At this point a relatively straightforward factual story, blessed with a simple moral message as well, becomes considerably more complex, given the foibles and practices of the oddest primate of all. Haeckel’s drawings, despite their noted inaccuracies, entered into the most impenetrable and permanent of all quasi-scientific literatures: standard student textbooks of biology. I do not know how the transfer occurred in this particular case, but the general (and highly troubling) principles can be easily identified. Authors of textbooks cannot be experts in all subdisciplines of their subject. They should be more careful, and they should rely more on primary literature and the testimony of expert colleagues. But shortcuts tempt us all, particularly in the midst of elaborate projects under tight deadlines.
A famous chart of the embryological development of eight different vertebrates, as drawn by Haeckel for a 1903 edition of his popular book, The Evolution of Man. Haeckel exaggerated the similarities of the earliest stages (top row) with gill slits and tails.
Therefore, textbook authors often follow two suboptimal routes that usually yield adequate results, but can also engender serious trouble: they copy from previous textbooks, and they borrow from the most widely available popular sources. No one ever surpassed Haeckel in fame and availability as a Darwinian spokesman who also held high professional credentials as a noted professor at the University of Jena. So textbook authors borrowed his famous drawings of embryonic development, probably quite unaware of their noted inaccuracies and outright falsifications—or (to be honest about dirty laundry too often kept hidden) perhaps well enough aware, but then rationalized with the ever-tempting and ever-dangerous argument, “Oh well, it’s close enough to reality for student consumption, and it does illustrate a general truth with permissible idealization.” (I am a generous realist on most matters of human foibles. But I confess to raging fundamentalism on this issue. The smallest compromise in dumbing down by inaccuracy destroys integrity and places an author upon a slippery slope of no return.)
Once ensconced in textbooks, misinformation becomes cocooned and effectively permanent because, as stated above, textbooks copy from previous texts. (I have written two previous essays on this lamentable practice—one on the amusingly perennial description of the “dawn horse” eophippus as “fox terrier” in size, even though most authors, including yours truly, have no idea about the dimensions or appearance of this breed; and the other on the persistent claim that elongating giraffe necks provide our best illustration of Darwinian natural selection versus Lamarckian use and disuse, when, in fact, no meaningful data exist on the evolution of this justly celebrated structure.)
We should therefore not be surprised that Haeckel’s drawings entered nineteenth-century textbooks. But we do, I think, have the right to be both astonished and ashamed by the century of mindless recycling that has led to the persistence of these drawings in a large number, if not a majority, of modern textbooks! Michael Richardson of the St. George Hospital Medical School in London, a colleague who deserves nothing but praise for directing attention to this old issue, wrote to me (letter of August 16, 1999):
If so many historians knew all about the old controversy [over Haeckel’s falsified drawings] then why did they not communicate this information to the numerous contemporary authors who use the Haeckel drawings in their books? I know of at least fifty recent biology texts which use the drawings uncritically. I think this is the most important question to come out of the whole story.
The recent flap over this more-than-twice-told tale—an almost comical manifestation of the famous dictum that those unfamiliar with history (or simply careless in reporting) must be condemned to repeat the past—began with an excellent technical paper by Richardson and six other colleagues in 1997 (“There is no highly conserved embryonic stage in the vertebrates: implications for current theories of evolution and development,” Anatomy and Embryology, volume 196: 91–106; following a 1995 article by Richardson alone in Developmental Biology, volume 172: 412–21). In these articles, Richardson and colleagues discussed the original Haeckel drawings, briefly noted the contemporary recognition of their inaccuracies, properly criticized their persistent appearance in modern textbooks, and then presented evidence (discussed below) of the differences in early vertebrate embryos that Haeckel’s tactics had covered up, and that later biologists had therefore forgotten. Richardson invoked this historical tale in order to make an important point, also mentioned below, about exciting modern work in the genetics of development.
