Belief in Search of Evidence

An occasional commenter on this blog, Dan, recently posted on his own blog, Rationality Now , a reflection inspired by a church sign. The sign said “Begin to weave and God will provide you with the thread”—which, as Dan pointed out, could be taken as a variant on the common saying (among Christians), “Just have faith in Jesus and He will show Himself to you.”

I should point out that this is only one possible reading of the sign. One might also take the intended message to be about projects that urgently need to be done, but for which you lack the resources. On this reading, the sign is meant to encourage you to begin such projects anyway, operating in the hope that once you are underway, the needed resources will begin to come in (provided by God).

And yet it seems plausible to claim that faith in Jesus, properly construed, is really a kind of life project. If so, then Dan’s interpretation of the church sign might not be so far off the mark. I suspect that most Christians will agree that “having faith in Jesus” is much more than just believing in a set of propositions. It’s a way of leading one’s life. (Agreement among Christians is likely to break down as soon as we ask what way of life is implied by faith in Jesus.)

But even if faith in Jesus is much more than belief in a set of propositions, the way of life implied by such faith will certainly presuppose a set of beliefs. To have faith is, in part, to live one’s life as if certain things are true. In the broadest terms, having faith in Jesus means living as if Jesus’ life and ministry express the ultimate reality, the divine, in some unique and profound way. And having faith in Jesus as savior means living as if Jesus has secured the redemption of the world; as if the evils that shatter human lives and infect human hearts are never the final word; as if somehow, because of Christ, even the most devastating horrors and malignancies have been stripped of the power to deprive our lives of meaning and value.

Living as if something is true—that is, exhibiting both outward behaviors and inner dispositions that make sense only given a set of presuppositions—is at least one definition of what it means to believe something. More specifically, this is the pragmatic sense of belief. When you live as if X is true, then you are giving pragmatic assent to X even if you withhold your intellectual assent. On an intellectual level you might say, “I have no idea if X is the case”—by which you probably mean that you have no more reason to think that X is true than you have to think otherwise, or something to that effect. But if you still live your life as if it is true, then you express pragmatic belief.

In his post, Dan pointed out that the common Christian saying, “Have faith in Jesus and He will show Himself to you,” amounts to the injunction to reach a conclusion before seeing any evidence for it—an injunction that many agnostics and atheists view with great suspicion, to say the least. After all, shouldn’t we form our beliefs on the basis of the evidence? Dan clearly thinks so, calling this kind of "belief-in-search-of-evidence" a species of "cart before the horse" thinking. He then goes on to criticize what he takes to be the essence of religious faith—which involves, as he sees it, not merely the act of embracing a conclusion before seeing any evidence (in the promise of evidence to come), but a broader indifference to evidence.

Now I agree that much of what goes by the name of “faith” in the lives of religious believers (but not all) is just as Dan describes it here: belief without any regard for evidence. And I also think that many who follow the dictum, “believe it first and evidence will follow,” do so improperly. Doing so would obviously be improper if one ought to withhold belief until the evidence comes in. And we can all agree that waiting for the evidence is quite frequently the proper way to proceed. But is that always the proper way to proceed? This is the question I want to explore here.

Put simply, the question is whether there are occasions when it’s legitimate to “believe first” (at least on some sense of “belief”) in order to see if the evidence follows. To assess this question, it may be helpful to consider some examples.

First, there is the matter of hunches. At least sometimes, a researcher or investigator has a strong sense that some hypothesis is true, and this sense inspires her to keep doggedly pursuing evidence in favor of the hypothesis despite setbacks that would have discouraged those less convinced (sending them off in pursuit of an alternative hypothesis).

Put in ordinary language, the researcher “has a hunch” and acts on it. To “have a hunch” means that you are willing to invest time and resources into investigating a possibility that, on the clear or explicit evidence available, seems no more worthy of investigation than countless other possibilities. You have “the hunch” that something is going on which, if it were, would leave certain traces. And so you don’t just proceed to look for these traces, but do so with an attitude of expectation. Perhaps, because of the strength of your hunch, you look longer than you might have otherwise. If you finally do uncover the traces that confirm the hunch, it may well be that the only reason you did so is because you kept looking long after others would have given up.

Were there explicit clues that suggested you should start your investigation with this hypothesis rather than some other one, you wouldn’t be acting on a hunch. You’d be acting on the initial evidence. To have a hunch means in part that, based on the explicit evidence, you could as readily have started with any number of other lines of investigation. But in the case of a hunch, your decision to start where you do isn’t arbitrary. It isn’t based on the mere fact that you have to start somewhere, and starting here is as good as anywhere. It is, instead, based on some level of conviction—some initial sense, not rooted in the explicitly available evidence, that this hypothesis is true. You choose to start here because there is a part of you that expects to find the traces you are looking for.

Put yet another way, the person with a hunch believes in the hypothesis more than do those who are simply going by the available evidence. There is at least a level of belief here that precedes the evidence and which may actually help to generate confirming evidence. At least on one level, the belief at issue here is pragmatic: the person with a hunch had different behaviors and dispositions than do those who lack the hunch. On the level of what they are inclined to do and expect and feel (e.g., surprise when the traces aren't there), they are like the person who thinks the hypothesis is true.

In short, there’s a real sense in which a hunch is a certain kind of belief-in-search-of-evidence. But I can’t imagine any researcher or investigator who would categorically condemn those of their colleagues who have hunches and act on them. In fact, I doubt that there are many researchers out there who don’t have occasional hunches, or who consistently resist them when they do.

