name to contrast ourselves with all you folks who study modern organisms in human or ecological time. You therefore become neontologists. We do recognize the unbalanced and parochial nature of this dichotomous division — much like my grandmother's parsing of Homo sapiens into the two categories of 'Jews' and 'non-Jews.'
* Futuyma remains quite skeptical of punctuated equilibrium in general, and I would place him more among our critics than our supporters. But he does accept the empirical pattern, and he is an expert on speciation. Thus, when he developed an original way to resolve the paradox of why punctuations might correlate with events of speciation, even if processes of speciation don't accelerate the rate of evolution, he published his ideas as a constructive contribution to the general debate. Even though Futuyma disagrees with our claims for the general importance of punctuated equilibrium (while he, obviously, does not deny the phenomenon), he has granted us serious attention and has acknowledged the intellectual interest of the debate we provoked — and no one could ask for more from a good critic. Futuyma wrote (1988, p. 225), in stressing the need to integrate “synchronic” approaches as pursued by neontologists interested in evolutionary mechanisms with the “historical” themes favored by systematists and paleontologists — all (to borrow a line from elsewhere) “in order to form a more perfect union.”
We need to identify and to define rigorously questions to which both synchronic and historical evolution can make truly indispensable contributions. Some such questions have already been posed, so we now find systematists and population geneticists converging on the analysis of macromolecular sequences, geneticists publishing in Paleobiology (thanks to the healthy stimulus of punctuated equilibrium), systematists and students of adaptation finding a rapprochement in the use of phylogenetic information to test hypotheses of behavioral, physiological, and other adaptations.
* As an example of the conceptual stranglehold that gradualism once imposed upon such data, the major study done before punctuated equilibrium on the evolution of these rodents presupposed anagenetic gradualism at a constant rate: “This treatment assumes that a regular increase in size continued at approximately the same rate throughout Orellan time” (Howe, 1956, p. 74). When Howe then detected accelerated change at two paleosols marking boundaries of substages within the Orellan, he assumed (without any direct evidence) that the paleosols must mark diastems, or time gaps, compressing a true gradualism of change into the literal appearance of a small hiccup. But Heaton found no evidence for any temporal hiatus at these boundaries.
Heaton's results included both punctuated equilibrium (two stable species changing only in relative abundance) and gradualism (minor size increase within the larger species at the end of its range) in a total pattern, but he concluded (correctly, I think, in my own biased way) that punctuated equilibrium had shown greater utility in challenging previous assumptions that had stymied proper conclusions. He closed his paper by writing (1993, p. 307): “So, Ischyromys displays features of both 'punctuated equilibria' and 'phyletic gradualism' as defined by Eldredge and Gould (1972). But the primary revelation of this study is that what was thought to be a single gradually evolving lineage must now be seen as the replacement of one stable species by another.” (In the fairness of full disclosure, Heaton did his graduate work under my direction. But I really do encourage independence and contrariness, and some of my students have documented gradualism, even in their Ph.D. dissertations, when truly (and for the only time) beholden to my “official” approval — e.g. Arnold, 1982.)
* Interestingly, Barnosky's (1987) published version of his oral presentation refined his conclusion and tabulated a strong majority for punctuated equilibrium, even when compiled from an existing literature biased by previous traditions for ignoring stasis as non-data, and favoring apparent cases of gradualism. In his compendium for Quaternary mammals, Barnosky (1987) found punctuated equilibrium “supported twice as often as phyletic gradualism . . . The majority of species considered exhibit most of their morphological change near a speciation event, and most species seem to be discrete entities.”
* I also wish to reemphasize that I assert no exclusivistic claim in this formulation. Supporters of the hierarchical theory must not repeat the parochial error of their forebears by arguing that their newly specific, higher-level mechanisms can explain everything by reaching down, just as Darwinian traditionalists tried to develop a complete causal theory by extrapolating up. Thus, we do not challenge either the efficacy or the cardinal importance of organismal selection. As previously discussed, I fully agree with Dawkins (1986) and others that one cannot invoke a higher-level force like species selection to explain “things that organisms do” — in particular, the stunning panoply of organismic adaptations that has always motivated our sense of wonder about the natural world, and that Darwin described, in one of his most famous lines (1859, p. 3) as “that perfection of structure and coadaptation which most justly excites our admiration.” But should we not regard as equally foolish, and equally vain (in both senses of the word), any proposal that insists upon explaining all “things that species and clades do” as extrapolated consequences of organismic adaptation? I would not invoke species selection to explain the marvelous mechanics of beetle elytra, but the same theme of appropriate scale also leads me to equal confidence that the excellent adaptive design of beetle organisms cannot fully explain why this order so vastly predominates in species diversity, even among the most speciose of all metazoan classes — to the point of inspiring Haldane's canonical quip about God's “inordinate fondness” for these creatures (see Gould, 1993a, for an exegesis of this famous anecdote).
