How We Believe, 2nd Ed.

by Michael Shermer

  Most of these authors have criticisms of Gould’s theory, and some are valid. Fontana and Buss contend that plenty would be conserved if the tape were rerun again. Kauffman argues for necessitating laws of self-organization that defy contingency. Cohen and Stewart point out: “Nowhere in Wonderful Life does Gould give an adequate treatment of the possible existence of evolutionary mechanisms, convergences, universal constants, that might constrain the effects of contingency.” Kelly has actually run Gould’s thought experiment in a sandbox with contrary results: “First thing you notice as you repeat the experiment over and over again, as I have, is that the landscape formations are a very limited subset of all possible forms.” McRae concludes: “Gould’s argument for contingency ultimately returns to the notions of progress and predictability it set out to challenge.” And Dennett calls Gould “the boy who cried wolf,” a “failed revolutionary,” and a “refuter of Orthodox Darwinism.”

  THE MISMEASURE OF CONTINGENCY

  One of the surprising things about all of these criticisms is that they appear to have missed or misunderstood the meaning of contingency and what Gould believes is its relationship to necessitating laws of nature. The reason for these misunderstandings is twofold. The first is the problem of meaning—contingency does not mean random, chance, or accident. The second is the problem of emphasis—contingency does not exclude necessity. Identifying and solving these problems can not only show us what is right about Gould’s dangerous idea, but also helps us understand how to find meaning in a contingent universe.

  The Problem of Meaning

  Many of those who oppose the idea of a predominantly contingent universe have misread contingency for accidental or random. Jack Cohen and Ian Stewart, for example, have stated explicitly that, “The survivors, who produced us, did so by contingency, by sheerest accident;” “Gould [argues] that contingency—randomness—plays a major role in the results of evolution …”, and Gould “sees the evolution of humanity as being accidental, purely contingent.” Yet Gould states quite clearly in Wonderful Life:

  I am not speaking of randomness, but of the central principle of all history—contingency. A historical explanation does not rest on direct deductions from laws of nature, but on an unpredictable sequence of antecedent states, where any major change in any step of the sequence would have altered the final result. This final result is therefore dependent, or contingent, upon everything that came before—the unerasable and determining signature of history. [Emphasis added.]

  As Gould notes, contingency is an unpredictable sequence of antecedent states, not randomness, chanciness, or accident.

  Daniel Dennett likewise takes Gould to task in a chapter entitled “Tinker to Evers to Chance,” a play on words linking Gould’s love of baseball—the three names represent the most famous double-play combination in baseball history—to chance, which Dennett identifies with contingency. But contingency does not mean chance, nor does it mean random, despite Dennett’s conclusion: “The fact that the Burgess fauna were decimated in a mass extinction is in any case less important to Gould than another conclusion he wants to draw about their fate: their decimation, he claims, was random.” True, mass extinctions may seem random, as when an asteroid hits the Earth. But by contingency Gould means a conjuncture of preceding states that determine subsequent outcomes. Just as astronomers knew exactly when and where Comet Shoemaker-Levy 9 was going to strike Jupiter in July of 1995 (and nailed the timing and location precisely), astronomers from (say) Mars, observing Earth 65 million years ago could have calculated the collision with the Yucatán peninsula with pinpoint accuracy. But the effects of those impacts could not have been adequately computed (and in the case of the Jupiter hit were not), because of the number of contingencies involved.

  The eventual rise of Homo sapiens, is even more contingent with millions of antecedent states in our past. Each event in the sequence has a cause, and thus is determined, but the eventual outcome is unpredictable because of contingency, not randomness or chance. The Burgess extinction may have been determined, but the sequence of events leading up to it, and those following, all the way to humans, were contingent. On this point Dennett says he is confused about what Gould means by “we” when he says we would not be here again if we reran the tape:

  There is a sliding scale on which Gould neglects to locate his claim about rewinding the tape. If by “us” he meant something very particular—Steve Gould and Dan Dennett, let’s say—then we wouldn’t need the hypothesis of mass extinction to persuade us how lucky we are to be alive … . If, at the other extreme, by “us” Gould meant something very general, such as “air-breathing, land-inhabiting vertebrates,”he would probably be wrong.

  Dennett’s confusion seems, well, confusing. By “we” Gould means the species Homo sapiens, no more, no less, and he has stated so on numerous occasions, including in Wonderful Life: “Replay the tape a million times from a Burgess beginning, and I doubt that anything like Homo sapiens would ever evolve again.”

