Stephen Jay Gould
Stephen Jay Gould; drawing by David Levine


An astonishing fact about Charles Darwin’s theory of evolution is that it was conceived in the absence of two major kinds of support that would have been helpful to it—evidence from the then meager fossil record and knowledge of the mechanism of inheritance. The endurance of a theory so conceived—almost from first principles—has excited the admiration of generations of biologists. It has also given them the opportunity, as discoveries in genetics and paleontology have been made, to determine if this new evidence merits a revision of Darwinism, as the theory is now known.

The Structure of Evolutionary Theory is the latest—and perhaps the grandest—of such attempts. In effect it is a summation of Stephen Jay Gould’s life work, building on Darwinism to provide a novel synthesis of how evolution has shaped the living world. Explaining his motivation in writing it, Gould, who was trained as a paleontologist, says it

does fire my very best shot in the service of…general theory. I am a child of the streets of New York City; and although I reveled in a million details of molding on the spandrel panels of Manhattan skyscrapers…I guess I always thrilled more to the power of coordination than to the delight of a strange moment—or I would not have devoted 20 years and the longest project of my life to macroevolutionary theory rather than palaeontological pageant.

Gould employs two metaphors to assist the reader in understanding the complex theory that the book develops: the Duomo of Milan and a fossil coral. The Duomo, Gould says, is similar to Darwin’s evolutionary theory in that it has been built on in ways that add significantly to it but nonetheless leave the original form and intent recognizable. The fossil coral (see illustration on page 54), which was discovered near Messina and figured in Agostina Scilla’s La vana speculazione disingannata dal senso, published in 1670, has special significance for Gould because its branches mimic Gould’s theoretical structure.

Gould envisages the trunk and each branch of the coral as representing a hierarchy within Darwinism. The trunk is Darwin’s theory of evolution by natural selection itself. The first three branchings represent the three fundamental principles of Darwinian logic, which Gould characterizes as “agency [the central branch], efficacy [the left branch] and scope [the right branch].” The three fundamental principles so characterized arise from Darwin’s thesis that evolution produces new species by natural selection (efficacy) acting on individuals (agency), resulting in small changes that accumulate over very long periods of time (scope).

The potentially confusing structure of The Structure of Evolutionary Theory itself can only be understood in these terms. Gould says of his book that it “cycles through the three central themes of Darwinian logic at three scales—by brief mention of a framework in [the introduction], by full exegesis of Darwin’s presentation in Chapter 2, and by lengthy analysis of the major differences and effects in historical (part 1) and modern critiques (part 2) of these three themes in the rest of the volume.” To many readers, the arguments that Gould mounts in support of his evolutionary synthesis will not be new. Most have appeared in articles and exchanges over the past quarter-century.1 Here, they are brought together for the first time.

The historical overview presented in Part One consists of a reflective commentary on thought relating to evolution as it has developed over the past three hundred years. Gould provides an admirably comprehensive review of the literature, dealing with great scientists such as Richard Owen, Georges Cuvier, and Étienne Geoffroy St. Hilaire, as well as many lesser-known but interesting and worthy contributors. This section includes many fresh insights into how thinking about evolution slowly changed, and who built on whose prior work. Central to the discussion is an exegesis of Darwin’s work, where considerable emphasis is given to the idea that Darwin was a gradualist; that is, that he saw evolution resulting from the aggregation of countless small changes over a vast period of time. This is followed by a discussion of the ideas of Darwin’s “brilliant and eccentric cousin” Francis Galton, who challenged Darwin’s concept of gradualism by suggesting that species come into existence almost instantaneously. Galton illustrated his idea with the metaphor of the polyhedron; a shape that rests stably on any of its facets, but that when disturbed switches rapidly from one resting position (facet) to another. Species, Galton argued, are likewise stable, not changing or evolving until something happens; then they are rapidly transformed into new (stable) species. Here Gould finds an intellectual predecessor, for his theory of punctuated equilibrium (of which more presently) is more akin to Galton’s polyhedron than to Darwin’s gradualism.

One fascinating aspect of evolutionary thought to which Gould gives prominence is the idea that Darwin’s concept of natural selection owes much to Adam Smith’s economic arguments as presented in The Wealth of Nations. It is known that Darwin devoted much time in the late 1830s to studying Smith. Gould goes so far as to say that “the theory of natural selection is, in essence, Adam Smith’s economics transferred to nature,” adding that while “Adam Smith’s economics doesn’t work in economics,” it does admirably well in “amoral nature.”


