Can I Be Obliged to Believe?

Abstract

We build an argument directed to agnostics who think there’s a realistic possibility some specific revelatory claim is true (for instance, the Christian, or Judaic, or Islamic claim) and who find that claim more plausible than its theistic competitors. Though such agnostics may have serious reservations about the claim, perhaps not even deeming the chance it’s true to be at least fifty-fifty, we contend that—surprisingly—it’s obligatory for them to assent to the claim if it provides a means for remediation of wrong-doing. Our focus is the Christian revelatory claim, but the argument’s template can be applied to other religions that, like Christianity, promise to fix the world’s ills in an afterlife.

1. Overview of the Argument

Might there be an obligation to believe in Christianity—or some other religion that advertises it can fix the problems of this world in an afterlife, especially problems we create ourselves?

Yes. That is the answer we’ll defend in this paper. We’ll begin with the fundamental reality that all of us, all too often, act very badly towards our fellow human beings, sometimes with frightful outcomes, consequences we can’t contain or repair. From this fundamental fact an argument-template can be constructed:

(1)

Some of my grievous wrongs to others have generated debts that cannot be paid in this world.

(2)

I judge there’s a realistic possibility x is true [where x is a religion or revelatory claim]; x is more plausible and appealing to me than any of its theistic competitors; and I judge that x, if true, offers a path to recompense those I’ve wronged beyond present repair.

(3)

If (1) and (2), then I’m obligated to assent to x.

(4)

So, I’m obligated to assent to x.

We will focus on an instantiation of the argument-template that substitutes “Christianity” for x (those who find other religions more plausible and appealing can consider alternative instantiations):

(1)

Some of my grievous wrongs to others have generated debts that cannot be paid in this world.

(2)

I judge there’s a realistic possibility Christianity is true; Christianity is more plausible and appealing to me than any of its theistic competitors; and I judge that Christianity, if true, offers a path to recompense those I’ve wronged beyond present repair.

(3)

If (1) and (2), then I’m obligated to assent to Christianity.

(4)

So, I’m obligated to assent to Christianity.

The first-person form of the premises makes this argument dialectical, in Aristotle’s terminology—proceeding from a particular individual’s beliefs—rather than demonstrative. A demonstrative argument would proceed from the facts irrespective of what anybody concedes. We think a demonstrative form of our reasoning is defensible. Surely, the first premise above can be generalized: all of us have grievously wronged others, generating debts impossible to discharge in our world. And it’s not an individual judgment whether some specified religion or revelatory claim, as commonly understood, promises to wipe every tear away. Similarly, whether a revelatory claim, such as the Christian claim, has some plausibility, is not, in the end, an individual judgment: the credentials of revelatory claims can be evaluated objectively.

However, though we believe a demonstrative version of our argument is defensible, it’s beyond this paper’s scope to support all the generalized claims, so we’ll confine ourselves to the dialectical form of the argument. This means, of course, that the argument won’t work for everyone: atheists and some agnostics are likely to reject premise (2) forthwith. But many agnostics will not reject it outright. Many agnostics believe that theism, and specifically Christianity, is what we will term “realistically possible”. They might say: “I’m not at all sure Christianity is true, or that it’s more likely than not, or even that the chance it’s true is at least fifty-fifty; but I see some reason for believing it. Maybe it’s true, though maybe it’s false.” The argument we advance in this paper is aimed at such individuals. It’s a broad audience. And though we will be developing the argument with a focus on Christianity, the general pattern of argument is transferrable to other religions, which widens its receptibility.

Various lines of reasoning categorized as “practical moral arguments” or “pragmatic moral arguments” for God’s existence conclude that “believing God exists is rational”, or that a theistic commitment is permissible, despite a lack of super-abundant evidence.1 We argue for the much stronger claim that believing God exists is obligatory—for one who judges that there’s a realistic possibility Christianity is true (and that Christianity is more plausible than its theistic competitors). Our argument is intended to encourage fence-sitters, those who think their assent to Christianity would be rational but also that their non-assent is rational, to move off the fence, to act.

2. Defense of the First Two Premises

The first premise of our argument, recall, was:

(1)

Some of my grievous wrongs to others have generated debts that cannot be paid in this world.

