Bully for Brontosaurus

by Stephen Jay Gould

  The explanation now offered for Io’s intense activity had been predicted just a few days before the photos arrived. Io is so close to giant Jupiter that the interplay between Jupiter’s gravitational tug and the reverse pull of the three other large satellites from the opposite side keeps the interior of Io fluid enough to resist rigidification.

  As this information arrived, I could only stand by in awe and reflect that Io had been misnamed. The four Galilean satellites honor some of Jove’s many lovers—an ecumenical assortment including his homosexual partner Ganymede; the nymph Callisto; and Europa, the mother of King Minos. Io, the fallen priestess, was changed to a heifer by jealous Hera, afflicted with a gadfly, and sent to roam Europe, where she forded (and indirectly named) the Bosporus (literally, the cow crossing) and finally emerged, human again, in Egypt. I thought that this innermost moon should be renamed Semele, to honor another lover who made the mistake of demanding that Jove appear to her in his true, rather than his disguised (and muted) human, form—and was immediately burned to a crisp!

  Is the size hypothesis therefore wrong because Io violated its prediction? The principle of surfaces and volumes, as a basic law of physics and the geometry of space, is surely correct. Io does not challenge its validity, but only its scope. The size hypothesis does not merely claim that the surface-to-volume principle operates—for this we can scarcely doubt. The hypothesis insists, rather, that the surface-to-volume principle so dominates all other potential forces that we need invoke nothing further to understand the history and topography of rocky planetary surfaces. Io does not refute the principle. But Io does prove dramatically that other circumstances—in this case proximity to opposing gravitational sources—can so override the surface-to-volume rule that its predictions fail or, in this case (even worse), are diametrically refuted.

  Planetary surfaces lie in the domain of complex historical sciences, where modes of explanation differ from the stereotypes of simple and well-controlled laboratory experiments. We are not trying to demonstrate the validity of physical laws. Rather, we must try to assess the relative importance of several complex and interacting forces. The validity of the surface-to-volume principle was never at issue, only its relative importance—and Io has challenged its domination.

  We must therefore know, in order to judge the size hypothesis, whether Io is a lone exception in a singular circumstance or a general reminder that the surface-to-volume principle ranks as only one among many competing influences—and therefore not as the determinant of planetary surfaces and their history. The test will not center upon arguments about the laws of physics, but upon observations of other bodies, for we must establish, empirically, the relative importance of a hypothesis that worked until Voyager photographed Io.

  Venus was the next candidate. Our sister planet, although closer to us than any other, had remained shrouded (literally) in mystery by its dense cover of clouds. But Russian and American probes have now mapped the Venusian surface with radio waves that can penetrate the clouds, as wavelengths in the visible spectrum cannot. Results are ambiguous and still under analysis, but proponents of the size hypothesis can scarcely react with unalloyed pleasure. Venus and Earth are just about the same size and Venus should, by the surface-to-volume hypothesis, be as active as our planet. Our sister world is, to be sure, no dead body. We have seen high mountains, giant rifts, and other signs of extensive tectonic activity. But Venus also seems to maintain too much old and cratered terrain for a body of its size, according to the principle of surfaces and volumes alone.

  Scientists have advanced many explanations for the difference between Venus and Earth. Perhaps tidal forces generated by the moon’s gravity keep Earth in its high state of geological flux. Venus has no satellite. Perhaps the high surface temperature of Venus, generated by a greenhouse effect under its dense cover of clouds, keeps the surface too pliable to form the thin and rigid plates that, in their constant motion, keep Earth’s surface so active.