Part Two of The Structure of Evolutionary Theory returns to the three principles of Darwinian logic in the context of contemporary evolutionary debate. Here the author and his collaborators take center stage as the reformulators of two of these three principles. The theory of punctuated equilibrium, conceived by Niles Eldredge and Gould, is seen as having revised the principle of agency, because Gould believes that it lays the theoretical groundwork for natural selection to act upon species as well as individuals. Gould and E.S. Vrba’s ideas on internal constraint are said to have performed the same service for efficacy, because they challenge the omnipotence of natural selection in shaping species. Of the remaining principle, that of scope, Gould says that although his contribution to the understanding of mass extinction and catastrophic events which challenge Darwin’s concept of gradualism “cannot claim much novelty…. I do explicate [their significance] perhaps more fully than before.”

Niles Eldredge and Gould first coined the term “punctuated equilibrium” in 1971 and published it the following year. The theory seeks to explain a persistent pattern in the fossil record whereby a species suddenly appears, then persists unchanged for a very long time before going extinct. This pattern is seen in a wide variety of contexts, from marine creatures such as shellfish and sea urchins to mammals and birds. Punctuated equilibrium posits that these species come into existence relatively rapidly (over tens of thousands of years), though just how (and indeed if) this happens is hotly debated. An opposing explanation is that these species have evolved much more slowly somewhere else, and their “sudden” appearance is the result of migration. While, as Galton’s polyhedron suggests, the concept of punctuated equilibrium was not entirely new to paleontology, Eldredge and Gould’s formulation of it was timely and coherent. Even among its supporters, however, argument has raged over its significance, with many questioning whether it really challenges Darwin’s concept of gradualism. (After all, tens of thousands of years is sufficient time for species to evolve “gradually.”) Most researchers, though, recognize that the concept has been invaluable in encouraging paleontologists to examine the fossil record with a rigor and attention to detail that previously was largely lacking.

Punctuated equilibrium has forced paleontologists to focus not only on the origin of species, but also on their often long, unchanged persistence in the fossil record. Before punctuated equilibrium, this phenomenon tended to be accepted as simply the way things were—yet it is an intriguing problem. One of the best examples given by Gould concerns the mammals (including rhino, pig, and camel relatives) of the White River Chronofauna of the American West, which endured almost unchanged through a period of astounding climatic upheaval. During this time the rain forests of Nebraska were transformed into forested grassland and the mean annual temperature dropped by 13 degrees Celsius. Revelations such as this remind us of how much we have to learn about our planet’s past.


What is particularly satisfying about reading Gould’s views of punctuated equilibrium is the sense he gives that a fundamental problem—that of how species originate and then persist over geological time—is not yet fully understood. If young biologists ever reach deep into his vast tome, they may find here materials with which to build careers.

Following a thorough treatment of the literature relating to punctuated equilibrium in nature, Gould brings his discussion into the social realm, debating how the concept might apply to changing human cultures and even to businesses in complex societies. He notes, not without some chagrin, that creationists have welcomed the theory because it has enabled them to argue that the fossil record does not show us “evolution in action” and that therefore evolution does not exist. The first part of this observation is indeed a corollary of the theory, which posits that the changes leading to new species usually happen so rapidly that they go undetected, but the second is clearly an unwarranted “leap of faith.”

Gould’s discussion of punctuated equilibrium concludes with an overview of nonscientific objections to the theory, including the accusation that he developed it as part of a Marxist political agenda. Gould resents “this absurd misreading,” writing that his politics are “a private matter that I do not choose to discuss in this forum.” Under the heading “THE MOST UNKINDEST CUT OF ALL” he discusses the dismissal of his theory as both “trivial and devoid of content,” a charge he intensely resents and feels is deeply unfair.


On the twenty-first anniversary of the publication of the theory of punctuated equilibrium, Gould and Eldredge wrote in Nature: “In developing punctuated equilibrium, we have either been toadies and panderers to fashion, and therefore destined for history’s ashheap, or we had a spark of insight about nature’s constitution.” In The Structure of Evolutionary Theory Gould appears to have revised this view, perhaps as a result of the criticism that the theory is extraordinarily difficult to test, because life is so diverse and so many differing interpretations of the fossil record are possible. He now believes that no one expects an either-or result, but only (eventually) a sense of what proportion of life’s diversity evolves gradually and steadily, and how much in leaps and bounds.