Consider Zacchaeus, the tax collector who climbed into a sycamore tree to see Jesus over the crowds (Luke 19:1-10). He’s cheated the Jews. He admits it. Seeking to make amends, he vows to give half his goods to the poor and restore four-fold what he’s stolen. But one of Zacchaeus’s victims would have bought a house, imagine—and now can’t, due to the fraud, because houses are ten times more expensive. Another victim might have educated his son if he’d had the funds—but they’ve been stolen from him; and the son, lacking a good education, took a risky job as a stonecutter and was killed. There’s no making full recompense for wrongdoing because we can’t control the forward paths of our actions. That’s the case with respect to simple situations where the wrongdoer has cheated somebody of a specific amount of money—scenarios where we want to say, if recompense is ever possible, it’s possible here. And in more complex circumstances, it’s hard to know how even to begin the process of restitution. How does the woman caught in adultery make amends to those she’s wronged (John 7:53-8:11)? No amount of money can undo the harm to her family and the family of her adulterous partner. We cannot change the past; but that doesn’t mean we don’t own it. We own it and have incurred debts we cannot fully pay through natural means.

We carry those debts to the grave—and beyond. Norman Malcolm reports that Wittgenstein thought the idea of immortality could acquire meaning “through one’s feeling that one has duties from which one cannot be released, even by death.”2 Wittgenstein attempted partially to remedy some of his own transgressions and mis-steps through confessing them to friends and apologizing, where possible, to those he’d harmed—even searching out children he’d hurt decades earlier as a teacher, to try to atone. Such ameliorative efforts never suffice. This world does not provide means for true redress.

But a supernatural order might allow fully restorative justice. This brings us to the (instantiated) argument’s second premise:

(2)

I judge there’s a realistic possibility Christianity is true; Christianity is more plausible and appealing to me than any of its theistic competitors; and I judge that Christianity, if true, offers a path to recompense those I’ve wronged beyond present repair.

Each of the three conjuncts in this premise deserves comment. We begin with the last of the three, the judgment that Christianity—if true—offers a way of discharging debts that cannot be paid in this world.

Wittgenstein was not a Christian; nevertheless, he wrote these striking lines:

What inclines me to believe in Christ’s Resurrection? It is as though I play with the thought.—If he did not rise from the dead, then he is decomposed in the grave like any other man. He is dead and decomposed. In that case he is a teacher like any other and can no longer help; and once more we are orphaned and alone.

(Malcolm 1994, p. 9)

Evidently, even non-theists who would not go so far as to say there’s a “realistic possibility” that Christianity is true (we take it Wittgenstein wouldn’t have said this) feel the allure of another world—one that establishes right order, that provides for payment of debts, that connects us to deep community and fathomless aid.

It’s clear that if Christianity were true, it would offer a means of remediating wrongdoing, of discharging our debts. Christianity claims a providential God who draws great good from wrongdoing, good for the injured, and allows sinners to collaborate in the restorative process through prayer and sacrifice. Further, Christianity claims a savior who rescues us from orphanhood and solitude, who satisfies debts and unites us as children of God. If Christianity were true, there’d be a means of remediation.

Exactly what the means of remediation would be is not clear. The various Christian communities and their associated theological traditions don’t agree about all details of the story of salvation: they differ regarding the nature and import of original sin, for instance, and the structure of atonement, and the role of restorative justice in upbuilding the City of God. We find a similarly variegated array of takes on other major components of Christian teaching. Consider, for example, the doctrine of the incarnation. Where does the union of divine and human natures take place? In a person? In a nature? In a “supposit”?3 Or what? Theologians argue about the answers, but the inability to explain how God was incarnate doesn’t block belief that God was incarnate. And if there was an incarnation, then various consequences flow from that—such as the reliability of Christ’s message of salvation. Of course, if a central Christian doctrine is demonstrably incoherent, or inconsistent with well-established facts, or immoral, that will tell decisively against the doctrine and the religion itself. Some atheists and agnostics think that no Christian soteriology can pass muster. But many inquirers think that the broad outline of the Christian story of salvation and remediation is not implausible, though the details of the account are obscure.

