The Flamingo’s Smile

by Stephen Jay Gould

  La Grande Galerie is the granddaddy of this superseded style. Built in 1889, and unchanged since, its skeletons and stuffed animals occupy every available inch. The great central pyramid almost reaches the high glass ceiling. One side is all zebras, another all antelopes; six giraffes crown the summit. The dust is thick, the hall dark and empty; eerie silence marks a dingy majesty.

  The companion hall, La galerie d’anatomie comparée, is smaller, well lit, and still open. Its style is identical—row upon endless row, tier upon tier of blanched skeletons. I wandered up and down the aisles, marveling at a long row of walruses and five superposed tiers of monkey skulls. Then I passed by cabinet 106 and stopped short. It contains a sideshow to offset the neighboring forest of sleek lions, and to remind complacent Victorians that nature can be capricious and cruel, as well as bountiful. Cabinet 106 holds a collection of teratological specimens, skeletons of deformed and abnormal births. Most are human and most represent that puzzling and frightening phenomenon of joined birth, or “Siamese” twinning. Skeleton A8597 has two heads, three arms, and two legs; A8613 has four arms, two legs, and two heads projecting from the ends of a joined vertebral column; A8572 is almost normal, but a tiny, headless brother with arms and legs projects from his chest. All are small and clearly died at birth or soon thereafter.

  One skeleton stands out for its considerably larger size. A8599 is (or are)—and this is the issue we shall soon discuss—twin girls with two well-formed heads and upper bodies with four full arms. Two distinct vertebral columns nearly join at their base, and only two well-formed legs extend below. The label reads monstre humain dicéphale, or “two-headed human monster.” But A8599 was born live and survived several months. The twins were baptized and given names. The label records this poignant detail and includes, under the number and description, the simple identification “Ritta-Christina.”

  I mused much over Ritta and Christina, wondering about their life and death. Yet I would not have made the transition from troubled thought to essay had I not discovered, quite by accident, a dusty old tome in a bookstore two days later—volume 11, for 1833, of the Memoirs of the Royal Academy of Sciences. It contained a long monograph by the great medical anatomist Etienne Serres: Théorie des formations et des déformations organiques, appliquée à l’anatomie de Ritta Christina, et de la duplicité monstrueuse (“Theory of organic development and deformation, applied to the anatomy of Ritta Christina, and to duplicate monsters in general”).

  Anyone who does not grasp the close juxtaposition of the vulgar and the scholarly has either too refined or too compartmentalized a view of life. Abstract and visceral fascination are equally valid and not so far apart. Two days before, I had seen young schoolchildren standing before Ritta-Christina in open-mouthed awe or horror, soon masked by forced humor. Now I learned that France’s finest medical anatomist had dissected Ritta-Christina and used her to support a general theory of organic (not only human) embryology. Both themes seemed equally compelling to me; indeed, I had wallowed in both myself for two days. The children might not have generalized, but I have no doubt that M. Serres once gulped, as well as thought. I bought the book.

  Ritta and Christina were born on March 23, 1829, to poor parents in Sardinia. Times were hard and social mobility scarcely possible in ordinary circumstances. Parents today would receive pity and experience only sorrow; in 1829, realistic people, whatever their private feelings, must have recognized that such a child represented potential and substantial revenue, otherwise quite unobtainable.* Thus, the parents of Ritta-Christina scraped together some funds and brought her to Paris, hoping to display her at fancy prices. The Hottentot Venus had provoked enough protest fifteen years earlier (see essay 19); but she was whole, however exotic. Public sensibilities had limits, and the authorities forbade any open display of Ritta and Christina. But she was shown privately, many times too often—for she died, in part from overexposure, after five months of life.

  Ritta-Christina, drawn from life. FROM SERRES, 1833.

  The skeleton of Ritta-Christina after Serres’s preparation and still on display in Paris. FROM SERRES, 1833.

  I have consciously switched back and forth from singular to plural in describing Ritta-Christina. When the vulgar and scholarly meet, a common question often underlies our joint fascination. One question has always predominated in this case—individuality. Was Ritta-Christina one person or two? This issue inspired the feeble jokes of my terrified schoolchildren. It also motivated Serres’s scientific investigation. The same question underlay public fascination in 1829. When Ritta-Christina died, a Parisian newspaper wrote: “Already it is a matter of grave consideration with the spiritualists, whether they had two souls or one.”

  One or two? Through all scholarly excursions and sideshow huckstering, this single question has focused our fascination since Siamese twinning received its name. The originals, Eng and Chang, were born of Chinese parents in a small village near Bangkok in 1811 (Thailand was then called Siam). During the late 1820s and 1830s, they exhibited themselves in Europe and America and became quite wealthy. They decided to live in North Carolina where, at age 44, they married two sisters of English birth and settled down in two neighboring households to a comfortable life as successful (and, yes, even slaveholding) farmers. They switched houses at three-day intervals, traveling the one-and-a-half-mile distance by carriage. By the customs of the day, Chang was unquestioned master in his domicile, while Eng gave the orders chez lui. The unions were undeniably productive, for Chang had 10 children and Eng 12.