The emphasis on punctuated equilibrium opens the way for The Structure of Evolutionary Theory’s most audacious proposal—the idea of species selection. Gould sees punctuated equilibrium as a prerequisite for species selection because it bestows on species some of the properties inherent in individuals. Thus Gould envisages a sort of life cycle of species: born via punctuated equilibrium, they live a life in mature (stable) form, and then die through extinction. Species selection is a central aspect of the book, and by it Gould means that species as a whole—not just individuals or populations—are being acted on by natural selection.

Gould admits that few (if any) examples of species selection have come to light—but argues that this is because scientists have not looked hard enough. Consequently we must make do with hypothetical examples, of which the author gives us one of his own making. It has to do with variability within species, which Gould sees as the best example of a “species character.” This is a necessary prerequisite of species selection since otherwise natural selection at the level of species would have nothing to act on:

Suppose that a wondrously optimal fish, a marvel of hydrodynamic perfection, lives in a pond. This species has been honed by millennia of conventional Darwinian selection, based on fierce competition, to this optimal organismic [individual] state. The gills work in an exemplary fashion, but do not vary among individual organisms for any option other than breathing in well-aerated, flowing water. Another species of fish—the middling species—ekes out a marginal existence in the same pond. The gills don’t work as well, but their structure varies greatly among organisms. In particular, a few members of the species can breathe in quite stagnant and muddy waters.

Organismic selection favors the optimal fish, a proud creature who has lorded it over all brethren, especially the middling fish, for ages untold. But now the pond dries up, and only a few shallow, muddy pools remain. The optimal fish becomes extinct. The middling species persists because a few of its members can survive in the muddy residua. (Next decade, the deep, well-aerated waters may return, but the optimal fish no longer exists to reestablish its domination).

Gould argues that the persistence of the more variable, middling species and the extinction of the less variable, optimal form result from natural selection acting upon entire species rather than individuals, because the outcome ensues from the greater variability (a character of the species) of the surviving “middling fish.” While it is true that entire species die out because they cannot adapt to new circumstances (such as dinosaurs during large asteroid impacts), this is not, of itself, an argument for species selection. For me, this particular example raises more questions than it resolves. For instance, the supposed “species character” of greater variability of the middling fish is itself eliminated by the drought-caused natural selection, because just a few individuals at the extreme end of variation survive. Does this mean that the survivors constitute a new species, with different “species characters”? Not even punctuated equilibrium posits such virtually instantaneous generation of species.

Gould’s concept of species selection is embedded within a second concept of a hierarchy of selection, which leads from the gene to (presumably) the whole of life itself. This brings him into direct conflict with that formidable Gouldian nemesis Richard Dawkins, whose selfish gene theory posits that natural selection acts on genes rather than on individuals or groups of individuals. By way of illustration, on the subject of sexual selection among birds, Dawkins says, “‘Pressure’ from choosy females drove the evolution of male pheasants’ sumptuous feathers. What this means is that a gene for a beautiful feather is especially likely to find itself riding a sperm into a female’s body.”2 Dawkins’s hypothesis has arguably revolutionized evolutionary theory, and while actively contested, it enjoys widespread acceptance.

One of the more controversial aspects of Dawkins’s work is the concept of “memes,” which he defines as being the cultural equivalent of genes. The concept is intriguing because it suggests a new way of examining how ideas “behave” in human societies. Interestingly, Gould dismisses Dawkins’s selfish gene theory as “an impotent meme,” claiming that the entire concept is based on a flaw of logic—that of confusing what Gould calls “bookkeeping” with “causality.” I have tried to tease out exactly what Gould means by this, but find it hard to pin down. At its base, though, is Gould’s idea that natural selection acts on individuals or groups (such as species) and only by way of them on genes themselves. Not only does Gould see a fatal flaw of logic here, but he goes on to argue that the recent discovery that organisms possess fewer genes than previously thought is a death knell to Dawkins’s hypothesis. Yet this is such a new and rapidly changing field of research—particularly on the structure and function of proteins—that announcing the demise of the selfish gene theory on these grounds seems premature.

