Replication and Quality Control
Although unique observations (those reported in a single paper) are common in some areas of science, particularly whole-organism biology like evolution and ecology, in most fields, including chemistry, molecular biology, and physics, results are constantly being replicated by other observers. In those areas results become “true” only when they’re repeated often enough to gain credibility. The discovery of the Higgs boson in 2012, for which Peter Higgs and François Englert received a Nobel Prize the next year, was deemed prize-worthy because it was confirmed by two completely independent teams of researchers, each using rigorous statistical analysis.
A sufficiently novel or startling result will immediately inspire doubtful scientists to repeat it, often bent on disproving it. Other scientists, assuming your results are correct, might try to build on them to find new things, and part of that involves verifying your original results. The whole edifice of modern molecular genetics depends on the accuracy of the double-helix model of DNA, its process of replicating by unzipping and using each strand as a template to build another, and on the notion that the genetic code involved triplets of bases, each triplet coding for one unit (amino acid) of a protein. If any of this had been wrong, it would have been discovered very quickly as the field advanced. Likewise, each advance tested by proxy all the preceding ones.
Science has additional features that keep us from fooling ourselves by conscious or unconscious finagling with experiments or data. These include statistical analyses that tell us how likely our results might have been due to chance alone rather than to our new theory; blind testing, in which the researcher is prevented from knowing what material she’s testing (“double-blind studies,” in which neither researcher nor patient knows the identity of the treatment being given, are the gold standard for drug testing); and data sharing, which requires scientists to provide their raw data to anyone who asks, ensuring that those who want to can search for anomalies and run their own statistical tests.
Parsimony
Scientific theories invoke no more factors than necessary to adequately explain any phenomenon. This, like everything in the toolkit of science, is not an a priori requirement of the scientific method, but simply a method developed over centuries of experience. In this case, ignoring things that seem irrelevant keeps us from distracting ourselves with false leads. If we can completely explain the presence of smallpox by infection with a virus, why even consider factors like whether the patient ate too much sugar, or, indeed, whether, as was once thought, he was being divinely punished for immorality?
One unparsimonious method is invoking gods. Our experience that supernatural hypotheses have never advanced our understanding of the cosmos has, as we’ll see later, led to the idea of philosophical naturalism: the notion that supernatural entities not only fail to help us understand nature, but don’t seem to exist at all.
Living with Uncertainty
One of the most common statements we hear in science is “I don’t know.” Scientific papers, even those that report fairly solid findings, are hedged with statements like “this suggests that . . . ,” or “if this finding is correct . . . ,” or “this result should be verified by further experiments.” Granted, scientists are people, and we’d like to know all the answers, but in the end it’s our ignorance that moves science forward. It’s no shame to admit it, for without the unknown, there would be no science, nothing to spark our curiosity. But that attitude assumes that there are some answers we might never know.
One of these is how life originated. We know it happened between 4.5 billion years ago, when the Earth was formed, and 3.5 billion years ago, when we already see the first bacterial fossils. And we’re virtually certain that all living creatures descended from one original life-form, for virtually all species share the same DNA code, something that would be a remarkable coincidence if the code arose several times independently. But because the first self-replicating organism was small and soft-bodied and thus could not fossilize (it was likely a molecule, perhaps one surrounded by a cell-like membrane), we don’t have a way of recovering it.
Now, we may be able to create life in the laboratory under conditions thought to prevail on the early Earth—I predict we’ll do this within fifty years—but that tells us only that it could happen, not how it did happen. Like historians lacking data on crucial events (was there a real Homer who wrote the Iliad and the Odyssey?), students of historical sciences like cosmology and evolutionary biology are often forced to live with uncertainty. (The uncertainty is not about everything, however: we know when both the universe and life on Earth began; we’re just not sure how.) Living with uncertainty is hard for many people, and is one of the reasons why people prefer religious truths that are presented as absolute. But many scientists (I am one) share the feelings of Richard Feynman, who expressed his comfort with ignorance in an interview with the BBC:
I can live with doubt, and uncertainty, and not knowing. I think it’s much more interesting to live not knowing than to have answers which might be wrong. I have approximate answers and possible beliefs and different degrees of certainty about different things. But I’m not absolutely sure of anything, and there are many things I don’t know anything about, such as whether it means anything to ask why we’re here, and what the question might mean. I might think about it a little bit; if I can’t figure it out, then I go on to something else. But I don’t have to know an answer. I don’t feel frightened by not knowing things, by being lost in the mysterious universe without having any purpose, which is the way it really is, as far as I can tell—possibly. It doesn’t frighten me.
