It is often asserted by defenders of religious faith that science, especially within an atheist / materialist context, requires just as much faith as belief in God. This is often asserted without specifying either how one measures degrees of faith and without considering issues of actual truth. The fact that one assertion requires “just as much faith” as another proves the truth of neither one. Both may be wrong. But aside from that, perhaps the most serious omission is the failure to consider that the same word – in this case, “faith” – can connote very different meanings in different contexts. What “faith” means within the context of rationality is very different from what “faith” means within a religious context. This is actually a subject for a whole book, more likely, many books, than for a brief blog column. But for the sake of brevity, I will restrict my attention to two more tractable parameters that mean radically different things in their respective contexts: rationality (specifically, though not exclusively, science) and religion.
Let’s begin by acknowledging up front that, yes, there is a certain sense in which science – even believing in the credibility of science, and even more so practicing science – does indeed require a type of faith. For one thing and speaking very strictly, scientists do indeed accept on faith, first, the uniformity of physical law. How do we know, for example, that, say, the speed of light – 300,000 kilometers per second – is the same in, say, the Andromeda Galaxy as it is on the earth? That Newton’s gravitational constant (approximately 6.674 x 10-8 cm3/ gm / sec2) is the same? That water at sea level, i.e., the earth’s sea level and given the parameters of earth’s atmosphere, boils at 100 degrees C.? Speaking in the strictest possible sense, we don’t. No one has ever been to the Andromeda Galaxy and measured the speed of light or Newton’s constant or the boiling point of water empirically. Secondly, and in a far deeper sense – even in a sense far transcending the merely empirical – what assurance do we have that the language of mathematics is entirely adequate to describe and to predict natural phenomena? In fact, people who work in the field of the foundations of mathematics (1) cannot prove the unqualified consistency of mathematics itself and (2) can even prove that they cannot prove that consistency. (In fact, if you take a graduate-level course in the foundations of mathematics, and if you get your tuition-money’s worth, one of the milestones of the course will be the demonstration of the impossibility of proving the consistency of mathematics.) But scientists and mathematicians proceed on the basis of the assumption that physical law is uniform throughout the cosmos and that mathematics is likewise categorically consistent. In both cases, the uniformity of physical law and the consistency of mathematics, the following maxim is followed with, so to speak, religious rigor: “So Far, So Good”.
Furthermore, there is another reason, arguably as deep and as unverifiable as the consistency of mathematics, why, in a certain sense, the scientific enterprise proceeds on the basis of faith: the effectiveness of the scientific method itself. Even in principle, and least of all in practice, the effectiveness of the scientific method cannot itself be empirically demonstrated. How would you go about doing that? How would you go about formulating an experiment that would prove, even on the basis of “preponderance of evidence” and never mind “beyond reasonable doubt”, that the application of the scientific method results in “true truth” about the way the world works? What method would you use? If you answer “the scientific method”, then you have entered a vicious logical circle: you are using the very method whose validity you are seeking to assess. And even if you formulate an alternative methodology to the scientific method, the same question recurs: how would you go about assessing the effectiveness of that alternative? So accepting the effectiveness of the scientific method, like accepting the uniformity of physical law and the consistency of mathematics, is, at bottom, an act of faith on the part of the scientific community.
So what is it that distinguishes “scientific / rational faith” from religious faith? Do religious apologists perhaps have a valid point when they assert that belief in science and in rationality requires just as much faith as belief in God? My answer is “No”, and for a couple of interrelated – arguably synonymous – reasons.
First of all, like scientific data and scientific theories, even the deepest principles of science, in fact, of all fields of rational inquiry, are always held tentatively, as it were, lightly resting on the back of one’s hand like a goose-down feather waiting for a vagrant breeze to blow it off. At any given time, literally everything – data, theories, verification / falsification protocols and methodologies, etc. … everything – is always in principle up for grabs. I say “in principle” because, the more entrenched a theory is, of course, the greater the volume of data that substantiates it, the greater must be the volume of data required for disconfirmation. As the late Carl Sagan remarked in the same context, “Extraordinary claims require extraordinary evidence”.
An excellent recent example of this tentativeness of even the most venerable and exhaustively substantiated of scientific theories is what was at one point believed to be the possible observation of faster-than-light neutrinos at the European Center for Nuclear Research (CERN … the acronym is French) in Geneva, Switzerland. If confirmed by other investigators, faster-than-light neutrinos would have resulted in the collapse of one of the main supporting pillars of 20th- and 21st-century physics: Einstein’s special theory of relativity. In a few weeks, it became evident that the neutrinos appeared to exceed the speed of light because of improperly configured equipment, both at the emitting and detecting ends. Once the equipment was correctly configured, the appearance of superluminal velocities for neutrinos vanished. For my purposes, the most important take-away is this: scientists at CERN were willing to consider this possibility and investigate the data obtained from the CERN instruments, even if it meant that Albert Einstein and his cornerstone theory – along with most of contemporary physics founded thereon – were seriously wrong.
With religion, by contrast, there are certain teachings that are taken to be “irreformable”, therefore unchangeable. We usually think of the “irreformability” of religious teachings as characteristics of very conservative religious groups, most obviously fundamentalist sects. But all religions, even the most liberal, have such. Even with, say, liberal Anglicanism / Episcopalianism not everything is “up for grabs”. With regard to these teachings and principles, which vary from religious group to religious group, the maxim to be followed is most decidedly not “So Far, So Good”. The Bible, for example, is – in some sense, no matter how highly nuanced and qualified – seen as authoritative – in some sense, no matter how highly nuanced and qualified. Furthermore, if one relaxes the constraints on religious doctrine, and dials down “not up for grabs-ness”, to the point that literally everything is “up for grabs”, then one arrives at a religion that is in principle indistinguishable from science – or at least, from some rational area of inquiry. One ends up doing physics or chemistry or math or history, just wearing a mitre and carrying a crozier. The nature of all religious faith – properly so called – is that some things, however few, are seen as non-negotiable.