As I pondered the Gould–Dawkins rivalry, it occurred to me that Gould’s hypothetical example of the pond’s two fish species might explain more about Gould and Dawkins than it does about evolution. Read at another level (and surely the playful Gould is mischievous enough to have given it a second meaning), the “wondrously optimal fish” might be Richard Dawkins and his successful selfish gene theory (which indeed takes a narrow focus on natural selection), while the “middling fish” could be the author himself (who sees natural selection operating on a wide field). This left me wondering what revolution Gould thinks will muddy the waters and eliminate the competition, thus permitting the Gouldian “middling fish” to triumph? While I suspect that the concept of species selection is destined not to survive, at least in its present form, this is such a strongly contested field of biology that I would certainly not lay money—even at short odds—against Gould’s eventual triumph.

Gould’s discussion of the second of the fundamental Darwinian principles, efficacy, centers on the importance of historical and structural constraints on development, and the curious phenomena of spandrels and “exaptation,” of which spandrels are an example. A principal theme here is that many features of organisms do not arise through natural selection, but through contingency or the inherent nature of the organisms themselves. One of Gould’s favored examples concerns male nipples and the clitoris, both of which, he argues, exist not because of their function, but because they are necessary elements of the essential body plan from which adult males and females differentiate.

Gould argues that historical constraint (the legacy of all that has gone before in a biological lineage, including its genes) is important. While this may seem self-evident, there are other factors to consider, not the least of which is ecology. Gould attacks Theodosius Dobzhansky’s ecologically based theory of “adaptive peaks” because it cedes too much ground to natural selection and ecology. To illustrate the point, Gould uses the example of cats and dogs. Dobzhansky argues that cats and dogs exist as discrete types, with not much in between, because ecological niches exist for dog-like and cat-like creatures, but not for in-between kinds. Gould, however, argues that the two discrete types owe their existence to historical constraint in the form of their own, separate inheritances from an ancestral dog-like and cat-like creature.

There is a simple test of these competing theories. If ecology were responsible, it would be reasonable to expect that other mammal groups would have thrown up cat- and dog-like forms during their evolution, with little in between—and indeed, this is precisely what we see. Both South America’s and Australia’s marsupials have given rise to dog-like and cat-like species, while Madagascar’s carnivores (which are related to ferrets) have given rise to the cat-like fossa. Here is good evidence that Dobzhansky’s theory, in this instance at least, is preferable.

If the word “exaptation” is unfamiliar to some readers, it is because it has been only recently coined, by Gould and Vrba in 1982. To use Gould’s definition, it denotes “features coopted for a current utility following an origin for a different function (or for no function at all.)” By way of example, he cites the case of an African lizard whose extremely flattened head is an adaptation to life in crevices, but which also assists the creature in gliding. Spandrels—an architectural term that Gould has introduced to biology—are an extreme form of exaptation, in that they may not result from natural selection at all. Like the spandrels on the interior surface of a cathedral dome, they just have to be there as a result of the underlying structure of a dome atop four supports; yet whether in snails or mammals or other animals, we find spandrel-like structures that have acquired functions.

Perhaps the best example comes from snails, Gould’s own special subject. Certain groups of snails have co-opted the umbilicus (a tube around which the shell coils) as a chamber for protecting their eggs. The umbilicus must exist if the shell is to coil, yet it had no internal function of its own until a few species co-opted it as an egg chamber. Much later in The Structure of Evolutionary Theory, Gould discusses how his sense that the inherent properties of organisms (such as spandrels), rather than deterministic and all-explaining adaptation, is important personally to him because it allows room for individuality and essential humanity in his own worldview.

The third Darwinian principle, scope, is dealt with more cursorily than the other two. At the heart of the matter for Gould is the existence of seemingly random events such as mass extinctions and other similar catastrophes. He explains that they are “more frequent, more rapid, more intense, and more different” than Darwin ever imagined. Moreover, their discovery represents an important extension of Darwinian logic, because they further disrupt Darwin’s concept of “gradualism.” Yet these events are not strictly biological in nature—and thus outside Gould’s own expertise—so he deals with them only briefly.