Feynman went a bit far in claiming that he wasn’t absolutely sure of anything, for he surely knew that he’d die one day (sadly, that day came too soon), and that he’d fall if he stepped off his roof. But his statement does encapsulate the doubt that is endemic—and necessary—in science. It’s not only endemic: it’s one of science’s attractions. A scientist lacking a big, juicy unsolved problem is a scientist bereft. H. L. Mencken compared the scientific investigator to “the dog sniffing tremendously at an infinite series of rat-holes,” and that was meant as a compliment. Our living with doubt contrasts strongly with the way many people regard their religion. True, some believers wrestle with doubt and uncertainty, but it’s a mind-set that’s neither encouraged, common, nor comfortable. Clerics and coreligionists usually urge the doubter to wrestle with those uncertainties and, in the end, resolve them. But with religion, there’s no real way to resolve them, for there’s no procedure for checking whether your doubts are justified. You’re then faced with either returning to your original faith or becoming an unbeliever.
Collectivity
One of the best parts of science’s toolkit is its international character, or rather, its transcendence of nationality, for although there are scientists throughout the world, we all work by the same set of rules. The participants in the discovery of the Higgs boson, for instance, came from 110 countries, with 20 of those nations being official collaborators in the project. When I visit Turkey, Russia, Austria, or India, I can discuss my work with my colleagues without any cultural awkwardness or misunderstandings. Although scientists come in all faiths, including no faith at all, there is no Hindu science, no Muslim science, and no Jewish science. There is only science, combining brainpower from the whole world to produce one accepted body of knowledge. In contrast, there are thousands of religions, most differing profoundly in what they see as “true.”
Curiously, the earliest scientific test I know of is actually described in the Bible. If you look at the First Book of the Kings (18:21–40), you’ll find a controlled experiment designed to reveal which god is real—Baal or the Hebrew god Yahweh. The test, proposed by the prophet Elijah, involved two bullocks, each killed, dressed, and placed on a separate pyre. Worshippers of each god were then told to ask their deity to ignite the pyre. Importuning Baal had no effect, even when his acolytes cu
t themselves with knives and lancets. But the Hebrew god came through, for even when his pyre was drenched with water, it burst into flame. Score one for Yahweh, scientifically shown to be the true god. In this case the penalty for being wrong was severe: the worshippers of Baal were summarily slain. But this story also invalidates the common claim that God won’t be tested, for God willingly participated in this experiment.
Along with the methodology that I’ve described as “science” come the accoutrements of professional science: having grant support for one’s research (usually from peer-reviewed applications to the government), submitting papers that are refereed by your peers before publication, having a job that can be categorized as “a scientist,” and so on. But these are ancillary to the methods themselves, which are in fact used by many people who aren’t usually considered scientists. In fact, I see science, conceived broadly, as any endeavor that tries to find the truth about nature using the tools of reason, observation, and experiment. Archaeologists use science when they date and study ancient civilizations. Linguists use science when they reconstruct the historical relationships between languages. Historians use science when they try to discover how many people died in the Holocaust, or refute the claims of Holocaust deniers. Art historians use science when dating paintings or trying to discern whether one is a forgery. Economists and sociologists use science when they try to understand the causes of social phenomena, although “truths” in those areas can be elusive. Native peoples use science when figuring out which local plants are useful in illness. (The use of quinine to cure malaria, for instance, was derived from the Quechua of Peru, who made an early version of “tonic” by mixing sweetened water with the bitter ground bark of the cinchona tree.) Even biblical scholars use science when reconstructing how and when the Bible was written. Not all of these areas, of course, are entirely scientific: much of the writing about history, for instance, involves untestable speculation about what caused various events.
The methods of science aren’t even limited to academics. Car mechanics use science when working out a problem in your electrical system, for they make and test hypotheses about where the defect lies. Plumbers use science, and their knowledge of hydraulics, when finding the source of leaks. The kinship between “professional” science and plumbing was engagingly described by Stephen Jay Gould. In 1981, Gould was in Little Rock, Arkansas, testifying in the famous trial of McLean v. Arkansas, during which a federal judge adjudicated (and eventually rejected) a state law requiring “balanced treatment” of evolution and creationism in public schools. On that visit Gould encountered a plumber:
As I prepared to leave Little Rock last December, I went to my hotel room to gather my belongings and found a man sitting backward on my commode, pulling it apart with a plumber’s wrench. He explained to me that a leak in the room below had caused part of the ceiling to collapse and he was seeking the source of the water. My commode, located just above, was the obvious candidate, but his hypothesis had failed, for my equipment was working perfectly. The plumber then proceeded to give me a fascinating disquisition on how a professional traces the pathways of water through hotel pipes and walls. The account was perfectly logical and mechanistic: it can come only from here, here, or there, flow this way or that way, and end up there, there, or here. I then asked him what he thought of the trial across the street, and he confessed his staunch creationism, including his firm belief in the miracle of Noah’s flood.
As a professional, this man never doubted that water has a physical source and a mechanically constrained path of motion—and that he could use the principles of his trade to identify causes. It would be a poor (and unemployed) plumber indeed who suspected that the laws of engineering had been suspended whenever a puddle and cracked plaster bewildered him. Why should we approach the physical history of our earth any differently?