This is true because another difference between scientific / rational faith, on the one hand, and religious faith, on the other, is that, in the former, there is fundamental agreement on two critical points: (1) the relevance of evidence, and (2) what counts as evidence. In science and in other forms of rational inquiry, hypotheses live and die according to the evidence, be it evidence for or against. In fact, in at least one important way, evidence against is even more critical and determinative than evidence for. This is why it is often said that scientific theories must be falsifiable. Suppose we hypothesize "All crows are black". We may find thousands, millions, billions ... of instances of black crows. But the hypothesis is proven false by finding a single white -- actually, any non-black -- crow. So important is falsifiability that hypotheses that are not falsifiable are not even science. Poetry, maybe. Narrative fiction, maybe. But not science. In a very real sense, scientific theories cannot be proven, only falsified. Similar remarks apply to other fields of rational inquiry. Roughly a century ago, it was widely believed that the British romantic poet William Wordsworth was always celibate ... until it was discovered that, on his visit to France shortly after the Revolution, he became romantically involved with a woman, Annette Vallon. The hypothesis of the celibate Wordsworth was falsified with a single counterexample. Annette Vallon was William Wordsworth's white crow. This also tells us what counts as evidence in rational inquiry: data, empirical and verifiable facts. Consequently, rational faith consists in the willingness to assume a given hypothesis holds true until -- but only until -- the hypothesis is disconfirmed by a single counterexample. This is why I said earlier that scientific theories are held on the back of the hand like a goose-down feather awaiting a single breath of wind.
The place of evidence is one of the most critical ways in which rational faith (science, but not only science) differs from religious faith, both in terms of (1) and (2), above (the relevance of evidence and what counts as evidence, respectively). In fact, and tragically, wars have been fought and blood has been shed over both these issues, and the matter is still not settled. Does the Pope's pronouncement count as evidence in favor of the Perpetual Virginity of Mary? Does the Shroud of Turin count as evidence for the actual, space-time-historical Resurrection of Jesus Christ? You interpret the Bible one way; I, another. Furthermore, the two interpretations are logically and diametrically opposed. Both cannot be right. Both may be wrong. What would count as evidence either way?
One may well object that expecting scientific evidence -- or, anyway, evidence that would satisfy the demands of a rational intellect -- for religious / faith assertions is to commit a "category mistake", like asking for evidence that the square root of two is green. Rational assertions and "faith assertions", this argument continues, are two fundamentally different species of sentences, and, accordingly, require different types of evidence and different methodologies to substantiate them. This was, in fact, the late Prof. Stephen Jay Gould's position: science / rationality and religious faith represent what Gould called "non-overlapping magisteria" (NOMA -- the acronym was coined by Prof. Daniel Dennett of Tufts University and popularized by Prof. Richard Dawkins). This is, I believe, the most compelling account of the difference between rational faith and religious faith, and the categories of one cannot be used to critique the other. Again, whole libraries could be written on this issue, more space and time than I obviously have in a blog post.
But the NOMA response does does imply, as a corollary, the danger of founding law and public policy purely and exclusively on religious-faith assertions, e.g., homosexuality, abortion, the relationship between Church and State, etc. The nature of rational faith is such that it is founded upon, and open to revision by, evidence that is objectively real and evident to all inquirers. The speed of light is 300,000 km / sec for all observers, irrespective of location. Accordingly, laws and public policies that are binding on everyone in a free, constitutionally governed society should be based on considerations that cross all lines of race, socio-economic class, ethnicity,etc. (Considerations of common law and custom also enter in, not just "hard data". For example, one argument commonly advanced in favor of capital punishment is its universality within the Western legal tradition. But even here, the custom must be virtually universal, almost like the speed of light.) This is not an infringement on freedom, unless one wants to argue that one should be "free" to "choose" to go faster than the speed of light. However, religious faith, by the nature of NOMA, impinges on areas that are largely personal and historically conditioned, circumstances that, in such a society, one is free, from the standpoint of civil law, to abide by or not. In undertaking rational discourse, one is speaking and investigating in terms that are, in the highest sense, universally binding -- not in terms of "ought", but in terms of "is". In undertaking religious discourse, one is speaking and investigating in terms that are, by their very nature, binding only in terms of "ought", and even then only with respect to oneself and one's experience as part of a religious community.
James R. Cowles
(Postscript: In the interest of fairness, I should note for the record that Prof. Dawkins disagrees with Gould on the matter of NOMA. Dawkins believes that the existence of God is a scientific hypothesis that can be addressed scientifically, because there are not, as Gould argues, two "magisteria", only one. For Gould's position, see his 1997 essay in the magazine Natural History, "Non-Overlapping Magisteria" and his subsequent  book Rock of Ages. For Dawkins' position, see his book The God Delusion. I have problems with both. Gould's approach perhaps works for non-historical religions. But the three great monotheisms -- Christianity, Islam, and Judaism -- are emphatically not non-historical: for all three, God leaves Her / His fingerprints all over the universe and human history, and the moment you begin to talk about God being present in and acting in history, the wall between the two "magisteria" has been breached. Dawkins needs to specify a prior criterion by which we can determine whether or not a supposed artifact is of intelligent or natural origin. This is especially difficult because, as far as we know, the universe we inhabit only came into being one time. See Hume's critique of Paley's "Divine Watchmaker" argument.)