There is—inescapably—an indefiniteness to objects of belief. Descartes thought that certitude was guaranteed by clear and distinct ideas. But clarity and distinctness can be inversely related to the confidence we have in propositions. You may see off in the distance in northern Minnesota a large animal coming towards you. You’re positive that it’s a large animal, but unsure whether it’s a wolf or a German shepherd. You can be confident that genes play a role in evolution without knowing exactly what a gene is, or what its function is. You can be certain that one plus one equals two, though you can’t decide between Platonist and anti-Platonist accounts of the number “one”. You can think there’s a realistic possibility that Christianity offers a means of remediation without being able to detail the operation. We often judge statements to be true or false, and assent to them, when they are less than crystal-clear and precise.

A wrong-doer might engage in or contribute to the process of remediation in various ways. Simply by making the act of belief a person might participate. Among the promises made to those who embrace the faith is the assurance that the harms they’re responsible for will be put to rights. Is it in virtue of your belief that those you’ve wronged are repaired, made whole? The nature of the rectification might, indeed, depend on you. Perhaps your participation in Christian community through faith amounts to a form of prayer, and if your prayer is efficacious, God supplies generous remedy for your transgressions. Prayer matters. A skeptic might say there’s no point in asking God to make things right for those you’ve injured, because a good God would do that anyway—would balance the scales of justice. But there is no injustice if God is abundantly good to your victim, but not as good as God would be if you didn’t make the act of belief.

While a degree of participation in remediation may come through making an act of belief or faith, a deeper level may be reached by a believer making an act of charity or love bound up with the suffering of the cross. (Indeed, the act of love might be seen as a deeper form of assent.) Christian teaching on this matter is profound and mysterious, and we don’t claim to understand it. But the tradition has held that believers may will to unite their sufferings with Christ’s and call upon the wounds of Christ to provide the remedy for sin. God ultimately rectifies, but it’s via an act of love on our part. “In my flesh I complete what is lacking in Christ’s afflictions for the sake of his body, that is, the Church”, the Apostle Paul writes (Col. 1:24). The pronouncement is startling to a Christian hearing it for the first time because it suggests Christ’s suffering was in some way inadequate. Theological reflections on Paul’s meaning may help us fathom the teaching.4 But however the process of remediation is supposed to work, Christian doctrine does teach that we can collaborate in restorative justice through assenting to Christianity.

What does it mean to judge there’s a realistic possibility that all this is true (or that any other Christian doctrine, or any doctrine of another religion, is true)? It doesn’t mean one assigns it some specific probability, such as a probability according to which it’s more likely than not. Experts don’t agree upon a form of probabilistic argument that fits every-day and scientific reasoning: frequentism, Bayesian reasoning, and likelihoodism are all seriously problematic outside tightly controlled set-ups (e.g., where one is predicting the color of a ball to be drawn from an urn containing known numbers of black and red balls, and balls of no other color). Rather than speaking of probability, we prefer to speak in terms of plausibility. To judge there’s a “realistic possibility” a statement is true is to judge at least that it’s not highly implausible. It certainly doesn’t require knowing that Christianity is true, or even believing that the case for Christianity is stronger than the case against—indeed, it’s not clear one can get either case on the table.

Though the notions of “realistic possibility” and “plausibility” can’t be spelled out in terms of the probability calculus, we can say a little more about their import through recourse to analogies. Here’s one. Suppose you believe that there are quarks, tiny little things that compose neutrons as well as other entities. That’s correct, you believe, or nearly correct (whatever “nearly” means). Now the history of science shows that many hypotheses get discarded after a time. And in the case of quarks, much is said about quantum field theory that puts the existence of quarks, conceived of as particles of some sort, in doubt. Still, you accept the proposition “There are quarks”, along with propositions you find strewn through books on particle physics explaining what quarks do. If you’re asked what your level of confidence is, you might say, mindful of the history of science: “I’m not certain physicists’ accounts of quarks are correct; but the notion that there are quarks has some plausibility—it’s a realistic possibility that there are quarks.”

In connection with our defense of premise (3) in the next section, we will develop another analogy, an extended analogy, that further illustrates what we mean by “realistic possibility”.

