Why People Believe Weird Things: Pseudoscience, Superstition, and Other Confusions of Our Time

by Michael Shermer

  Hobbes' introduction to scientific thinking came at the age of forty, when he happened upon a copy of Euclid's Elements at a friend's home and turned to a theorem he could not understand until he examined the preceding definitions and postulates. In one of those flashes of insight so important in the annals of science, Hobbes began to apply geometrical logic to social theory. Just as Euclid built a science of geometry, Hobbes would build a science of society, beginning with the first principle that the universe is composed of material matter in motion. His second principle was that all life depends on "vital motion," just as, in Hobbes' words, "the motion of the blood, perpetually circulating (as hath been shown from inany infallible signs and marks by Dr. Harvey, the first observer to it) in the veins and arteries" (1839-1845, vol. 4, p. 407). Through the senses, the brain detects the mechanical motion of objects in the environment. Since all simple ideas come from these basic sense movements, complex ideas must come from combinations of simple ideas. Thus, all thought is a type of motion in the brain called memories. As the motion fades, the memory fades.

  Humans are also in motion, driven by passions—appetites (pleasure) and aversions (pain)—to maintain the vital motion of life itself. To gain pleasure and avoid pain, one needs power. In the state of nature everyone is free to exert power over others in order to gain greater pleasure. This Hobbes calls the right of nature. Unequal passions among individuals living in nature lead to a state of "war of all against all." In the most famous passage in political theory, Hobbes imagines life without government and the state: "In such condition there is no place for industry because the fruit thereof is uncertain ... no account of time, no arts, no letters, no society, and which is worst of all, continual fear and danger of violent death and the life of man, solitary, poor, nasty, brutish, and short" ([1651] 1968, p. 76). Fortunately, Hobbes argues, humans have reason and can alter the right of nature in favor of the law of nature, out of which comes the social contract. The contract calls for individuals to surrender all rights (except self-defense) to the sovereign who, like the biblical Leviathan, is responsible only to God. Compared to a war of all against all, a sovereign presiding over the state is far superior and forms the basis for a rational society in which peace and prosperity are available on a mass scale.

  I have oversimplified the steps in Hobbes' complex theory, but the point is that his reasoning was Euclidean and his system mechanical. He began with metaphysical first principles and ended with an entire social structure. Moreover, because many political theorists consider Hobbes the most influential thinker of the modern age, the connection Hobbes made between politics and science is not dead yet. Science and culture are interactive, not separate and independent, despite attempts by scientists to keep them separate. One of the founders of modern science, Isaac Newton, in the third edition (1726) of his great work, the Principia, claimed, "Hitherto I have not been able to discover the cause of properties of gravity from phenomena, and I feign no hypothesis; and hypotheses, whether metaphysical or physical, whether of occult qualities or mechanical, have no place in experimental philosophy" ([1729] 1962, vol. 2, p. 547). Yet Olson has demonstrated just how often Newton did feign hypotheses, "such as the conjecture that light is globular and resembles tennis balls, which is clearly presented in the first optics paper" (1991, p. 98). Moreover, says Olson, even with regard to the law of gravity—Newton's greatest achievement—he feigned hypotheses: "It is undeniable that he did speculate about the cause of gravity—not only privately, but also in print. It has even been argued very convincingly that, so far as the study of experimental natural philosophy in the eighteenth century is concerned, Newton's conjectures and hypotheses ... were more important than the antihypothetical tradition of the Principia" (1991, p. 99). What could be more occult and metaphysical, in fact, than the "action at a distance" gravity produces. What is gravity? It is the tendency for objects to be attracted to one another. Why are objects attracted to one another? Because of gravity. In addition to being tautological, this explanation sounds rather ghostly, which brings us to the resolution of Pirsig's Paradox.