  Voyager then moved toward Uranus and a final test. By this time, buffeted by Io and Venus, I was holding out little hope for the size hypothesis (and also wishing that the Rubáiyát had not spoken so truly about the moving finger, and that my publishers might deep-six all unsold copies of The Panda’s Thumb, with its reprint of my original 1977 essay). In fact, anticipating final defeat for that elegantly simple proposal of earlier years, I actually managed to turn disappointment into a modest professorial coup. I have long believed that examinations have little intellectual value, existing only to fulfill, and ever so imperfectly at that, any large institution’s need for assessment by number. Yet, for the first time, the moons of Uranus allowed me to ask an examination question with some intellectual interest and integrity.

  I realized that the final examination for my large undergraduate course had been set for the morning of January 24, at the very hour that Voyager would be relaying photographs of Uranian moons to earth. I therefore predicted that Miranda, although the smallest of five major moons, would be most active among them, and asked the students to justify (or reject) such a speculation—though the conjecture itself is absurd under the size hypothesis with its evident prediction that Miranda, as the smallest moon with the highest surface-to-volume ratio, should be cratered and devoid of internal activity. I asked the students:

  As you take this exam, Voyager 2 is sending back to Earth the first close-up pictures of Uranus and its moons…. On what basis might you predict that Miranda, although the smallest of these moons, is most likely to show some activity (volcanoes, for example) on its surface? We will probably know the answer before the exam ends!

  (When I first wrote the exam in early January, I couldn’t even provide my students with the moons’ diameters, for they had not yet been measured precisely, though we knew that Miranda was smallest. Between writing and administering the exam, Voyager measured the diameters, and we rushed to the printers with an insert. Science, at its best, moves very quickly indeed.)

  So I was ready for the final undoing of the size hypothesis, but not for the actual result of Miranda’s countenance. The conjecture of my question turned out to be quite wrong. I had been thinking of Io and the gravity of a nearby giant planet. Since Miranda is closest to Uranus, I supposed that it might be lit with modern volcanoes. But no volcanoes are belching forth sulfur, or anything else, on Miranda. The actual observations, however, spoke even more strongly against the size hypothesis and its prediction of a cold, cratered world.

  I had made one good prediction, probably for the wrong reason: Miranda is the most geologically active of Uranian moons, despite its small diameter of but 300 miles. (The moons of Uranus, outdoing even the mythic splendor of Jupiter’s satellites, bear lovely Shakespearean names—in order from Uranus out: Miranda, Ariel, Umbriel, Titania, and Oberon. In addition, Voyager has discovered at least ten additional and much smaller moons between Miranda and the planet’s surface.) The first photos of Miranda stunned and delighted the boys in Pasadena even more than her namesake had mesmerized Ferdinand on Prospero’s island. Laurence Soderblom, speaking for the Voyager imaging team, exclaimed: “It’s just mind boggling…. You name it, we have it…. Miranda is what you would get if you can imagine taking all the bizarre geological features in the solar system and putting them on one object.” A brave new world, indeed. So much for the size hypothesis and its uniform blanket of craters.

  Moreover, all the Uranian moons are surprisingly active (except for Umbriel, odd man out in more ways than one, as the only non-Shakespearean entry) in a gradient of increasing turmoil from the outermost king of Midsummer Night to the innermost daughter of the Tempest. “As you move closer to Uranus,” Soderblom added, “we see an increasing ferocity, as though these bodies have been tectonically shuffled in a cataclysmic fashion.”

  I must save the details for another time, but for starters, the surface of Miranda is a jumble of frozen geological activity—long valleys, series of parallel grooves, and blocks of sunken crust. Most prominent, and also most
notable for their lack of any clear counterpart on other worlds, are three structures that seem related in their formation. One has been dubbed a stack of pancakes, the second a chevron, and the third a racetrack. They are series of parallel grooves, or cracks, shaped to different forms according to their nicknames and full of evidence for massive slumping, rifting, and cliff making.