What would support a judgment that it’s realistically possible that Christianity (including doctrines concerning remediation) is true? That question is largely beyond the scope of this paper; in setting it aside we rely on our sense that a good number of agnostics do, in fact, make the judgment, and the argument of this paper should be of interest to them. Still, we can offer a few remarks about sorts of evidence for Christianity that may be particularly useful to an inquirer drawn towards premise (2), but hesitant to assert it.

Theistic arguments with nearly indubitable premises but relatively modest conclusions, conclusions less grand than “God exists”, can be highly valuable to persons reflecting on whether they can assent to premise (2). For instance, it can be argued that it’s not highly implausible there’s a little-g god, a cause of the world’s existence:

• It’s not implausible that the physical universe came to be. [That’s the consensus of contemporary scientific opinion.]
• It’s not implausible that whatever comes to be, has a cause—a non-logical necessary condition for its existence. [That’s been the considered view of virtually all philosophers at least in the West, including, perhaps surprisingly, Hume.]
• Given the above, it’s not highly implausible that the physical universe has a cause, a non-logical necessary condition for its existence. [Multiplying the probabilities of two probabilistic claims generates the probability of their conjunction, a probability that will be lower than the probability of either of the two original claims; a similar, though non-quantitative, move can reasonably be made regarding plausibility claims.]

There is a substantial gap between the conclusion of this argument and the conclusion theists ultimately are after. This argument doesn’t show that there’s only one cause of the universe and doesn’t show that the cause is a conscious agent. However, once it’s recognized that it’s not highly implausible there’s a non-logical necessary condition (a cause) for the universe, the door is open for an argument that this cause is a conscious agent, and one to which moral terms such as “good” and “wicked” can be applied (i.e., not an amoral agent).5 So, the short argument directly above provides a foot in the door—and for some inquirers, maybe a shoulder as well.

The bulleted argument (and other arguments for a little-g god) can lead to serious consideration of revelatory claims, claims that God has vouchsafed a revelation to humankind. And if a particular revelatory claim, such as the Judaic or Christian or Islamic claim, becomes a live option, the content of that claim may enable an inquirer simultaneously to judge that the revelatory claim, and the embedded proposition that God (a big-G God) exists, are more likely true than not, and assent to them. A fortiori, the content may support the weaker judgment that there’s a realistic possibility Christianity is true. Just how the content might support these judgments is a longer story than we have room here to tell.6 We mention the potential of the line of argument because in our experience it is powerful. But of course, it’s not the only possible means by which an agnostic inquirer might be persuaded that there’s a realistic possibility Christianity is true.

Premise (2) has an inquirer asserting that Christianity is more plausible and appealing to the inquirer than its theistic competitors. Many agnostic inquirers do, in fact, have this perspective.7 An inquirer raised or schooled in the West presumably will have some familiarity with religious traditions other than Christianity. But he or she may rationally judge it a mistake to spend much time trying to plumb the depths of other traditions or resolve apparent perplexities. For instance, although karmic traditions present a process for affecting one’s own karma and the karma of others (thus redressing one’s wrongdoing), the process involves a long series of reincarnations, which may seem excessively complicated. Serious study of the karmic accounts could conceivably address the concerns. But the inquirer may decide the final goal of the series of reincarnations (annihilation of self) isn’t, on its face, sufficiently attractive to undertake such labor. The enterprise, which would require immersion in the literature, perhaps learning relevant languages, would make sense for this inquirer only if the tradition’s final goal had a stronger pull—for this inquirer.

It can be a vexed matter to decide how much work in comparative religion is appropriate prior to opting for, assenting to, a particular religion. The decision is prudential and needs to be made by the individual inquirer, in view of individual circumstances, just as the individual must decide how much time to spend studying the languages and history relevant to assessing the specific revelatory claim that initially appealed. Important though these investigations may be, the inquirer will want to avoid being caught in Kierkegaard’s “infinite parenthesis”, examining by-ways that divert attention from the vital decision about assent. As Newman insists: time is short, and death is certain.

3. Defense of the Third Premise

Skepticism about our (instantiated) argument is likely to settle on its third premise.8 Here, again, is that premise, along with the lead-up:

(1)

Some of my grievous wrongs to others have generated debts that cannot be paid in this world.

(2)

I judge there’s a realistic possibility Christianity is true; Christianity is more plausible and appealing to me than any of its theistic competitors; and I judge that Christianity, if true, offers a path to recompense those I’ve wronged beyond present repair.