  Do ghosts exist? Do scientific laws exist? Is there no difference between ghosts and scientific laws? Of course there is, and most scientists believe in scientific laws but not ghosts. Why? Because a scientific law is a description of a regularly repeating action that is open to rejection or confirmation. A scientific law describes some action in nature that can be tested. The description is in the mind. The repeating action is in nature. The test confirms or rejects it as a law. The law of gravity, for example, describes the repeating attraction between objects, and it has been tested over and over against external reality, and thus it has been confirmed. Ghosts have never been successfully tested against external reality (I do not count blurry photographs with smudges on them that can be explained and replicated by lens distortions or light aberrations). The law of gravity can be considered factual, meaning that it has been confirmed to such an extent that it would be reasonable to offer temporary agreement. Ghosts can be considered nonfactual because they have never been confirmed to any extent. Finally, although the law of gravity did not exist before Newton, gravity did. Ghosts never exist apart from their description by believers. The difference between ghosts and scientific laws is significant and real. Pirsig's Paradox is resolved: all description is in the mind, but scientific laws describe repeating natural phenomena while pseudoscientfic claims are idiosyncratic.

  Pseudoscience and Pseudohistory

  Okay, so ghosts are bunk, along with most claims that fall under the heading of pseudoscience, by which I mean claims presented so that they appear scientific even though they lack supporting evidence and plausibility. The search for extraterrestrial life is not pseudoscience because it is plausible, even though the evidence for it thus far is nonexistent (the SETI—Search for

  Extraterrestrial Intelligence—program looks for extraterrestrial radio signals). Alien abduction claims, however, are pseudoscience. Not only is physical evidence lacking but it is highly implausible that aliens are beaming thousands of people into spaceships hovering above the Earth without anyone detecting the spacecrafts or reporting the people missing.

  But what about historical events? How do we know they happened since they do not repeat, either in nature or in the laboratory? As we shall see in chapters 13 and 14, there is a significant difference between history and pseudohistory. Most people would argue that history is not a science. Yet they would agree that Holocaust deniers and extreme Afrocentrists are doing something different from what historians are doing. What is that difference? In chapter 1, I emphasized that external validation through observation and testing is one of the key characteristics of science. We are told by believers in alien abductions that there is no way to test their claims because the experience was, in a way, a historical event, and we were not there to observe for ourselves. Further, the abduction experience itself is often a memory reconstructed through "regression hypnosis," which makes external validation even more difficult.

  Yet historical events can be tested. External validation is possible. For example, classicist Mary Lefkowitz has written a thoughtful reply to Afrocentric claims that Western civilization, philosophy, science, art, literature, and so on came out of Africa, not Greece and Rome. Her book, Not Out of Africa, raised storms across America, and she was accused of being everything from racist to politically incorrect. Lefkowitz wrote her book after attending a lecture given in February 1993 at Wellesley College (where she teaches) by Dr. Yosef A. A. ben-Jochannan, a noted extreme Afrocentrist. Among the outrageous statements made in the lecture was the claim that Aristotle stole the ideas that became the foundation of Western philosophy from the library of Alexandria, where Black Africans had deposited their philosophical works. During the question-and-answer session, Lefkowitz asked ben-Jochannan how this could be since the library was built after Aristotle was dead. The response was enlightening:

  Dr. ben-Jochannan was unable to answer the question, and said tha
t he resented the tone of the inquiry. Several students came up to me after the lecture and accused me of racism, suggesting that I had been brainwashed by white historians. . . .

  . . . As if that were not disturbing enough in itself, there was also the strange silence on the part of many of my faculty colleagues. Several of them were well aware that what Dr. ben-Jochannan was saying was factually wrong. One of them said later that she found the lecture so "hopeless" that she decided to say nothing. . . . When I went to the then dean of the college to explain that there was no factual evidence behind some Afrocentric claims about ancient history, she replied that each of us had a different but equally valid view of history. . . .

  . . . When I stated at a faculty meeting that Aristotle could not have stolen his philosophy from the library of Alexandria in Egypt, because that library had not been built until after his death, another colleague responded, "I don't care who stole what from whom." (1996, pp. 2, 3, 4)

  Therein lies the problem. Each of us may have a different view of history, but they are not all equally valid. Some are historical, and some are pseudohistorical, namely, without supporting evidence and plausibility and presented primarily for political or ideological purposes.

