Leonardo’s Mountain of Clams and the Diet of Worms (20 page)

Kovalevsky had no misgivings about the factuality of his sequence.
He wrote about the first step in his English monograph of 1874: “There can be, in my opinion, no reasonable doubt that the Horse descended from the
Paleotherium
.” He expressed equal assurance about step two in his French treatise of 1873: “In its skeleton,
Anchitherium
is a genus so intermediary, so transitional, that if the theory of transmutation were not already so firmly established, this
genus would have provided one of the most important pillars of support” (my translations throughout). His German monograph of 1876 exuded confidence in the full sequence of four steps: “We have a form in the upper Eocene, which must surely be counted as an ancestor of horses; and we now have apparently persuasive data that this form,
Paleotherium medium
, evolved through the Miocene
Anchitherium
and
Hipparion
to modern horses.” The French monograph of 1873 made the same claim in a more succinct and forceful manner: “There can be no reasonable doubt that the four forms
—Paleotherium medium, Anchitherium, Hipparion
and the horse—form a relationship of direct descent.”

I would add a third point, more technical and professional and therefore not part of Kovalevsky’s public reputation, to
explain his high status among working paleontologists. Kovalevsky won kudos for his meticulous attention to descriptive detail. I have quoted the few general comments from his publications; the bulk of his text provides exhaustive information about every bump on every bone and every nubbin on every tooth. Moreover, Kovalevsky did not engage in such laborious work for the mindless and obsessive reasons
that motivate many practitioners, but explicitly because he viewed this level of detail as a prerequisite for adequate documentation of such grand ideas as evolution, natural selection, and adaptation. Directly following his paean to Darwin from his English monograph (quoted earlier), and as an excuse for such flowery generality, he wrote: “The foregoing observations are intended only as a sort
of apology for the somewhat minute osteological details into which it seemed to me necessary to enter in my description.”

Kovalevsky began his French treatise of 1873 with his most explicit defense of detailed empiricism as a scientific method. He states that he was able to succeed in documenting the intermediary status
of Anchitherium
only because he had so many more fossils to study:

But Kovalevsky, true to the foremost ideal of objectivity, then denies that any a priori preference for evolution could have influenced his interpretations. In the
best tradition of Sergeant Friday—“just the facts, ma’am”—he states that he had reached an inescapable conclusion by freeing his mind of preconceptions and letting the fossils speak in their undeniable factuality:

With this brave claim, we come to a wonderful irony that raises the story of Vladimir Kovalevsky from an antiquarian tale for the heartstrings, to a lesson of central relevance for the practice of modern science. Kovalevsky developed a case study that remains preeminent as a triumphant tale of evolution documented in the fossil record. (Does any paleontological
museum
not
include an evolutionary “line” of horses?) He expressed complete confidence in his sequence of four European genera. He won his reputation for unsurpassed care in meticulous and detailed description. He strongly espoused the classic doctrine of utter objectivity in observation, and he claimed that the fossils themselves had dictated an unimpeachable factual conclusion.

And Kovalevsky
was wrong—for a clear and interesting reason that brings no discredit upon his work. He and Huxley had regarded the sequence of four European fossils as an unbroken lineage of direct transformation—that is, as a series of ancestors and descendants. They did not realize—and could not have done so in the absence of published evidence—that horses had evolved in America and had migrated several times
into Europe. The three “ancestors” of Kovalevsky’s sequence—
Paleotherium, Anchitherium
, and the European
Hipparion
—are all side branches that migrated to Europe and became extinct without issue in their new and peripheral home. (As a historical irony that would cost the Aztecs dearly and greatly facilitate the bloody schemes of the Conquistadores, horses then died out in America, leaving the Old
World descendants of yet another migration as the source of all modern horses.) Kovalevsky’s four genera did represent a temporal series of stages in a trend, but not—as he had asserted so confidently—a direct sequence of filiation. He had, so to speak, misidentified an indirect series of my paternal grandpa’s brother, my mother’s brother, and me, as a direct genealogy of my paternal grandpa, my
father, and me.

The denouement occurred in an interesting manner when T. H. Huxley made his only trip to the United States to present an inaugural lecture for the opening of Johns Hopkins University, and to participate in several other events—including a lecture in New York on the evolution of horses—surrounding America’s centennial celebration of 1876. He visited the American paleontologist
O. C. Marsh at Yale and, in great excitement mixed with a bit of chagrin, saw enough beautifully transitional fossils to know that Europe had been a periphery, and America a breeding ground. Marsh later wrote of his magnificent show-and-tell:

Huxley scrapped his New York lecture, and hastily prepared a new version.

If science, as a stereotype proclaims, truly operated as an automatic process of objective documentation, then we should castigate Kovalevsky—for
he firmly advertised his conclusions as factually driven and truly proven thereby. Moreover, if theories rest upon crucial cases, as another stereotype holds, then don’t we call evolution itself into question by scratching Kovalevsky’s horses from the race of life? Doesn’t Kovalevsky then become an incubus rather than a hero—an unwitting foil for creationists rather than a preeminent student
of evolutionary paleontology—because his error might give comfort to the enemy?

We need, instead, to reject these simplistic stereotypes and heed Marsh’s wise statement about “the generosity of true greatness.” The only universal attribute of scientific statements resides in their potential fallibility. If a claim cannot be disproven, it does not belong to the enterprise of science. New fields
(like evolutionary paleontology in Kovalevsky’s day) based on imperfect data (the fossil record) are especially subject to error—and brave scientists must do their best and risk the consequences, secure in the knowledge that corrected mistakes (however personally embarrassing to the perpetrator) bring as much clarity as valid discoveries.