  In short, Io failed the size hypothesis by its position too close to Jupiter. Venus may not conform by a particular history that left it moon free and cloud covered. Miranda has failed, we know not why, by showing signs of a frantic past when the hypothesis predicted a passive compendium of impacts. The physical principle invoked by the size hypothesis—the law of surfaces and volumes—is surely correct, but not potent enough to overwhelm other influences and lead to confident predictions by itself. As we learn more and more about the historical complexity of the heavens, we recognize that where you are (Io) and what you have been (Venus and Miranda) exert as much influence over a planet’s surface as its size. After an initial success for our moon, Mercury, and Mars, the size hypothesis flunked all further tests.

  The story of a theory’s failure often strikes readers as sad and unsatisfying. Since science thrives on self-correction, we who practice this most challenging of human arts do not share such a feeling. We may be unhappy if a favored hypothesis loses or chagrined if theories that we proposed prove inadequate. But refutation almost always contains positive lessons that overwhelm disappointment, even when (as in this case) no new and comprehensive theory has yet filled the void. I chose this tale of failure for a particular reason, not only because Miranda excited me. I chose to confess my former errors because the replacement of a simple physical hypothesis with a recognition of history’s greater complexity teaches an important lesson with great unifying power.

  An unfortunate, but regrettably common, stereotype about science divides the profession into two domains of different status. We have, on the one hand, the “hard,” or physical, sciences that deal in numerical precision, prediction, and experimentation. On the other hand, “soft” sciences that treat the complex objects of history in all their richness must trade these virtues for “mere” description without firm numbers in a confusing world where, at best, we can hope to explain what we cannot predict. The history of life embodies all the messiness of this second, and undervalued, style of science.

  Voyager photograph of Miranda, showing fractured and reaggregated terrain. PHOTO COURTESY NASA/JPL.

  Throughout ten years of essays firmly rooted in this second style, I have tried to suggest by example that the sciences of history may be different from, but surely not worse than, the sciences of simpler physical objects. I have written about a hundred historical problems and their probable solutions, hoping to illustrate a methodology as powerful as any possessed by colleagues in other fields. I have tried to break down the barriers between these two styles of science by fostering mutual respect.

  The story of planetary surfaces illustrates another path to the same goal of lowering barriers. The two styles are not divisible by discipline into the hard sciences of physical systems and the soft sciences of biological objects. All good scientists must use and appreciate both styles since large and adequate theories usually need to forage for insights in both physics and history. If we accepted the rigid dichotomy of hard and soft, we might argue that as physical bodies, planets should yield to predictive theories of the hard sciences. The size hypothesis represented this mode of explanation (and I was beguiled by it before I understood history better)—a simple law of physics to regulate a large class of complex objects. But we have learned, in its failure, that planets are more like organisms than billiard balls. They are intricate and singular bodies. Their individuality matters, and size alone will not explain planetary surfaces. We must know their particularities, their early histories, their present locations. Planets are physical bodies that require historical explanations. They break the false barrier between two styles of science by forcing the presumed methods of one upon the supposed objects of the other.

  Finally, we should not lament that simple explanations have failed and that the “messy” uniqueness of each planet must be featured in any resolution. We might despair if the individuality of planets dashed all hope for general explanation. But the message of Io, Venus, and Miranda is not gridlock, but transcendence. We think that we understand Io, and we strive to fathom the moons of Uranus. Historical explanations are difficult, damned interesting, and eminently attainable by human cleverness. Whoever said that nature would be easy?

  Prospero, after saving his foes from the tempest, asserts that he cannot relate the history of his life too simply, for “’tis a chronicle of day by day, not a relation for a breakfast.” The tale is long and intricate, but fascinating and resolvable. We can also know the richness of history in science. Proper explanation may require a tapestry of detail. Our stories may recall the subtle skills of Scheherazade rather than the crisp epitome of a segment in Sixty Minutes, but then who has ever been bored by Sinbad the Sailor or Aladdin’s magic lamp?

