Evidence in Science and Religion, Part Two

Stanley Fish

Stanley Fish on education, law and society.

In the post previous to this one, I revisited the question of the place of evidence in the discourses and practices of science and religion. I was prompted by a discussion on the the show “Up w/ Chris Hayes” (MSNBC, March 25) in which Steven Pinker and Richard Dawkins stated with great force and confidence that a key difference between science and religion is that the conclusions of the former are based on evidence that has emerged in the course of rigorous rational inquiry publicly conducted, while the conclusions of the latter are based on dogma, faith, unexamined authority, subjectivity and mere trust.

In response, Hayes observed that as laypersons, with respect to most areas of science we must take on trust what practitioners tell us. I took Hayes’s point further than he might be willing to take it, and suggested that because trust is common to both enterprises, the distinction between them, at least as it is asserted by Pinker and Dawkins, cannot be maintained.

Readers responded by pouring the proverbial ton of bricks on my head. The chief objection, repeated by many posters, was to the positing of an “equivalence” (a word that appeared often) between science and religion.

Michael K. declares that “the equivalence between the methodological premises of scientific inquiry and those of religious doctrine is simply false.” I agree, but I do not assert it. Neither do I assert that because there are no “impersonal standards and impartial procedures … all standards and procedures are equivalent” (E.). What I do assert is that with respect to a single demand — the demand that the methodological procedures of an enterprise be tethered to the world of fact in a manner unmediated by assumptions — science and religion are in the same condition of not being able to meet it (as are history, anthropology, political science, sociology, psychology and all the rest).

This means that all standards are equivalently mediated, not that all standards are equivalent in every respect. Apart from the shared characteristic of not being directly in contact with something called reality, science and religion are different in many, familiar ways, and by and large the differences correspond to the tasks we typically ask them to perform.

If you want to build a better mousetrap or computer, you will look to scientists and engineers. If you want to improve your marriage or learn how to win friends and influence people, you will look elsewhere, perhaps to couples counselors or to a religious tradition. If you want to figure out what a poem means, you consult and deploy the vocabulary and categories of literary criticism. And in each instance you will do this not because you have some metaphysical belief about the adequacy of a method to its independent object, but because, in your experience, the resources for solving this problem or addressing this issue are to be found over here and not over there.

This, I take it, is what many readers meant when they said, in a tone of triumph, that science works. Yes, it does, but so does literary criticism (it settles interpretive disputes, at least for a while) and so does therapy (it enhances the ability to socially interact, at least sometimes), and so does religious faith (it gives meaning and direction to life, at least for some people). The parenthetical qualifications in the previous sentence acknowledge that the certainty these practices give us is, at least from the perspective of the long run, provisional; it can be replaced or overturned or dislodged. But so can the certainties science gives us. Johnny E points out that not long ago “in geology everybody believed in geosynclines, there was lots of published data about them, but … now geosynclines don’t exist and everybody … believe[s] in seafloor spreading.” Now you see them, now you don’t.

Yes, the apostle of science will reply: that just shows that science is progressive and can correct its mistakes, while religion lacks a mechanism for detecting and purging error. This argument (made by many posters) assumes that when science “changes its mind,” it is because more precise and powerful techniques have given it a better purchase on the world it had previously perceived only dimly (“Now we see through a glass darkly”). The world has stayed still; only the devices of perception have changed and brought us closer to it.

But this Baconian model of scientific progress in which data sits waiting to be revealed by superior instruments is now, the Princeton philosopher Thomas Kelly tells us, “universally rejected by philosophers” (“Evidence,” in the Stanford Encyclopedia of Philosophy). “It is now appreciated,” Kelly continues, “that at any given time, which theories are accepted … typically plays a crucial role in guiding the subsequent search for evidence which bears on these theories.”

In the stronger versions of the shift Kelly reports, “evidence which bears on these theories” would be replaced by “evidence that becomes available in the light of those theories.” The strong thesis is that rather than being confirmed or disconfirmed by independent evidence, theories determine what will count as evidence. And therefore, as Kelly notes, if theory is prior to evidence, then the idea of an “appeal to evidence that could be appreciated by both sides” becomes problematical.

That is the idea many readers rely on when they characterize the scientific method as one that features independently available evidence as the neutral arbiter of disputes. But if evidence is never independent and is only evidence within the precincts of a particular theory, “adherents of rival theories,” Kelly explains, “will irremediably differ as to the appropriate description of the data itself,” and agreement between them cannot be brought about by simply pointing to the data.

Indeed, the phrase “data itself” — data independent of any theory currently in place — will be without a sense. Nor can any sense be given to the claim that because scientific conclusions must “stand up to scrutiny” (Jan), they enjoy a superior status. Scrutiny, like evidence, is something that occurs within a theory and will have a theory-specific shape. Scrutiny is not a practice that escapes or corrects the boundaries of perspective; it is a feature of a perspective it cannot transcend.