  One might claim that these misunderstandings are caused by the fact that Gould has not offered a formal definition of contingency. That is true, so one must read him broad and deep. But it is there in dozens of examples and several informal definitions. In his essay “The Panda’s Thumb,” Gould shows that the thumb—actually the radial sesamoid bone of the panda’s wrist—is not a predictable design of nature’s necessitating laws of form, but an improvised contraption constructed from the history of what came before. In “The Panda’s Thumb of Technology,” Gould argues that the evolution of the QWERTY typewriter keyboard (denoting the first six letters from the left on the top letter row) supports his theory of contingency: “To understand the survival (and domination to this day) of drastically suboptimal QWERTY, we must recognize two other commonplaces of history, as applicable to life in geological time as to technology over decades—contingency and incumbency.” He then defines contingency as “the chancy result of a long string of unpredictable antecedents, rather than as a necessary outcome of nature’s laws. Such contingent events often depend crucially upon choices from a distant past that seemed tiny and trivial at the time. Minor perturbations early in the game can nudge a process into a new pathway, with cascading consequences that produce an outcome vastly different from any alternative.”

  This process is sometimes called path dependency, where systems get slotted into channels, and the QWERTY example is illuminating. Regular users of computers are locked by history into the QWERTY keyboard, designed for nineteenth-century typewriters whose key striking mechanisms were too slow for human finger speed. Even though more than 70 percent of English words can be produced with the letters DHIATENSOR, a quick glance at the keyboard will show that most of these letters are not in a strong striking position (home row struck by the strong first two fingers of each hand). All the vowels in QWERTY, in fact, are removed from the strongest striking positions, leaving only 32 percent of the typing on the home row. Only about 100 words can be typed exclusively on the home row, while the weaker left hand is required to type over 3,000 different words alone not using the right hand at all. Another check of the keyboard reveals the alphabetic sequence (minus the vowels) DFGHJKL. It appears that the original key arrangement was just a straight alphabetical sequence, which made sense in early experiments before testing was done to determine a faster alignment. The vowels were removed to slow the typist down, to prevent key jamming. This problem was eventually remedied, but by then QWERTY was so entrenched in the system (through manuals, teaching techniques, and other social necessities) that it became virtually impossible to change. Unless the major typewriter and computer companies, along with typing schools, teachers and publishers of typewriter manuals, and a majority of typists all decide to change simultaneously, we are stuck with the QWERTY system indefinitely.

  Gould’s biological version of this process is what he calls the Panda Principle: “The complex and curious pathways of history guarantee that most organisms and ecosystems cannot be desi
gned optimally.” Extending this principle to technology we might call it the QWERTY Principle: “Historical events that come together in an unplanned way create inevitable historical outcomes.”

  The Problem of Emphasis

  In the philosophy of history journal Clio, Murdo William McRae writes: “In spite of all his dedication to contingency and its attendant questioning of progress and predictability, Gould equivocates often enough to cast doubt upon the depth of his revolutionary convictions … . At times he insists that altering any antecedent event, no matter how supposedly insignificant, diverts the course of history; at other times he suggests that such antecedents must be significant ones.” The reason for the apparent “equivocation” is that Gould knows contingency interacts with necessity, but in his writings he sometimes emphasizes the former over the latter to make a particular point. Again, Gould does not offer a formal definition of necessity, yet it is there in his writings. After he first defined what he meant by contingency, in 1987, he immediately noted that “incumbency also reinforces the stability of a pathway once the little quirks of early flexibility push a sequence into a firm channel. Stasis is the norm for complex systems; change, when it happens at all, is usually rapid and episodic.” And in Wonderful Life Gould asks and answers the question of emphasis:

  Am I really arguing that nothing about life’s history could be predicted, or might follow directly from general laws of nature? Of course not; the question that we face is one of scale, or level of focus. Life exhibits a structure obedient to physical principles. We do not live amidst a chaos of historical circumstance unaffected by anything accessible to the “scientific method” as traditionally conceived. I suspect that the origin of life on earth was virtually inevitable, given the chemical composition of early oceans and atmospheres, and the physical principles of self organizing systems.

  Daniel Dennett goes much farther, accusing Gould of attempting to refute the quintessential driving mechanism of evolution itself, natural selection: “Can it be that Gould thinks his thesis of radical contingency would refute the core Darwinian idea that evolution is an algorithmic process? That is my tentative conclusion.” It is hard to imagine how Dennett came up with this notion since it is not to be found in Gould’s writings. The problem, it would seem, stems from the fact that when one wants to emphasize a previously neglected facet of nature, it might appear that something is being displaced. I asked Gould about Dennett’s charge and he responded as follows:

  My argument in Wonderful Life is that there is a domain of law and a domain of contingency, and our struggle is to find the line between them. The reason why the domain of contingency is so vast, and much vaster than most people thought, is not because there isn’t a lawlike domain. It is because we are primarily interested in ourselves and we have posited various universal laws of nature. It is because … we want to see ourselves as results of lawlike predictability and sensible products of the universe in that sense.