* One might argue that this focus only records another of Bacon's idols rather than an evident empirical reality. Bacon's idola tribus, or idols of the tribe, refer to mental biases deeply rooted in inherent modes of mental functioning, or human nature itself. Humans are pattern-seeking and story-telling creatures — and we prefer to tell our stories in certain modes that may reflect particula
r cultural traditions as well as universal preferences of thought. We shun randomness and non-directionality in favor of stories about movement in particular directions for definable reasons subject to moral judgment. We compiled the entire Bible as a grand and extended narrative in this mode, and then granted just one uncomfortable chapter to Ecclesiastes as the loyal opposition, where “time and chance happeneth to all” and “there is no new thing under the sun.” Thus, our chosen focus upon trends in the paleontological record may only record their salience in piquing our interests and preferences — and not a genuinely high relative frequency among all clades in nature. This subject deserves a great deal of thought and extended study. A remarkable article by Budd and Coates, 1992, on the predominance of non-trending in the evolution of montastraeid corals may point the way to substantial reform — see p. 937.
* Soon after I wrote this section, Science published a special issue on evolution (25 June 1999), featuring the work of Lenski's lab in a news article entitled, “Test tube evolution catches time in a bottle.” The twelve populations have now been evolving for 24,000 generations. Although all have shown similar increases in cell size and fitness, the genetic bases of change have been highly varied and unpredictable. Moreover, alteration in environmental and adaptive regimes yields no common response. When, after 2000 generations of growth on glucose (with similar evolutionary responses), the 12 populations were switched to a different sugar (maltose), some populations flourished, but others grew poorly. After 1000 generations on maltose, all twelve populations did improve in fitness, but not nearly so much (and, more importantly, not nearly so consistently) as on glucose. The starting genotype for the 12 populations had been identical for the first experiment with glucose, but different (after 2000 generations of evolution on glucose for each population) for the initiation of the subsequent maltose experiment. Apparently, any departure from simple and controlled experimental conditions towards the genetic and environmental variation invariably encountered in the natural world greatly decreases the predictability, while emphasizing the contingency, of outcomes.
* I dare not even begin to enter the deepest and most difficult of all issues raised by differences between scientific and humanistic practice: why does the history of scientific ideas, even when proceeding in a punctuational mode, marked by quirky, unpredictable and revolutionary shifts, undeniably move to better understanding (at least as measured operationally by our technological successes) — that is, and not to mince words, to progress in knowledge — whereas no similar vector can be discerned in the history of the arts, at least in the sense that Picasso doesn't (either by any objective measure or by simple subjective consensus) trump Leonardo, and Stravinsky doesn't surpass Bach (although later ages may add new methods and styles to the arsenals of previous achievement). The naive answer — that science searches for a knowable, objective, external reality that may justly be called “true,” whereas art's comparable standard of beauty must, to cite the cliche, lie in the eye of the beholder — is probably basically sound, and probably explains a great deal more of this apparent dilemma than most academic sophisticates would care to admit. (In this belief, I remain an old-fashioned, unreconstructed scientific realist — but then we all must take oaths of fealty to our chosen profession.)
But I also acknowledge that the question remains far more complicated, and far more enigmatic, than this fluffy claim of such charming naievete would indicate. After all, we only “see” through our minds (not to mention our social organizations and their pervasive biases). And our minds are freighted with a massive cargo of all the inherent structural baggage that Kant called the synthetic a priori, and that modern biologists would translate as structures inherited from ancestral brains that built no adaptations for what we designate as “consciousness.” In this light, why should we be “good” at knowing external reality? After all, our vaunted consensuses — and on this point, Kant remains as modern as the latest computer chip — may record as much about how our quirkily constructed brains must parse this “reality,” as about how external nature truly “works.” But enough of unanswerable questions! I only note that Kuhn himself raises this great issue in his closing thoughts on the special character of science: “It is not only the scientific community that must be special. The world of which that community is a part must also possess quite special characteristics, and we are no closer than we were at the start to knowing what these must be. That problem — What must the world be like in order that man may know it? — was not, however, created by this essay. On the contrary, it is as old as science itself, and it remains unanswered.”
* Interestingly, the King James Bible (a 17th century document) uses the word constraint ten times, nine in the positive sense of directing or forcing an action in a particular way. The most popular of 20th century “updates,” the Revised Standard Version, keeps the word “constrains” in some passages, but often changes the King James entry to “urge,” “compel,” or “make,” to emphasize the obviously intended positive meaning.