(3)

If (1) and (2), then I’m obligated to assent to Christianity.

Our initial defense of (3) comes by way of an analogy.9

Imagine that an experienced cave explorer is trapped in a cave, along with a less seasoned companion. Nobody knows their location. The companion was badly injured after they both fell: he’s unconscious, needing life-saving medical attention, and needing it soon. Attempts to establish communication with the outside world have failed, though at times the explorer believes that he sees a light, even that he hears a voice, from above. He’s not sure the only way he’ll survive is by attempting escape—possibly if he waits a while, he’ll be found—but he’s virtually certain his partner will die if they wait for rescue. To make matters worse, the veteran explorer is responsible for his associate’s injury. The less-experienced explorer hadn’t wanted to attempt the maneuver that resulted in their both falling, but the expert had pushed. The skilled climber, resenting the accolades the up-and-coming novice had been receiving from the spelunking community, harbored the dark thought before leading his partner into danger: I hope he takes a tumble here; that’ll show him what true climbing mastery involves. In short, our expert explorer has badly wronged his companion. Shocked by his own malice and its appalling consequence, he is desperate to right the wrong insofar as he’s able. He thinks he spots a way to scale the cliff they face, to reach an exit. He sees no other plausible egress. Will he be able to negotiate the cliff and avoid plummeting to his death? He’s far from certain. He can’t assign a specific probability to the chance of success. But he thinks it’s realistically possible he’ll succeed. It’s not highly implausible; it’s not a suicide mission. If he weren’t responsible for his companion’s injury, maybe he’d postpone for a time the attempt at ascent, hoping to be rescued. But under the circumstances, he judges, with good reason, that it’s obligatory for him to act: obligatory to attempt to scale the cliff.

We might, with questions regarding religious pluralism in mind, extend the analogy. Imagine that halfway up the explorer’s attempted exit, he glances to the side and spots, at a distance, a different route that could lead up and out. He didn’t initially see it as providing an escape, but from his present vantage point, it looks viable. If he and his companion had been stuck elsewhere in the cave, it might have been the only sensible route to attempt. But if it’s reachable at all from where he is (and it might not be), it would require exhaustive effort, and it’s ill-advised for him to climb down and hazard the alternate track. Perhaps it passes through his mind that the path he is on, and the alternative to the side, could converge at a higher juncture, leading to a single exit from the cave.10

The analogy could be extended differently. Imagine that halfway up, the explorer spots a different route, one that had thus far been hidden from sight; it looks like a better bet than the one he’s on and seems accessible. Indeed, the path he’s on may now have unanticipated blockages, and the alternate path looks like it may be the only one offering a viable exit. In this case it makes sense for the explorer to change his escape route.

The cave explorer is obliged to act, to take the first step of committing to the best realistically possible path up and out, and to keep going. The agnostic inquirer we have been describing is obliged to act, to take the first step of committing to the path of assent or belief, and to keep going. In neither case is the first step sufficient; but in both cases, the initial commitment, the initial step onto the path, is necessary. The cave explorer hopes he’ll finish the path, surmounting barriers to exit the cave. The agnostic inquirer (or the agnostic who crosses the threshold to belief) hopes she’ll finish the path, overcoming obstacles to reach a final beatitude and communion encompassing those she’s wronged and those who’ve wronged her.

Notice that despite the inquirer’s hope to complete the path, it is a mistake to see our argument as underwriting an obligation merely to hope that Christianity is true, and not an obligation to believe that it is true. The first two premises of our argument generate an obligation to assent or believe because assent or belief (we use the terms interchangeably here) is required for the recompense mentioned in premise (2) of the argument. Mere hope does not suffice for that recompense, though of course hope for the future is part of the worldview one steps into in assenting to Christianity.

Notice further that the agnostic inquirer’s journey along the path she chooses involves a series of steps regarding belief itself. Belief in (assent to) Christianity is something that grows and develops through time: “Lord, I believe; help my unbelief” (Mark 9:24). Belief is, in this respect, like love. We are commanded to love one another, Christianity tells us. But that is a lifetime project. People often present themselves in quite unlovable ways. As we make acts of love to develop the full virtue and habit of love, so also do we make acts of belief to develop robust belief and faith. We do what we can to acquire the fixed dispositions.