From this excellent and accurate beginning, the reassertion of Haeckel’s old skullduggery soon spiraled into an abyss of careless reporting and self-serving utility. The news report in Science magazine by Elizabeth Pennisi (September 5, 1997) told the story well, under an accurate headline (“Haeckel’s embryos: fraud rediscovered”) and a textual acknowledgment of “Haeckel’s work, first found to be flawed more than a century ago.” But the shorter squib in Britain’s New Scientist (September 6, 1997) began the downward spiral by implying that Richardson had discovered Haeckel’s misdeed for the first time.
As so often happens, this ersatz version, so eminently more newsworthy than the truth, opened the floodgates to the following sensationalist (and nonsensical) account: a primary pillar of Darwinism, and of evolution in general, has been revealed as fraudulent after more than a century of continuous and unchallenged centrality in biological theory. If evolution rests upon such flimsy support, perhaps we should question the entire enterprise and give creationists, who have always flubbed their day in court, their day in the classroom.
Michael Behe, a Lehigh University biologist who has tried to resuscitate the most ancient and tired canard in the creationist arsenal (Paley’s “argument from design,” based on the supposed “irreducible complexity” of intricate biological structures, a claim well refuted by Darwin himself in his famous discussion of transitional forms in the evolution of complex eyes), reached the nadir in an op-ed piece for the New York Times (August 13, 1999), commenting on the Kansas School Board’s decision to make instruction in evolution optional within the state’s science curriculum—an antediluvian move, fortunately reversed in February 2001, following the political successes of scientists, activists, and folks of goodwill (and judgment) in general, in voting fundamentalists off the state school board and replacing them with elected members committed to principles of good education and respect for nature’s factuality. (In fairness, I liked Behe’s general argument in this piece, for he stayed away from irrelevant religious issues and attacked the Kansas decision by saying that he would never get a chance to present his supposed refutations if students didn’t study evolution at all.)
As his putatively strongest refutation of Darwinism, Behe cites the ersatz version of Richardson’s work on Haeckel’s drawings. (Behe presents only two other arguments, one that he accepted as true [the evolution of antibiotic resistance by several bacterial strains], the second judged as “unsupported by current evidence” [the “classic” case of industrial melanism in moths], with only this third point—the tale of Haeckel’s drawings—declared “downright false.” So if this piece represents Behe’s best shot, I doubt that creationists will receive much of a boost from their latest academic poster boy.) Behe writes:
The story of the embryos is an object lesson in seeing what you want to see. Sketches of vertebrate embryos were first made in the last 19th century by Ernst Haeckel, an admirer of Darwin. In the intervening years, apparently nobody verified the accuracy of Haeckel’s drawings . . . If supposedly identical embryos were once touted as strong evidence for evolution, does the recent demonstration of variation in embryos now count as evidence against evolution?
From this acme of media hype and public confusion, we should step back and reassert the two crucial points that accurately site Haeckel’s drawings as a poignant and fascinating historical tale and a cautionary warning about scientific carelessness (particularly in the canonical and indefensible practices of textbook writing)—but not, in any
way, as an argument against evolution, or a sign of weakness in Darwinian theory. Moreover, as a testament to greatness of intellect and love of science, whatever the ultimate validity of an underlying world-view honorably supported by men of such stature, we may look to the work of von Baer and Agassiz, Darwin’s most valiant opponents in his own day, for our best illustrations of these two clarifying points.
1. Haeckel’s forgeries as old news. Tales of scientific fraud excite the imagination for good reason. Getting away with this academic equivalent of murder for generations, and then being outed a century after your misdeeds, makes even better copy. Richardson reexamined Haeckel’s drawings for good reasons, and he never claimed that he had uncovered the fraud. But press commentary then invented and promulgated this phony version—and these particular chickens came home to a creationist roost (do pardon this rather mixed metaphor!).