  Chang and Eng were physically complete human beings connected by a thin band of tissue, three and a quarter inches at its widest and only one and five-eighths inches at its thickest. Each had a full set of parts down to the last toenail. They carried on independent conversations with visitors and had distinct personalities. Chang was moody and melancholy and finally took to drink; Eng was quiet, contemplative, and more cheerful. Yet even they, history’s most independent Siamese twins, apparently harbored private doubts about their individuality. They signed all legal documents “Chang Eng” and often spoke about their ambiguous feelings of autonomy.

  Eng and Chang, the original Siamese twins. PHOTO COURTESY OF THE GRANGER COLLECTION.

  But what of Ritta and Christina, whose bodily independence did not extend below the navel? They seemed, at first glance, to be two people above and only one below. The old cultural criterion of head and brain might have suggested an easy resolution—two heads, two people. But as a scientist, Serres resisted this facile answer, for he had also studied Siamese twins with one head, two arms, and four legs. He reasoned that a uniformity of process must underlie both types of twinning and could not accept the simplistic resolution—one person if zipped halfway but starting from the top; two if zipped from the bottom.

  Serres struggled with this momentous issue for 300 pages and finally concluded that Ritta and Christina were two people. His arguments and basic style of science belong to another era in the history of biology. They are worth recounting if only because few intellectual exercises can be more rewarding than an examination of how radically different systems of thought treat a common subject of mutual interest. I also believe that Serres was at least half wrong.

  Serres represented the great early nineteenth-century tradition of romantic biology, called Naturphilosophie (“nature philosophy”) in Germany and transcendental morphology in his native France. If modern morphologists study form either to determine evolutionary relationships or to discover adaptive significance by examining function and behavior, Serres and his colleagues pursued markedly different goals. They were obsessed with the idea that some overarching, transcendental law must underlie and regulate all the apparent diversity of life.

  These laws, in the Platonic tradition, must exist before any organism arises to obey their regularities. Organisms are accidental incarnations of the moment; the simple, regulating laws reflect timeless principles of universal order. Biology, as its primary task, must search
for underlying patterns amidst the confusing diversity of life. In short, biologists must seek the “laws of form.”

  Serres contributed to the transcendental tradition by extending its concerns to embryology. Most of his colleagues had emphasized the static form of adults by searching for underlying patterns in final products alone. But organisms grow their own complexity from egg to adult. If laws of form regulate morphology, then we must discover principles for dynamic construction, not merely for relationships among finished creatures.

  A Siamese twin pair of conformation opposite to Ritta-Christina: one upper and two lower bodies. FROM SERRES, 1833.

  Serres’s monograph on Ritta-Christina begins with an abstruse 200-page dissertation on the principles of morphology and their application to embryology. Unless we sneak a peek at the alluring plates in the back (including the three figures reproduced with this essay), we hear nothing of the famous Sardinian twins until our senses are numbed by generality. This organization, in itself, reflects a style of science strikingly different from our own. We maintain an empirical perspective and like to argue that generalities arise from the careful study and collation of particulars. Any modern embryologist would discuss Ritta-Christina first and only venture some short and cautious conclusions at the end. But Serres, as a transcendentalist, believed that laws of form existed before the animals that obeyed them. If abstraction preceded actuality in nature, why not in human creativity as well? Thought and theory first, application later. (Neither extreme well represents the intricate interplay of fact and theory that regulates our actual practice of science. Still, I suspect that Serres’s “inverted” order is no worse a distortion of complex reality than our modern stylistic preferences.)

  In the first pages of his monograph, Serres tries to reduce the embryology of all animals to three basic laws of “organology.” First, by the law of eccentric development, known otherwise as the law of circumference to center, organs form initially at the edge of the developing embryo and then migrate toward the center. Second, by the law of symmetry, organs that become single and central in adults begin as double symmetrical rudiments on opposite edges of the developing embryo. Third, by the law of affinity, these symmetrical rudiments are drawn towards each other until they fuse in the center to form a single adult organ. (Let me be charitable and simply state that these laws are unwarranted extensions of patterns that operate occasionally in development. Serres was writing before the establishment of cell theory and just a few years after Karl Ernst von Baer’s discovery of the human ovum. His formal approach to morphology, so foreign to a world that can assess cellular and even molecular causes, fit the knowledge and mores of his own era.)

  Two hundred pages later, when Serres finally discusses his dissection of Ritta-Christina, we understand why he devoted so much preceding space to the three primary laws of organology—for they provide his solution to the great dilemma of individuality. Ritta and Christina are two people, albeit imperfect, and the laws of form proclaim their status.

  No one quarreled with the double verdict on Ritta and Christina from the waist up; the dilemma had always centered upon their well-formed, but clearly single, lower half—one anus, one vaginal opening, two legs. If she were two people all the way from stem to stern, how could her lower half form so well in the shape of one? How could the incomplete parts of two separate creatures fuse and blend into a form indistinguishable from the lower half of such unambiguous singletons as you or me?