Recognizing that belief grows over time provides a response to the person who says: “You can’t just up and believe that Christianity is true; it’s not a matter of will.” Of course, there are plenty of propositions you can’t up and believe. Try forcing yourself to believe that your head is made of glass, or that you’re an arthropod, with six legs, or that whales are singing the sextet from Lucia di Lammermoor. Correlatively, plenty of propositions are such that you can’t help but believe them, and surely don’t will yourself to accept them (your head is not made of glass, etc.). Many other propositions, though, propositions with an alien aspect, are such that it’s not out of the question that you will yourself to believe them. Think, for instance, of the claim that there are different-sized infinities. Lots of well-educated people disbelieve the claim when they first hear it; indeed, they take it to be preposterous, partly because a good education includes exposure to the idea of infinity and the initial idea seems at odds with the notion that infinities come in different sizes. But once people read about the consensus of mathematicians or go through Cantor’s proof, they may tell themselves that they need to assent to the odd proposition that there are infinities of different sizes—even if they don’t understand it very well. Over time they assent more fully, and the belief seems less strange. And similarly, perhaps, with a host of other propositions that are not obviously true at first glance: The earth revolves around the sun; or Light and heavy bodies fall at the same velocity; or There’s no universal time: time is frame-relative; and so forth. There is evidence for these propositions, strong evidence—however, the evidence doesn’t initially present in that light.

Take a somewhat similar example, with a significant difference. Imagine that you are told by a good friend that you’re too judgmental, and that you should be less negative and give people the benefit of the doubt. You may decide to accept this advice as appropriate; you may decide to believe that your friend’s characterization of you is apt, and you are going to start thinking better of people. Assent to this proposition can be immediate, but internalizing the belief, getting to the point where you easily and effectively give others the benefit of the doubt, takes time. This example is like the mathematical/scientific examples in the preceding paragraph in that the belief one initially assents to has a foreign aspect at the outset; but it is different insofar as the mathematical/scientific examples are supported by overwhelming evidence (ready to be discovered, though you’re not aware of it at first), whereas the example involving your judgmental attitude may involve much less abundant and conclusive evidence. Enough evidence, though, for you to assent and operationalize the assent.

So, an obligation to believe Christian doctrine (or some other religious doctrine) is an obligation to start moving, to put yourself on a path. It’s not an obligation instantaneously to attain Abrahamic belief and faith, any more than is the obligation to love everyone a requirement to love perfectly, right away. Robust belief and love are not immediate, except in extraordinary circumstances where there’s a special divine gift. You do what you can to believe, and you do what you can to love, recognizing that in both cases, considerable effort may be required. In all the cases mentioned above involving assent to a proposition with an alien aspect, we can imagine the person making the assent saying, either right before they assent or even simultaneously with assent: “I want to believe this; but I’m having trouble making myself.” (Compare: “I want to love this person; yet I’m having trouble making myself.”) In any event, the examples illustrate that one can, often anyway, force oneself to assent … when one thinks there’s an obligation to do so.

As is clear from the way we’ve been speaking, the obligation to believe is, in our view, an evidence-based obligation. At least, for many people, including us, it is evidence-based: the obligation arises after reflection on evidence. On lots of evidence. The initial assent to a doctrine, though, the initial act of belief, isn’t simultaneous with consideration of the evidence. At some point, one must let go of the evidence, however much or little one has. One thinks the evidence supports an obligation to believe, and then one lets go of the evidence and makes the judgment: one puts oneself on the path and strives to stay on it. We take this to be a general feature of assent to propositions, whether religious or not: judgments are staccato, instantaneous, timeless (i.e., the psychological act isn’t spread out through time), though the preparation for them may be lengthy indeed. (And though they are timeless in the sense just mentioned, they are not irrevocable.)