That said, it is also true that, in scientific or forensic research, hunches can be dangerous. The decision to follow a hunch flirts with the kind of bias that undermines objective inquiry. A hunch becomes a bias, for example, when the researcher refuses to give up on it even once contrary evidence starts to pile up. It also becomes a bias when the researcher reads confirming evidence into ambiguous observations.

Let me clarify this last point. A body of clues often permits a variety of interpretations. If one interpretation supports one’s hunch while others don’t, it can be easy to accord preferential status to the supportive interpretation even though there are no other compelling reasons to do so. In short, one can ignore or fail to notice one’s own interpretive bias and so slide without realizing it from “The evidence is consistent with my hunch” to “The evidence favors my hunch over other hypotheses.” The police detective who falls prey to this danger when following a hunch may not only ignore a trail of evidence leading to the real killer, but may help put an innocent person behind bars.

But these risks do not constitute a decisive basis for ignoring hunches altogether and condemning those who don’t. If one is aware of the dangers, one can guard against them. And the fact is that acting on hunches sometimes bears important fruits.

Consider a different example: the act of trusting someone. Often, the only way you can find out whether or not someone is trustworthy is to actually trust them. That is, you begin by acting as if they are trustworthy. You invest them with responsibility and don’t hover over them to see whether they will fulfill it. You put yourself in their hands. You adopt behaviors and dispositions premised on the trustworthiness of the person, and in this sense adopt a belief in the pragmatic sense. When it comes to trustworthiness, it is often the case that only once you do this—only once you adopt a pragmatic belief—does the evidence for or against it becomes available.

Of course, here things get tricky, because often the decision to trust inspires someone to be more trustworthy than they might otherwise have been. Trusting can often operate as a self-fulfilling prophecy: give a dog a good name to live up to, and the dog is more likely to do so.

In my prison work with the Alternatives to Violence Project, we routinely introduce an array of trust exercises—trust circles and trust lifts—that call upon some participants of the workshop to voluntarily put their trust in others. Part of the point of these exercises is to give participants the chance to take the risk of trusting to see what happens (to believe in the trustworthiness of others as a way of discovering evidence of the same). But part of the point, I think, is also to inspire inner transformations in those who are given the experience of being trusted. The act of trust attributes to them a good character trait that they want to live up to. In doing so, they begin to own this character trait in a way that, perhaps, they hadn’t before. By being trusted, they become more trustworthy.

But this is an interesting point. Sometimes what we pragmatically believe shapes the reality around us—a fact that can operate both for good and for ill. In short, there may be circumstances in which it behooves us to believe the best hypothesis, even when there is no evidence to think it likely, precisely because expecting the best often makes the best more likely.

Finally, consider the case of a convicted murderer who seems to most everyone to be a kind of moral monster, a psychopath incapable of remorse or compassion towards his victims. But one person—a nun, say—is convinced that there is something good in every human being. Armed with this conviction, she seeks to reach that something in him. She really believes that it is there, and that a real human connection with this man, a mutual acknowledgment of shared humanity, is therefore possible.

There is, we can suppose, nothing in the man’s behavior that gives the nun any indication that he really possesses the capacity for such a connection with others. Everyone else has dismissed him as a psychopath lacking the kinds of human emotions and dispositions that make love, empathy, and compassion possible. His behavior is entirely consistent with such a dismissal of him. But this nun is convinced, in spite of this evidence, that somewhere buried beneath all the crud is a spark of human goodness. And so she reaches for it—persistently, sincerely.

Here’s what I think about such a case. If that “spark” is there, it’s only someone who is sincerely convinced that it’s there—someone like the nun—who’s likely to find it. Someone who doesn’t believe this will be neither persistent enough nor sincere enough to get through the crud. The vulnerable humanity of this criminal, if it’s there, is only accessible to those who are already convinced that it’s there, and who therefore have the behavioral dispositions and feelings necessary to reach it. Evidence for the existence of this spark will come only to those who believe in it before any evidence is available.

So here’s the question: Is the nun’s conviction inappropriate? Should we condemn her for it? Should we insist that belief must follow evidence, and thus conclude that she ought to reject the claim that this man has a tender human core she can reach, at least until evidence comes in—thereby guaranteeing that the evidence will never come in? Or should we, rather, declare along with William James that “a rule of thinking which would absolutely prevent me from acknowledging certain kinds of truth if those kinds of truth were really there, would be an irrational rule”?

The point is this: The decision to follow a hunch, or to trust a stranger, or to have faith in the potential of a human being who shows no evidence of potential, can be legitimate (even though it isn’t always so). But such decisions are cases of belief-in-search-of-evidence. That is, they involve pragmatic belief—acting as if something is true—before the evidence is in, while at the same time facilitating the acquisition of evidence.

And so I conclude that there are legitimate cases of believing in the hope that evidence will follow. The question, then, is whether there are theistic correlates for such cases—ways of having faith in God that exemplify legitimate belief-in-search-of evidence. Is there such a thing as a legitimate God-hunch, or a legitimate act of trusting in that which transcends the world of ordinary experience, or of seeking sincerely to establish a personal connection with the transcendent based on the conviction that it is there and has some the personal character that makes relationship possible? If so, what are the parameters within which such faith must fall in order to be legitimate?