  Serres used his laws of organology to render Ritta and Christina’s lower half as the conjoined product of two people. After all, the harmonious, well-formed single organs of ordinary individuals arise (by the law of symmetry) as separate and double parts at the embryo’s edge, and then move inward (by the law of circumference to center), eventually meeting and fusing (by the law of affinity) into one integral organ. If our single heart, stomach, and liver begin as two symmetrical rudiments (actually, they don’t, but Serres thought they did), then why should we view the presence of a single, well-formed organ in Ritta and Christina’s lower half as any argument against its construction from the mingled and melded parts of two embryonic individuals? If the twins have but one uterus, then the right half came from Ritta, the left from Christina. The two rudiments formed at the embryonic edges, in regions unambiguously assigned to Ritta or to Christina (law of symmetry). They moved toward the midline (circumference to center) and joined there (law of affinity) to form a single organ.

  Serres announced proudly that his laws of form had resolved the great dilemma in favor of duality: “How could we possibly have conceived that each child furnished half of an organ common to both, if the law of eccentric development had not taught us that single organs are, in their normal state, originally double.”

  Serres did not shrink from the decidedly peculiar logical implications of his solution. He noted that the large uterus had proper connections with the ovaries and vaginal canal and saw no reason why Ritta and Christina might not have borne children had they lived to maturity. (Serres also found a second, rudimentary uterus that would not have worked.) He concluded that the large uterus had formed half from Ritta and half from Christina, and admitted that any offspring developed within would have two natural mothers:

  This disposition of Ritta and Christina’s genital organs evidently shows…that while nature had taken measures to assure the lives of these children, she had not forgotten the possibility of their reproduction. Now, for this reproduction, nature had combined everything, so that all the pleasures and pains would be shared…. Supposing that conception occurred in the large uterus, a single child would have had two distinct mothers, a singular result of this associated life.

  Serres then discussed a pair of conjoined males with four legs and a single head, and opted for consistency and duality: the single well-formed brain shared the combined thoughts of two.

  There is a perfect unity produced by two distinct individualities. There are sense organs and cerebral hemispheres for a single individual, adapted to the service of two, since it is evident that there are two me’s in this single head [deux moi dans cette tête unique].

  Thus Serres made a valiant and consistent attempt to resolve a question that seemed hopelessly ambiguous. We may appreciate the effort and enjoy an excursion into the different view of biology that Serres maintained. But we must reject his conclusion.

  Fertilized human eggs usually develop into single individuals. Rarely, the dividing cells separate into discrete groups, and two embryos develop. These one-egg (or identical) twins are genetic carbon copies. In some ultimate, biological sense, they are the same iterated individual—and the psychological literature contains ample testimony to feelings of imperfect separation shared by many so-called identical twins. Yet, at least for definition’s sake, we experience no difficulty in identifying one-egg human twins as undeniably separate personalities for two excellent reasons: first, physical separation is the essence of our vernacular definition of individuality (see following essay); second, human personalities are so subtly and pervasively shaped by complex environments of life (whatever the quirky similarities between one-egg twins reared apart) that each person follows a unique path.

  With vastly greater rarity, the dividing cells of a fertilized egg begin to separate into two groups, but do not complete the process—and conjoined (or Siamese) twins develop. Conjoined twins span the entire conceivable range from a single individual bearing a few rudimentary parts of an imperfect twin, to superficially joined, complete individuals like Chang and Eng. Ritta and Christina fall squarely in the middle of this continuum. With our modern knowledge of their biological formation, I fear that we must reject Serres’s solution, and admit instead that his dilemma cannot be answered.

  We inhabit a complex world. Some boundaries are sharp and permit clean and definite distinctions. But nature also includes continua that cannot be neatly parceled into two piles of unambiguous yeses and noes. Biologists have rejected, as fatally flawed in principle, all att
empts by anti-abortionists to define an unambiguous “beginning of life,” because we know so well that the sequence from ovulation or spermatogenesis to birth is an unbreakable continuum—and surely no one will define masturbation as murder. Our congressmen may create a legal fiction for statutory effect but they may not seek support from biology. Ritta and Christina lay in the middle of another unbreakable continuum. They are in part two and in part one. And this, I am sorry to say, is the biological nonanswer to the question of the centuries.

  If this argument leaves you with an empty feeling after so much verbiage, I can only reply with the paradoxical phrase that is, so often, the most liberating response to an old mystery: The question has no answer because you asked the wrong question. The old question of individuality in Siamese twinning rests upon the assumption that objects can be pigeonholed into discrete categories. If we recognize that our world is full of irreducible continua, we will not be troubled by the intermediate status of Ritta and Christina.

  Dante punished schismatics by dismembering them in hell to exact a physical punishment worthy of their ideological crime: “Lo, how is Mohammed mangled…. Whom here thou seest, while they lived, did sow scandal and schism, and therefore thus are rent.”

  Let us value connections. As Dante analogized physical with ideological separation, perhaps we can learn from the indissoluble union of Ritta and Christina that our intellectual world revels in continuity as well.