We hope the analogy involving the cave-explorer vivifies our line of thinking. But analogies are only starting points for arguments. Ideally, one wants to strip off an analogy’s accidental features and extract a general principle—as from a model in science, one can generalize to laws. A mere analogy that does not ascend to a general principle leaves the argument open to attack through dissimilarities. So, we want to articulate a general principle suggested by the analogy, a principle that grounds premise (3). To undercut our defense of (3), then, a critic cannot merely find dissimilarities between the situations we deem analogous or re-interpret the analogy. The critic instead will need to argue against the general principle we extract. Here it is:

G: Whenever I’ve grievously wronged another, I’m obligated to take a particular path or course of action, P, if it seems to me that (i) P is the only realistically possible path to pay my debt, and (ii) P doesn’t violate other obligations or impose disproportionate burdens on myself or others.

Different principles than G might be drawn from the analogy, or from modified versions of the analogy. For instance, G refers to cases where one individual “grievously” wrongs another, but it might be possible to drop that qualifier, and speak more broadly of any case in which one person wrongs another. We’ll focus on the more specific principle involving grievous wrongs because it may be easier to defend than the principle associated with the more general case—and the sub-case suffices for our purposes, since all of us have grievously wronged others. Furthermore, G specifies that “P is the only realistically possible path”, but a principle more general still might be defended, one that allows for other realistically possible paths. Again, though, we’ll focus on the simpler sub-case.

Principle G refers to a “particular” path, P. In one way, there’s never a particular path, not down to the very last detail. That’s the situation with the cave explorer: Which hand does he put first in climbing the wall he’s resolved to scale? Does he pause regularly to try his cellphone, or does he leave it in his belt and focus all effort on the upward climb? And so, too, with assent to Christianity or any other revelatory claim: religions come with multiple sub-sects. We might divide forms of Christianity into two categories, for present purposes: those that, if true, would enable one to pay one’s debts, and those that don’t provide means for recompense (some people say they are Christian but don’t believe in an afterlife; this would be a form of Christianity that doesn’t include a means to pay the sort of debt we’re talking about here). Any form of Christianity in the first category serves for our purposes.

The tail-end of G brings in the idea of “disproportionate burdens”. We can get a grip on the notion by considering how it is applied in medical ethics. At the end of life, many would say, physicians are obligated to use ordinary means to preserve life but need not employ extraordinary means. Whether a means is extraordinary is tied to its costs and benefits. Physicians may be obligated to provide painful life-saving therapy for serious skin burns suffered by a 10-year-old, treatment that extends for months; the therapy counts as ordinary here. But in the case of a 95-year-old, the same treatment may be extraordinary, and not obligatory (perhaps not even permissible, unless the patient is demanding it): the burdens are disproportionate to the benefits.

Why include a clause regarding “disproportionate burdens” in G? Because there are counterexamples to the principle if the caveat is not added. Imagine, for instance, that your spiteful words have hurt somebody terribly and you owe them an apology. They’re dying, and you have a last chance to apologize. When you get to the person, though, you realize they’re in no shape to listen to your apology, which would remind them, painfully, of an earlier, bad stretch of time. It’s too late. You could pay what you owe but judge the burden would be too great.

Our talk of weighing benefits and burdens may bring Pascal’s Wager to mind. But the argument we are defending is not a Pascalian Wager. Pascal presented his Wager as a decision procedure for maximizing benefits to oneself. In our argument the aim is not to maximize benefits, either to oneself or to others—the aim is to meet one’s obligations, discharge one’s responsibilities, pay one’s debts. So, while the Wager looks broadly to anticipated benefits if Christianity is true, we focus on a specific aspect of Christianity, one that can be seen as a “benefit” only with some awkwardness (and is not catalogued as a benefit in discussions of the Wager with which we’re acquainted): the discharge of weighty obligations, the payment of serious debts.11

Finally, we will note that there may be circumstances in which somebody inclined to assent to Christianity might violate an obligation to other persons by the act of assent, assuming the act is operationalized. For instance, someone might marry, sincerely and solemnly promising their spouse to remain a committed atheist. In such a case, the act of assent to Christianity would require breaking a promise, violating an obligation, though perhaps not an ultimate obligation (some promises shouldn’t be made). For present purposes, cases of this sort can be set aside; they will be excluded by G-(ii). But there is a broader objection to premise (3) that invokes G-(ii), the clause in G that says other obligations aren’t violated, an objection that threatens to pull the rug out from under our argument. To this objection we now turn.

4. Response to an Objection Regarding Epistemic Obligations

It may be objected that assenting to Christianity in the circumstances described, where all one has is the judgment that Christianity is “realistically possible”, violates one or more epistemic obligations.