  A variety of sources independently attest to the life span of Aristotle (384-322 B.C.E.) and to the earliest date for the library of Alexandria (after 323 B.C.E.). It is a fact that Aristotle died before the library of Alexandria was built. One would have to posit a massive and widespread campaign of denial and fabrication to change this fact, which is exactly what extreme Afrocentrists do. True, humans are capable of almost anything and historical inferences have been wrong. Nonetheless, as Lefkowitz points out, "There is no reason why claims of conspiracy should be credited, if no real evidence can be produced to support it" (p. 8). Which brings us to another important point: pseudohistorians and historians do not treat their audiences equally and they use data differently. If Dr. ben-Jochannan wanted to argue that Aristotle was influenced by or acquainted with certain ideas circulating between Greece and Africa, he could examine the evidence for and against such a theory. Indeed, Lefkowitz does just that. But Dr. ben-Jochannan is not as interested in historical facts as he is in historical flavoring, not as interested in teaching the nuances of historiography as he is in instilling an Afrocentrist agenda. He takes a valid point about the influence of ideology on knowledge, stirs in the ignorance or apathy of an audience about historical events, adds a few historical facts and series of eccentric inferences about the past, and makes pseudohistory.

  The historical sciences are rooted in the rich array of data from the past that, while nonreplicable, are nevertheless valid as sources of information for piecing together specific events and confirming general hypotheses. The inability to actually observe past events or set up controlled experiments is no obstacle to a sound science of paleontology or geology, so why should it be for a sound science of human history? The key is the ability to test one's hypothesis. Based on data from the past the historian tentatively constructs a hypothesis, then checks it against "new" data uncovered from the historical source.

  Here is an example of this. I once had the opportunity to dig up a dinosaur with Jack Horner, curator of paleontology at the Museum of the Rockies in Bozeman, Montana. In Digging Dinosaurs, Homer reflected on the historical process in describing the two phases of the famous dig in which he exposed the first dinosaur eggs found in North America. The initial stage was "getting the fossils out of the ground; the second was to look at the fossils, study them, make hypotheses based on what we saw and try to prove or disprove them" (Horner and Gorman 1988, p. 168). The first phase of unsheathing the bones from the surrounding stone is backbreak-ing work. As you move from jack hammers and pickaxes to dental tools and small brushes, however, the historical interpretation accelerates as a function of the rate of bone unearthed, as does one's enthusiasm to keep digging. "Paleontology is not an experimental science; it's an historical science," Horner explained. "This means that paleontologists are seldom able to test their hypotheses by laboratory experiments, but they can still test them" (p. 168). How?

  In 1981 Horner discovered a site in Montana that contained approximately thirty million fossil fragments of Maiasaur bones, from which he concluded "at a conservative estimate, we had discovered the tomb of ten thousand dinosaurs" (p. 128). Horner and his team did not dig up thirty million fossil fragments. Rather, they extrapolated from selected exposed areas how many bones there were in the 1.25 by 0.25 mile bed. The hypothesizing began with a question: "What could such a deposit represent?" (p. 129). There was no evidence that predators had chewed the bones, yet many were broken in half, lengthwise. Further, the bones were all arranged from east to west—the long dimension of the bone deposit. Small bones had been separated from bigger bones, and there were no bones of baby Maiasaurs, just those of Maiasaurs between nine and twenty-three feet long. The find revealed more questions than answers. What would cause the bones to splinter lengthwise? Why would the small bones be separated from the big bones? Was this one giant herd, all killed at the same time, or was it a dying ground over many years?