This year also marks the publication of a very different Gouldian production. I Have Landed is Gould’s tenth and final volume of collected essays, most of which were written for Natural History magazine. These essays have entranced millions, from company presidents to penitentiary inmates, with the wonders of evolution. For the past twenty-five years Gould has produced an essay a month for Natural History, not stopping, as he says, for “cancer, hell, high water or the World Series.” In between he has also found time to write for Science, Time, The New York Times, The New York Review, and antiquarian book catalogs, as well as completing ten monographs on various aspects of evolution. But human energy is not limitless, and in I Have Landed Gould indicates that this prodigious flow of words is slowing down.

The book’s introduction celebrates his fascination with numerical coincidence. In earlier works, Gould’s interest in this topic was often concerned with baseball, but here it becomes altogether more personal. The book’s title derives from the coincidence of the author’s grandfather’s arrival in New York on September 11, 1901, and Gould’s interrupted pilgrimage to Ellis Island on the terrible centenary of that day. Other numerical coincidences are at play in Gould’s pattern of publication. His three hundredth essay appeared in the millennial issue of Natural History, followed in 2002 by his tenth book of essays and his tenth monograph. On realizing this, he “felt the double whammy of an equally ‘exact’ and notable twenty-five years (a quarter of the square of our decimal base) between two odd and fortuitous conjunctions in life’s passage.” He took this to mean that “something must be beaming me a marching order to move on.”

The new collection of essays brings us up to date with the thinking—and indeed even the personal experience—of one of America’s best-loved scientists. Gould has fashioned a admirable “second” career as a historian and philosopher of science, and this material dominates the compilation. Topics range from the nature of excellence as illustrated by the operas of Gilbert and Sullivan to Freud’s concepts of mental development, and the brief friendship between Karl Marx and the biologist E. Ray Lancaster (“Resolving Evolution’s Oddest Coupling”). In comparison with Gould’s earlier collections, many of the essays in I Have Landed lack an urgent scientific point; the prose is more detailed and less intense in tone.

While there is rarely the sense that Gould has stepped out of his depth, the essay I found least satisfying is the twenty-fourth, titled “An Evolutionary Perspective on the Concept of Native Plants.” Arguing that indigenes may be less well adapted to their local conditions than introduced organisms, Gould bases his case on the observation that introduced plants and animals can be wildly successful, often to the detriment of the locals. What is astonishing about this argument is that it takes no account of ecology. Gould fails to consider, for example, that California’s eucalyptus trees may have succeeded because they left behind in Australia all of their predators (such as koalas, which can kill trees), diseases, and competitors. Nor does he consider that Australia’s marsupials may have been disadvantaged, and the introduced placental mammals given an advantage, by the land management practices introduced by European colonists. These seem fundamental oversights.

In the closing lines of the title essay, Gould talks of his “own childhood dream—something that once seemed so mysteriously beyond any hope of realization to an insecure little boy in a garden apartment in Queens—to become a scientist and to make, by my own effort, even the tiniest addition to human knowledge of evolution and the history of life.” No one can deny Stephen Jay Gould this achievement, and there will be many who have been inspired by the spark of his particular genius as an essayist and popular communicator of the fascination of science.

How are we to judge Stephen Jay Gould’s additions to the knowledge of evolution and the history of life, particularly as presented in The Structure of Evolutionary Theory? I could not help but feel that the diverse matter fails to gel into a cohesive theory; yet remarkably Gould leaves us with a self-acknowledged paradox that may explain why this is so. He says that one of his principal themes has been the importance of nonadaptational features of evolution such as spandrels, and of the effects of accidents such as asteroid-induced mass extinctions on the course of evolution. By their very nature such phenomena resist being part of an inclusive evolutionary theory, yet they remain central to Gould’s view. If he is correct, then evolution may be one field of science that is not amenable to the application of a grand, overarching theory. This seems an entirely plausible if disconcerting possibility.

One could wish that the editors of The Structure of Evolutionary Theory had been more concerned about the book’s own structure. At 1,433 densely printed pages and almost five pounds in weight, Gould’s mammoth book takes too long to make its points. A very large book can, of course, be justified for those wishing to understand Gould’s corpus of ideas; but the lack of a clear, succinct summary of his work and the mostly unhelpful index (which even lacks an entry for Gould’s beloved Tyrannosaurus) frustrate this goal. Still, one of the joys of reading about good science is the chance not only to observe how scientific theory works, but also to participate in the workings of the mind behind the words. In Gould’s I Have Landed and The Structure of Evolutionary Theory, the reader will find such joy in abundance.

This Issue

May 23, 2002