  To distance his pure Darwinism from Gould’s contingently modified version, Dennett makes an intriguing distinction between two types of metaphorical building devices: skyhooks, or “miraculous lifters, unsupported and insupportable,” and cranes, “no less excellent as lifters, and they have the decided advantage of being real.” Skyhooks are for wishful-thinking whimps who can’t handle the cold, hard reality of natural selection’s crane: “A skyhook is a ‘mind-first’ force or power or process, an exception to the principle that all design, and apparent design, is ultimately the result of mindless, motiveless mechanicity. A crane, in contrast, is a subprocess or special feature of a design process that can be demonstrated to permit the local speeding up of the basic, slow process of natural selection, and that can be demonstrated to be itself the predictable (or retrospectively explicable) product of the basic process.” Dennett accuses Gould of trying to sneak in a skyhook while he and his brave brethren—the unalloyed Darwinians—face the crane maker with brutal honesty. In fact, Dennett spends no less than fifty typeset pages trying to convince his readers that Gould is a skyhooker. Me thinks the gentleman doth protest too much. In my opinion, Dennett, and some others who adhere to a strict Darwinian adaptationist program, may be trying to find in nature a nonexisting pattern that shows us—Homo sapiens—as the nearly inevitable result of evolution. Dennett’s crane of relentless natural selection is, for him, a skyhook—“a ‘mind-first’ force or power or process” that, run over and over, would produce us again and again. It is something akin to an evolutionary theology, a secular cosmogony that finds us as the pinnacle of progressive cerebral evolution.

  CONTINGENT-NECESSITY

  The issue of contingency and necessity remains one of the great issues of our time because it touches on such deeply meaningful issues as free will and determinism, fate and destiny, and our place in the cosmos and in history. No one captured this better than Karl Marx, who opened the second paragraph of The Eighteenth Brumaire with these now classic lines: “Men make their own history, but they do not make it just as they please; they do not make it under circumstances chosen by themselves, but under circumstances directly found, given and transmitted from the past. The tradition of all the dead generations weighs like a nightmare on the brain of the living.”

  For the next century historians sought out those transmitted circumstances in the form of historical “laws,” culminating in 1942 with the publication of Carl Hempel’s influential paper entitled “The Function of General Laws in History,” in which he concluded: “There is no difference between history and the natural sciences: both can give an account of their subject matter only in terms of general concepts, and history can ‘grasp the unique individuality’ of its objects of study no more and no less than can physics or chemistry.” Hempel was wrong about general laws, but right about history and the natural sciences; not, however, in the direction one might think. History is not governed by Hempel’s laws (which he describes as “universal conditional forms”), but neither are the physical and biological worlds to the extent we have been led to believe. Scientists are coming to realize that the Newtonian clockwork universe is filled with contingencies, catastrophes, and chaos, making precise predictions of all but the simplest physical systems virtually impossible. As noted, we could predict precisely when and where Comet Shoemaker-Levy 9 would hit Jupiter, but we could muster at best only a wild guess as to the effects of the impacts on the Jovian world. The guess was completely wrong. Why? Contingency.

  There is irony in Hempel’s quest for general laws in history. For decades historians chased scientists in quest of universal laws, but gave up and returned to narratives filled with capricious, contingent, and unpredictable elements that make up the past. Meanwhile, a handful of scientists, instead of chasing the elusive universal form, began to write the equivalent of scientific narratives of systems’ histories, integrating historical contingencies with nature’s necessities, as Gould observes: “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.” Indeed, contingency is not Gould’s idea at all. Twenty-five hundred years ago Aristotle explained “that which cannot be otherwise is necessarily as it is,” yet “an event might just as easily not happen as happen.” and this is contingency. In a sense, science has been one long struggle to tame the contingent beast by finding necessitating laws that govern nature. Contingency becomes dangerous in Gould’s hands because he is a scientist, demonstrating how even a subject as predictable and subservient to natural law as planets and their moons, when examined closely, reveal so much uniqueness and individuality that while “we anticipated greater regularity … the surfaces of planets and moons cannot be predicted from a few general rules. To understand planetary surfaces, we must learn the particular history of each body as an individual object—the story of its collisions and catastrophes, more than its steady accumulations; in other words, its unpredictable single jolts more than its dai
ly operations under nature’s laws.” Simply put, history matters.

  Historians and philosophers have been cognizant for millennia of this basic tension between what may not be at all and what cannot be otherwise, between the particular and the universal, between history and nature, between contingency and necessity. But such synonyms can only take us so far (and may lead to problems of meaning and emphasis). Precise definitions are needed to formulate a model of change. Thus in this analysis contingency will be taken to mean a conjuncture of events occurring without design, and necessity to mean constraining circumstances compelling a certain course of action. Contingencies are the sometimes small, apparently insignificant, and usually unexpected events of life—the kingdom hangs in the balance awaiting the horseshoe nail. Necessities are the large and powerful laws of nature and trends of history—once the kingdom has collapsed, the arrival of 100,000 horseshoe nails will not help a bit. Leaving either contingency or necessity out of the formula, however, is to ignore an important component in the development of historical sequences. The past is constructed by both components, and therefore it might be useful to combine the two into one term that expresses this interrelationship—contingent-necessity—taken to mean a conjuncture of events compelling a certain course of action by constraining prior conditions.