This anecdote shows not only the continuity of scientific methods (and “ways of knowing”) across disparate areas, but also the disparity between science and religion embodied in a plumber who believed in Noah’s flood.
What Is Religion?
Defining “religion” is a thankless task, for no single definition will satisfy everyone. Belief in a god would seem mandatory, but some groups that look like religions, such as Jainism, Taoism, Confucianism, and Unitarian Universalism, don’t even have that. Other “religions,” like Tibetan Buddhism, may not worship gods, but do accept supernatural phenomena like karma and reincarnation.
Rather than argue semantics, I’ll choose a definition that fits most people’s intuitive conceptions of religion, and certainly corresponds to the tenets of the three Abrahamic faiths—Judaism, Christianity, and Islam—that comprise about 54 percent of the world’s inhabitants. This is also the form of religion that most often conflicts with science. The definition is taken from the Oxford English Dictionary:
Religion. Action or conduct indicating belief in, obedience to, and reverence for a god, gods, or similar superhuman power; the performance of religious rites or observances.
One can derive three characteristics of religion from this definition, all part of the Abrahamic faiths. The first is theism: the claim that God interacts with the world. The notion of “superhuman power” implies that God’s power is exercised, and the ideas of obedience and reverence, as well as performance of rites, imply that God must not only observe you but judge you, and his approval implicitly carries rewards or punishment. This means that I am considering religion as largely theistic, rather than a deistic belief in a remote, noninteractive God. As we’ll see, few religionists are strict deists anyway. But even deism, though denying God’s influence in the world, conflicts with science by making claims about God’s existence, and often about his creation of the universe.
The second feature of religion is its embrace of a moral system. If the supernatural agent confers or denies approval based on obedience, that means there are behaviors and thoughts that are either worthy or unworthy of that approval, including obedience itself. This yields a framework of divinely based morality. Even faiths like Taoism and Jainism that, lacking gods, could be considered philosophies still have moral codes. (Jains, for instance, devoutly abjure harming any creatures, including insects, and even try to avoid injuring plants!)
Codes of morality imply the third trait of religion: the idea that God interacts directly with you in a personal relationship. In The Varieties of Religious Experience, William James saw the ideas of a moral code and a personal connection to God as the nucleus of all religions:
[T]here is a certain uniform deliverance in which religions all appear to meet. It consists of two parts [an uneasiness and its solution]:
1. The uneasiness, reduced to its simplest terms, is a sense that there is something wrong about us as we naturally stand.
2. The solution is a sense that we are saved from the wrongness by making proper connection with the higher powers.
Finally, what do we mean by a “supernatural agent”? As we’ll see, the term “supernatural” is slippery, for even supernatural powers can affect natural processes, bringing the supernatural into the realm of empirical study. I’ll rely again on the Oxford English Dictionary’s definition of the adjective: “Belonging to a realm or system that transcends nature, as that of divine, magical, or ghostly beings; attributed to or thought to reveal some force beyond scientific understanding or the laws of nature; occult, paranormal.” Here “beyond scientific understanding” means “outside the realm of the material world.” As a supernatural being, God is often seen as a “bodiless mind,” but one with humanlike emotions.
From now on I’ll concentrate on religions that make empirical claims about the existence of a deity, the nature of that deity, and how it interacts with the world. But what do we mean by “claims”? Are they the claims of the church itself (that is, official doctrine and dogma), the claims of theologians (which, of course, differ, even among clerics within a faith), or the claims of regular believers,
which needn’t coincide with those of either theologians or church doctrine? We all know Catholics, for instance, who consider themselves members of the church although rejecting its doctrines on homosexuality and abortion, as well as the theory of evolution, which is accepted by the Vatican but rejected by many Catholics. When I discuss the claims of “religion,” I’ll simply go back and forth between theologians, believers, and dogma, trying to make clear which I’m discussing. Except for those rarefied theologians whose claims are either terminally obscure or close to atheism, it makes little difference, for believers, dogma, and theologians alike make existence claims and promote “ways of knowing” that make their faith incompatible with science. But do religions really make such claims? One needn’t look far to discover that most do, although more sophisticated believers and theologians tend to downplay that fact.
Does Religion Look for Truth?
It seems obvious that if religion is based on the existence of a god, then that is a contention about reality, and such a reality constitutes a basis for belief. In other words, the existence of God is taken as a fact. Surprisingly, some theologians come close to denying this, saying that God cannot be described, or is beyond all ken, thus rejecting any empirical claims about a deity save its existence. Religion, they say, has little or nothing to do with facts, but is about morals, building a community, or finding a way of life. Here are two examples of such denial from believers, the first from Francis Spufford, a Christian, and the second from Reza Aslan, a Muslim:
Faith Versus Fact : Why Science and Religion Are Incompatible (9780698195516) Page 7