Haeckel’s expert contemporaries recognized what he had done, and said so in print. For example, a famous 1894 article by Cambridge University zoologist Adam Sedgwick (“On the law of development commonly known as von Baer’s law”) included the following withering footnote of classical Victorian understatement:
I do not feel called upon to characterise the accuracy of the drawings of embryos of different classes of Vertebrata given by Haeckel in his popular works. . . . As a sample of their accuracy, I may refer the reader to the varied position of the auditory sac in the drawings of the younger embryos.
I must confess to a personal reason, emotional as well as intellectual, for long and special interest in this tidbit of history. Some twenty years ago I found, on the open stacks of our Museum’s library, Louis Agassiz’s personal copy of the first (1868) edition of Haeckel’s Natürliche Schöpfungsgeschichte (The Natural History of Creation). After his death, Agassiz’s library passed into the museum’s general collection, where indifferent librarianship (before the present generation) led to open access, through nonrecognition, for such priceless treasures.
I noted, with the thrill that circumstances vouchsafe to an active scholar only a few times in a full career, that Agassiz had penciled copious marginal notes—some forty pages’ worth in typed transcription—into this copy. But I couldn’t read his scribblings. Agassiz, a typical Swiss polyglot, annotated books in the language of their composition. Moreover, when he wrote marginalia into a German book published in Roman type, he composed the notes in Roman script (which I can read and translate). But when he read a German book printed in old, but easily decipherable, Fraktur type (as in Haeckel’s 1868 edition), he wrote his annotations in the corresponding (and now extinct) Sütterlin script (which I cannot read at all). Fortuna, the Roman goddess, then smiled upon me, for my secretary, Agnes Pilot, had been educated in Germany just before the Second World War—and she, Gott sei Dank, could still read this archaic script. So she transliterated Agassiz’s squiggles into readable German in Roman type, and I could finally sense Agassiz’s deep anger and distress.
In 1868, Agassiz at age sixty-one, and physically broken by an arduous expedition to Brazil, felt old, feeble, and bypassed—especially in the light of his continued opposition to evolution. (His own graduate students had all “rebelled” and embraced the new Darwinian model.) He particularly disliked Haeckel for his crass materialism, his scientifically irrelevant and vicious swipes at religion, and his haughty dismissal of earlier work (which he often shamelessly “borrowed” without attribution). And yet, in reading through Agassiz’s extensive marginalia, I sensed something noble about the quality of his opposition, however ill-founded in the light of later knowledge.
To be sure, Agassiz waxes bitter at Haeckel’s excesses, as in the accompanying figure of his final note appended to the closing flourish of Haeckel’s book, including the author’s gratuitous attack on conventional religion as “the dark beliefs and secrets of a priestly class.” Agassiz writes sardonically: Gegeben im Jahre I der neuen Weltordnung. E. Haeckel. (Given in year one of the new world order. E. Haeckel.) But Agassiz generally sticks to the high road, despite ample provocation, by marshaling the facts of his greatest disciplinary expertise (in geology, paleontology, and zoology) to refute Haeckel’s frequent exaggerations and rhetorical inconsistencies. Agassiz may have been exhausted and discouraged, but he could still put up a whale of a fight, even if only in private. (See my previous 1979 publication for details: “Agassiz’s later, private thoughts on evolution: his marginalia in Haeckel’s Natürliche Schöpfungsgeschichte (1868),” in Cecil Schneer, ed., History of Geology [University of New Hampshire: The University Press of New England, 277–82].)
Agassiz proceeded in generally measured prose until he came to page 240, where he encountered Haeckel’s falsified drawings of vertebrate embryology—a subject of extensive personal research and writing on Agassiz’s part. He immediately recognized what Haeckel had done, and he exploded in fully justified rage. Above the nearly identical pictures of dog and human embryos, Agassiz wrote: Woher copiert? Gekünstelte Ähnlichkeit mit Ungenauigkeit verbunden, z.b. Coloboma, Nabel, etc. (Where were these copied from? [They include] artistically crafted similarities mixed with inaccuracies, for example, the eye slit, umbilicus, etc.)