The slogan of what Carl Hempel called “the narrow inductivist account of scientific inquiry” is “seeing is believing”; first you look around and see what’s there, then you formulate general propositions of theory or belief. The slogan that comes along with the account of inquiry Kelly rehearses (but does not endorse), an account that has been around at least since Hobbes, is “believing is seeing.” The very act of looking around is always and already performed within a set of fully elaborate assumptions complete with categories, definitions and rules that tell you in advance what kinds of things might be “discovered” and what relationships of cause and effect, contiguity, sameness and difference, etc., might obtain between them. In Hebrews 11:1, St. Paul speaks of the “evidence of things not seen.” In the up-to-date accounts of scientific inquiry, the corollary would be “the evidence of things not directly seen,” but things that can be brought to (indirect and provisional) visibility by the assumption and application of powerful theories and the procedures they call into being.

This does not mean, as some readers assume, that the things brought to visibility by theory and belief are not real because their status as fact cannot be universally verified by independent empirical measures. If you withhold the adjective “real” from anything that is not independently apprehended, you will never bestow it. The better course — and the one we follow anyway without thinking about it — is to give the label “real” to whatever appears perspicuous to us within the evidential lenses we happen to be wearing.

Of course, the things we give the label “real” to are not all real in the same way and with the same persuasiveness to everyone. There are important differences between the arguments and experiments that are taken to support the reality of quarks (see Andrew Pickering, “Constructing Quarks”), and the arguments and statistics that are taken to support the reality of faith-healing or the power of prayer.

It is certainly possible to distinguish between these sets of arguments and to conclude that one is better supported than the other. And there will be many reasons that might lead you to that conclusion, including the fit or non-fit of the arguments with other arguments you already accept and the authority within a discipline or within the institutions you belong to of those who propose them. But one reason that cannot be given is that one set of arguments enjoys the advantage of being in touch with the world as it is apart from anyone’s beliefs, allegiances, assumptions and theories. No set of arguments can claim that advantage, although any set of arguments can claim a host of others.

For many readers, however, that reason — the one I say cannot be given — is the key one; and these readers believe that if that reason — science is in touch with fact and religion is not — is removed, the entire edifice of rationality and objectivity falls. So, for example, gw asserts that my epistemology “lay[s] waste to any or all objective truth claims” and adds that although I preach it I don’t believe it because (he predicts) “you will step outside your door tomorrow and trust quite non-skeptically that the law of gravity will still hold.”

I am not laying waste to objective truth claims; like everyone else I make them all the time and, when I am asked to, I defend them. But any defense I offer will not proceed by citing unmediated and unchallengeable evidence, but by citing evidence that appears to me to be conclusive given the features of the world as I see it and the force of arguments I unproblematically affirm, at least for now. In short, I rely on the world that has been delivered to me by the traditions of inquiry and demonstration I currently have faith in. That is what objectivity means, going with the best arguments and bodies of evidence one has at the moment. A more severe definition of objectivity that would require a measure of validity outside any tradition of inquiry or paradigm or episteme or habitus is simply uncashable.

As for my stepping out of my door and trusting in the law of gravity, one is quite independent of the other. I don’t step out of my door or refrain from stepping out of my 9th floor window because I believe in the law of gravity. What I do or refrain from doing doesn’t depend on my or anyone else’s having beliefs of that sort at all, unless you want to argue — and it would be a strained argument, to say the least — that everyone who walks down the street is doing so in the grip of some theoretical conviction. Rather, it depends on admonitions my mother expressed and a felt sense of my own limitations (I can’t jump 70 feet; heck, I can’t jump 3 inches) and the commonplace observations I have been making my entire life.

What is difficult for many to grasp is the irrelevance of theoretical speculation about faith and evidence (or anything else) to the conduct of everyday life. As ja, a scientist, put it, an argument like the argument that both science and religion rely on “assumptions about their own first principle … seems to be of purely academic interest. Most humans spend the vast majority of their time worrying about practical … issues like the availability of food and water, the suitability of climate, maintaining health, and combatting illness.” Exactly! Not only is the argument that science and religion cannot be distinguished on the basis of fidelity to reality true (a word I do not shrink from for the same reason I do not shrink from the word “objectivity”), it is also harmless.

Finally, I cannot forbear noting the picture of religion assumed by some of the most caustic commentators who say that religious experts “don’t engage in … debate” (chuckwagon), that when a religious truth is announced “no further inquiry is permitted” (Kevin Brady), that “religious dogma brooks no debate” (Prakash Nadkarni), that the only argument believers have is “The bible says so” (Kevin Grierson) and that “Faith requires a belief system by fiat” (drdave). It is hard to know what to say in the face of such pronouncements, except to recommend a course of reading to those who make them. They might begin with The Book of Job, Augustine’s Confessions and Bunyan’s “Pilgrim’s Progress.”