  An early hypothesis that a mudflow buried the herd alive was rejected as "it didn't make sense that even the most powerful flow of mud could break bones lengthwise ... nor did it make sense that a herd of living animals buried in mud would end up with all their skeletons disarticulated." Applying the hypothetico-deductive method, Horner formulated a second hypothesis: "It seemed that there had to be a twofold event, the dinosaurs dying in one incident and the bones being swept away in another." Since there was a layer of volcanic ash a foot and a half above the bone bed, volcanic activity was implicated in the death of the herd. Deduction: because the fossil bones split only lengthwise, the damage to the bones came long after the event that caused death, which might have been a volcanic eruption, especially since volcanoes "were a dime a dozen in the Rockies back in the late Cretaceous." Conclusion: "A herd ofMaiasaura were killed by the gases, smoke and ash of a volcanic eruption. And if a huge eruption killed them all at once, then it might have also killed everything else around," including scavengers or predators. Then perhaps there was a flood, maybe from a breached lake, that carried the rotting bodies downstream, separated the big bones from the small bones (which are lighter), and gave them a uniform orientation. "Finally the ash, being light, would have risen to the top in this slurry, as it settled, just as the bones sank to the bottom." What about the baby Maiasaurs? "Perhaps the babies of that year were still in the egg or in nests when the volcano erupted, or perhaps nesting had not even begun." But what about babies from the previous season who would now be juveniles? Horner admits "that nobody knows for sure that these dinosaurs would have produced young each year" (pp. 129-133).

  Even in the first stage of a dig while fossils are being released from their rocky shroud, the hypothetico-deductive method is constantly applied. When I arrived at Horner's camp, I expected to find the busy director of a fully sponsored dig barking out orders to his staff. I was surprised to come upon a patient historical scientist sitting cross-legged before a cervical vertebra from a 140-million-year-old Apatosaurus and wondering just what to make of it. Soon a reporter from a local paper arrived (apparently a common occurrence as no one took notice) and inquired of Horner what this discovery meant for the history of dinosaurs. Did it change any of his theories? Where was the head? Was there more than one body at this site? And so on. Horner's answers were consistent with those of the cautious scientist: "I don't know yet." "Beats me." "We need more evidence." "We'll have to wait and see."

  This was historical science at its best. For example, after two long days of exposing nothing but solid rock and my own ineptness at seeing bone within stone, one of the preparators pointed out that the rock I was about to toss was a piece of bone that appeared to be part of a rib. if it was a rib, then the bone should retain its rib-like shape as more of the overburden was chipped away. This it did for about a foot, until it suddenly flared
to the right. Was it a rib or something else? Jack moved in to check. "It could be part of the pelvis," he suggested, if it was part of the pelvis, then it should also flare out to the left when more was uncovered. Sure enough, Jack's prediction was verified by further empirical evidence. And so it went day after day. The whole dig depends on such hypothetico-deductive reasoning. In a sense, historical science becomes experimental when predictions based on initial evidence are verified or rejected by later evidence. The digging up of history, whether bones or letters, is the experimental procedure of the historical scientist interested in putting a hypothesis to the test.

  I should note that there are differences between paleontological evidence and human historical evidence. The former is mostly first-order evidence—strictly physical, natural, and interpreted by extrapolating how natural laws apply now and in the past. The latter typically is second-order evidence—documents written by highly selective humans who add, delete, and alter the evidence. Historians have learned to treat historical evidence differently from archeological or paleontological evidence, to acknowledge that the gaps in historical evidence often have something to do with the fact that humans write about what interests them and what they think is important at the time. Nature does not delete the record of the socially marginalized. Still, as historian of science Frank Sulloway has shown in his controversial 1996 book, Born to Rebel, historical hypotheses can be tested (see chapter 16 for discussion of Sulloway's model). For the past hundred years, for example, historians have hypothesized that social class and social class conflict have been the driving forces behind revolutions, both political and scientific. Sulloway has tested this Marxian hypothesis by coding thousands of individuals in dozens of revolutions for their social class and then doing statistical analyses to see whether there really are significant differences in social class on opposing sides in revolutions. It turns out there is not. Marx was wrong, but it took a historian trained in the sciences to discover this fact by running a simple historical experiment.