* To make a personal statement, I was a dyed-in-the-wool adaptationist during my undergraduate and graduate years, but this particular claim — advanced by all my teachers as an article of faith — always bothered me. I saw no reason beyond an overweening faith in the strength and ubiquity of selection (Weismann's Allmacht all over again) to assume that variation in supply of raw material should exert so little effect on rates of evolutionary change — even though I remained quite willing to believe that natural selection always set the direction of change.
* As one more example of how “constraint” terminology can be biased by assumptions that equate “ordinary” evolution with selection and adaptation, Schwenk (1995, p. 251) argued that constraints “can have either negative or positive evolutionary effects at the lineage level (i.e., hamper or promote organismal adaptation).” I doubt that Schwenk truly believes what he literally says — that any “evolutionary effect” hampering organismal adaptation must be labeled as “negative” — but his statement illustrates the common conception that evolution “ought” to build better adapted organisms, and that any other result must be regarded as disappointing or somehow wrongheaded. But all manner of highly interesting phenomena beyond (and sometimes opposed to) organismal adaptation pervade such a richly varied, universal, many-leveled, highly complex process as organic evolution. Do we really want to label this lifeblood of fascination for our favorite subject as negative?
* The verbal pun upon “doxology” — the short, formulaic, and unvarying, prayers of Christian liturgy — inevitably comes to mind, although we trust that the norms of science will prevail to impose substantial and interesting improvements upon the current hoxology.
* “Whole animal biologists,” including the author of this book, can only experience enormous hope and gratification when colleagues trained in molecular and experimental traditions recognize the utility of data so often ignored and disparaged as antiquarian or superannuated. In fact, such a pattern has often been repeated in the history of science, as when the initial recognition of Mendelian mutations led early geneticists to reexhume old data long dismissed as mere description of phenomenological oddity — for example, the literature (dating to the earliest days of scientific publishing) on developmental anomalies. When molecular biologists value such classical data more highly (and utilize them more fruitfully) than practitioners in the classical fields manage to do themselves, then we may truly hope for an integrated biology based on the prospect, so often expressed but so little realized until recently, that molecular and organismic biology might finally consummate a union on the common field of development.
* Such metaphors from distant professions always pay the price of their utility in suggestive analogy by their occasional capacity to be confusing in their invocation of different systems with different causal bases. At this point, an architect who loves the strict Darwinian model would say: “but why is he calling a Corinthian
column an internal constraint rather than an adaptation? After all, I was commissioned to build a bank so I chose this element as a good fit with my project. The column is therefore an optimal adaptation based on my skilled selection.” But I then reply, “Yes, for you and your building in a non-Darwinian system. But if an organism carries the genes for Corinthian columns as a deeply intrinsic aspect of its developmental system, enmeshed in both upstream and downstream cascades of regulation as both result and promoter, then the organism is stuck with this eminently serviceable device, and can only construct itself in certain ways under the constraints imposed by this inherited, internal element.”
* I apologize for this second citation of a long quote in an overly ample book — see p. 527; but its uncanny appropriateness in these two different contexts leads me to beg your indulgence for this redundancy.
* D'Arcy Thompson couldn't resist a frequent Latin (or Greek, or Italian, or French, or German, or whatever) quotation, always untranslated. This little paragraph-ending phrase continues his theme of false modesty by proclaiming: At this point I close my composition and put back my oar.
* During the past 20 years, this term has been fading from use, in part (I believe) as a consequence of such critiques as this present section (and Gould and Vrba, 1982). Current graduate students may now encounter this term only rarely. But when I was a graduate student in the mid 1960's — admittedly a while ago, but not exactly Mesozoic either — preadaptation was a standard term in constant and continual use (Bock, 1959, for example). On the same theme of shifting terminology, reflecting a declining faith in the exclusivity of adaptation by natural selection as the basis of all evolutionary results, another common usage of my graduate years has completely disappeared. (I find it hard, even a bit embarrassing, to recall that we ever spoke so uncritically.) But, in these years (as all evolutionists of my generation will affirm), we used the term “adaptation” as a simple descriptive synonym — indeed the preferred name in professional circles — for any feature of a phenotype, with no intended implication about the origin or utility of the item. Merely to exist was to be an adaptation. We would, talking only descriptively about morphological features, the forelimbs of theropod dinosaurs, for example, say: “This adaptation was larger in Allosaurus than in Tyrannosaurus.”
The Structure of Evolutionary Theory Page 228