Moreover, truly grand and powerful theories—evolution
preeminently among them—do not and cannot rest upon single observations. Evolution is an inference from thousands of independent sources, the only conceptual structure that can make unified sense of all this disparate information. The failure of a particular claim usually records a local error, not the bankruptcy of a central theory. Kovalevsky mistook a series of collateral relatives for a true genealogical
sequence, but the notion of genealogy itself does not fall thereby. If I mistakenly identify your father’s brother as your own dad, you don’t become genealogically rootless and created
de novo.
You still have a father; we just haven’t located him properly.

With this tale of Kovalevsky’s fruitful error, we come face to face with the interesting and conjoined issues of the relationship between
fact and theory in science, and the phenomenon of being right for the wrong reason. Theory and fact are equally strong and utterly interdependent; one has no meaning without the other. We need theory to organize and interpret facts, even to know what we can or might observe. And we need facts to validate theories and give them substance. Kovalevsky made an eminently useful and rather wonderful mistake.
He grasped the explanatory power of a new and correct theory that would reformulate all of biology—and he yearned to apply this idea to difficult but crucial data. He reached a premature conclusion that was only half wrong in misidentifying collateral relatives as direct ancestors. But he provided the first powerful example of a workable methodology (inferences about adaptation from anatomical
changes in paleontological sequences) that can, with better data, document and support the most important theory subject to adjudication by the fossil record.

Lest anyone doubt the power of being right for the wrong reason, consider the central incident in the intellectual life of Kovalevsky’s hero, Charles Darwin. Thanks to his obsessive habits of record keeping, we can reconstruct the sources
of Darwin’s eureka when, in early 1837, he recognized that evolution must be true. Darwin first learned that he had been wrong in allocating many small birds from the Galápagos Islands to several distantly related families. All, in fact, were finches. How could adjacent islands house separate species of such closely related birds? He then learned that the distinctive rhea (a large flightless bird)
he had collected in southern Patagonia belonged to a new species (named
Rhea darwinii
in his honor by his ornithological consultant, John Gould). Why, Darwin wondered, did two so closely related birds share contiguous geographic territories, the ordinary rhea to the north, and Darwin’s new species to the south?

Darwin pondered these two examples of geographic replacement among closely related
modern birds. He then made a brilliant analogy: If both finches and rheas replace each other in space, then shouldn’t temporal succession also occur in continuity—that is, by evolution rather than successive creation? Darwin had collected large fossil mammals in South America. He regarded one genus, later named
Macrauchenia
, as a relative of the modern guanaco. If the two rheas are maximally related
and geographically contiguous, then the two camel-like creatures, the extinct
Macrauchenia
and the modern guanaco, as temporally contiguous, must also be joined by blood. In other words, temporal sequence must record evolutionary transformation. Eureka! Darwin jotted the key insight into a private notebook: “The same kind of relation that common ostrich [rhea] bears to Petisse [the new species,
Rhea darwinii
], extinct guanaco to recent; in former case position, in latter time.”

What a portentous moment; what a crux in the history of human thought! But Darwin based his brilliant analogy, and his astoundingly correct general conclusion, on a flub in particulars—an error that he could not have recognized, and that did not affect the validity of his larger insight. South America had been
an island continent for tens of millions of years before the Isthmus of Panama rose just a few million years ago (see chapter 20). Several independent orders of mammals had evolved on South America, though most perished following climatic changes and the influx of North American species after the isthmus formed. (The outstanding set of survivors, South America’s grand indigenous group, includes sloths,
armadillos, and anteaters of the unique order Edentata.)

One of these extinct and independent orders, the Litopterna, included creatures uncannily convergent upon unrelated mammals of other continents. (In the evolutionary phenomenon of convergence, distantly related groups evolve similar forms, as a result of independent adaptation to common environments. Ichthyosaurs are reptiles, and dolphins
are mammals—but they both look and work like fishes.) One group of litopterns evolved an amazing resemblance to horses, including a parallel loss of toes, culminating in
Thoatherium
, the one-toed litoptern! Another group, represented by
Macrauchenia
, converged upon the New World camels. Since true camels later crossed the Isthmus of Panama and colonized South America (to survive today as llamas,
guanacos, and alpacas), we can hardly blame Darwin for inferring a genealogical tie between modern guanacos and the remarkably similar but unrelated fossil
Macrauchenia.
Richard Owen, England’s greatest anatomist and Darwin’s friend in these early years (see chapter 6), had affirmed the link of
Macrauchenia
to modern guanacos.

Thus, at this most sublime moment in the history of biology, Darwin
grasped the truth of evolution by basing a brilliant analogy on a flat-out factual error. Theories rarely arise as patient inferences forced by accumulated facts. Theories are mental constructs potentiated by complex external prods (including, in idealized cases, a commanding push from empirical reality). But the prods often include dreams, quirks, and errors—just as we may obtain crucial bursts
of energy from foodstuffs or pharmaceuticals of no objective or enduring value. Great truth can emerge from small error. Evolution is thrilling, liberating, and correct. And
Macrauchenia
is a litoptern. The fossil record provides our best direct evidence for evolution at large scales. And European fossil horses are collateral relatives, not ancestors of modern
Equus.

As Mrs. Julia Ward Howe—the
enduringly famous member of the couple—wrote, we may obtain our inspiration from a full range of sources, both “in the beauty of the lilies,” and “where the grapes of wrath are stored” (to probe the entire botanical spectrum). Any inspiring light can indicate a path in the darkness of nature’s complexity. “I have read His righteous sentence by the dim and flaring lamps. His truth is marching on.”