  35 | The Horn of Triton

  THE ARGUMENTS of “iffy” history may range from the merely amusing to the horribly tragic. If Mickey Owen hadn’t dropped that third strike, the Dodgers might have won the 1941 World Series. If Adolf Hitler had been killed in the Beer Hall Putsch, the alliances that led to World War II might not have formed, and we might not have lost our war fleet at Pearl Harbor just two months after Owen’s miscue.

  I don’t think that we would be so fascinated by conjectures in this mode if we felt that anything could happen in history. Rather, we accept certain trends, certain predictabilities, even some near inevitabilities, particularly in war and technology, where numbers truly count. (I can’t imagine any scenario leading to the victory of Grenada over the United States in our recent one-day conflict; nor can I conjecture how the citizens of Pompeii, without benefit of motorized transport, could have escaped a cloud of poisonous gases streaming down Mount Vesuvius at some forty miles per hour.) I suspect, in fact, that our fascination with iffy history arises largely from our awe at the ability of individuals to perturb, even greatly to alter, a process that seems to be moving in a definite direction for reasons above and beyond the power of mere mortals to deflect.

  In opening The Eighteenth Brumaire of Louis Bonaparte, Karl Marx captured this essential property of history as a dynamic balance between the inexorability of forces and the power of individuals. He wrote, in one of the great one-liners of scholarship in the activist mode: “Men make their own history, but they do not make it just as they please.” (Marx’s title is, itself, a commentary on the unique and the repetitive in history. The original Napoleon staged his coup d’état against the Directory on November 9–10, 1799, then called the eighteenth day of Brumaire, Year VIII, by the revolutionary calendar adopted in 1793 and used until Napoleon crowned himself emperor and returned to the old forms. But Marx’s book traces the rise of Louis-Napoleon, nephew of the emperor, from the presidency of France following the revolution of 1848, through his own coup d’état of December 1851, to his crowning as Napoleon III. Marx seeks lessons from repetition, but continually stresses the individuality of each cycle, portraying the second in this case as a mockery of the first. His book begins with another great epigram, this time a two-liner, on the theme of repetition and individuality: “Hegel remarks somewhere that all facts and personages of great importance in world history occur, as it were, twice. He forgot to add the first time as tragedy, the second as farce.”)

  This essential tension between the influence of individuals and the power of predictable forces has been well appreciated by historians, but remains foreign to the thoughts and procedures of most scientists. We often define science (far too narrowly, I shall argue) as the study of nature’s laws and their consequences. Individual objects have no power to shape general patterns in such a system. Walter the Water Molecule cannot freeze a pond, while Sarah the Silica Tetrahedron does not per
turb the symmetry of quartz. Indeed, the very notion of Walter and Sarah only invites ridicule because laws of chemical behavior and crystal symmetry deny individuality to constituent units of larger structures. What else do we mean when we assert that hydrogen and oxygen make water or that silica tetrahedra sharing all their corner oxygen ions form quartz? (We could scarcely speak of a law if Ollie Oxygen willingly joined with Omar but refused to share with Oscar because they had a fight last Friday.) No actual quartz crystal has a perfect lattice of conjoined retrahedra; all include additions and disruptions known as impurities or imperfections—but the very names given to these ingredients of individuality demonstrate that scientific content supposedly lies in the regularities, while uniquenesses of particular crystals fall into the domain of hobbyists and aestheticians.

  (I don’t mean to paint the world of science as a heartless place of perfect predictability under immutable laws. We permit a great deal of play and doubt under the guise of randomness. But randomness is equally hostile to the idea of individuality. In fact, classically random systems represent the ultimate denial of individuality. Coin-flipping and dice-throwing models rest upon the premise that each toss or each roll manifests the same probabilities: no special circumstances of time or place, no greater chance of a head if the last five tosses have been tails, or if you blow on the coin and say your mantra, or if Aunt Mary will die as a consequence of your failure to score—in other words, no individuality of particular trials. Individuality and randomness are opposing, not complementary, concepts. They both oppose the idea of clockwork determinism, but they do so in entirely different ways.)