At least these two drawings displayed some minor differences. But when Agassiz came to page 248, he noticed that Haeckel had simply copied the same exact figure three times in supposedly illustrating a still earlier embryonic stage of a dog (left), a chicken (middle), and a tortoise (right). He wrote above this figure: Woher sind diese Figuren entnommen? Es gibt sowas in der ganzen Litteratur nicht. Diese Identität ist nicht wahr. (Where were these figures taken from? Nothing like this exists in the entire literature. This identity is not true.)
Finally, on the next page, he writes his angriest note next to Haeckel’s textual affirmation of this threefold identity. Haeckel stated: “If you take the young embryos of a dog, a chicken and a tortoise, you cannot discover a single difference among them.” And Agassiz sarcastically replied: Natürlich—da diese Figuren nicht nach der Natur gezeichnet, sondern eine von der andern copiert ist! Abscheulich. (Naturally—because these figures were not drawn from nature, but rather copied one from the other! Atrocious.)
2. Haeckel’s forgeries as irrelevant to the validity of evolution or Darwinian mechanisms. From the very beginning of this frenzied discussion two years ago, I have been thoroughly mystified as to what, beyond simple ignorance or self-serving design, could ever have inspired the creators of the sensationalized version to claim that Haeckel’s exposure challenges Darwinian theory, or even evolution itself. After all, Haeckel used these drawings to support his theory of recapitulation—the claim that embryos repeat successive adult stages of their ancestry. For reasons elaborated at excruciating length in my book Ontogeny and Phylogeny, Darwinian science conclusively disproved and abandoned this idea by 1910 or so, despite its persistence in popular culture. Obviously, neither evolution nor Darwinian theory needs the support of a doctrine so conclusively disconfirmed from within.
I do not deny, however, that the notion of greater embryonic similarity, followed by increasing differentiation toward the adult stages of related forms, has continued to play an important, but scarcely defining, role in biological theory—but through the later evolutionary version of another interpretation first proposed by von Baer in his 1828 treatise, one of the greatest works ever published in the history of science. In a pre-evolutionary context, von Baer argued that development, as a universal pattern, must proceed by a process of differentiation from the general to the specific. Therefore, the most general features of all vertebrates will arise first in embryology, followed by a successive appearance of ever more specific characters of particular groups.
Early embryonic stages of dog (left) and human (right), as drawn by Haeckel for an 1868 book, but clearly “fudged” in exaggerating and even making up some of the similarities. Louis Agassiz’s copy, with his angry words of commentary at top.
Agassiz’s angry comments (written above) upon Haecfyel’s false figure of simil
arities in the early embryonic stages of dog, chicken, and tortoise. Haeckel simply copied the same drawing three times.
Agassiz’s angry comments on Haeckel’s falsified drawings—ending with the judgment used as the title for this essay: “Abscheulich!”
In other words, you can first tell that an embryo will become a vertebrate rather than an arthropod, then a mammal rather than a fish, then a carnivore rather than a rodent, and finally good old Rover rather than Ms. Tabby. Under von Baer’s reading, a human embryo grows gill slits not because we evolved from an adult fish (Haeckel’s recapitulatory explanation), but because all vertebrates begin their embryological lives with gills. Fish, as “primitive” vertebrates, depart least from this basic condition in their later development, whereas mammals, as most “advanced,” lose their gills and grow lungs during their maximal embryological departure from the initial and most generalized vertebrate form.
Von Baer’s Law—as biologists soon christened this principle of differentiation—received an easy and obvious evolutionary interpretation from Darwin’s hand. The intricacies of early development, when so many complex organs differentiate and interconnect in so short a time, allow little leeway for substantial change, whereas later stages with fewer crucial connections to the central machinery of organic function permit greater latitude for evolutionary change. (In rough analogy, you can always paint your car a different color, but you had better not mess with basic features of the internal combustion engine as your future vehicle rolls down the early stages of the assembly line.)