What, exactly, is the epistemic obligation that the contemplated assent is supposed to violate? (What’s at least one?)

Our argument has assumed a situation in which the evidence for Christianity is less than overwhelming—perhaps even less than fifty-fifty. We have imagined that the evidence may support merely a “realistic possibility” that Christianity is true. And typically, in situations where evidence for the truth of a proposition is as low as this, it seems appropriate not to believe the proposition, but rather to keep it as a live option, perhaps deserving serious study, and reserve judgment until further evidence is in hand. Only with additional positive evidence, we typically think, will assent or belief be appropriate. Yet Christian doctrine exhorts us to believe: it’s a central injunction in the gospel of John and in other core teachings as well. So, in the circumstances we’ve imagined, with a case that may be below fifty-fifty, it looks like there’s an evidential gap, a gap between the evidence for the religion and the assent the religion enjoins. With this in mind, we can take a first crack at specifying a relevant epistemic obligation: It’s always wrong not to proportion the degree of belief to the evidence.

Now this primitive proportionality precept is seriously problematic. Through the exercise of enumerating some difficulties with it we can consider whether a more sophisticated version might serve, and thereby get at the heart of the problem with any version of a proportionality principle.

Begin with the question of what the “evidence” referenced in the precept is supposed to include. Not, presumably, all the evidence that there is, period. Non-human animals may have evidence we don’t. (“What is it like to be a bat?”) God, if there is a God, has evidence we don’t. The precept must reference evidence available to human beings. All evidence available to any human being? No one of us has this, and we don’t even have it collectively, because there’s evidence available that humans haven’t acquired.

Need we consider only evidence we can obtain without great exertion? Some evidence will be empirical, and relatively easily secured: in W. K. Clifford’s example of the ship owner with a duty to investigate his vessel’s seaworthiness, empirical investigation would suffice (Clifford 1879). But there are abstract questions, debated throughout the history of philosophy, that are not so easily settled, and these figure into deliberations.

Further, a key question in contemplated searches for evidence is this: Will obtaining the evidence violate an obligation? There are many situations where we could acquire new evidence if we were willing to violate obligations. By nosing around in somebody’s purse while she is out of the room, you can get all sorts of evidence about her. Flying a drone over your neighbor’s garden will get you evidence about his affairs. These investigations violate privacy, violate the rights of others. There are other situations where you may violate an obligation to yourself (perhaps others too) if you engage in a search for evidence. In earlier times, scholars spoke of the vice of curiositas—the inordinate seeking of knowledge. If you’re monomaniacal in searching out evidence regarding, say, best opening moves in chess, you may do a disservice to yourself and your family. Or: you may have subterranean information in your subconscious that could be brought to the surface through months and months of psychotherapy; but by devoting the time and energy you’d neglect obligations to yourself and others. An overriding obligation can undercut the duty to acquire evidence.

And even if these hurdles regarding evidence-acquisition are cleared, and you obtain a “reasonable” amount of evidence without undue exertion and without violating overriding obligations, there are questions about how you take that evidence and make the proportional judgments required by the precept. Assuming you can make an all-things-considered assessment of the strength of your evidence, how do you calibrate it to a particular strength of belief, given that different scales are involved? At best, one can produce a rough correlation.

And then, how do you handle re-calibrations? Do you proportion assent to the evidence every second of the day? That’s impossible. Every time a new piece of evidence comes along? What counts as new evidence? When you first hear of Zeno’s paradoxes, do they count as new evidence against the claim that there’s such a thing as motion—a claim you know to be true? And what weight is given to the psychology of learning: must children constantly question their parents’ teachings; may students never rely on their teachers’ assurances?

Here’s a revised version of the proportionality precept that attempts to accommodate the concerns just mentioned: It is always wrong not to proportion the degree of belief roughly to the evidence for a proposition when one is aware of unanswered, troubling objections to the proposition and does not know the proposition to be true, unless roughly proportioning belief violates an overriding obligation.

Much will hang on whether the rough proportioning violates an “overriding obligation”.

We want to provide two examples where overriding obligations vitiate any duty to proportion belief to the evidence. Both examples are pertinent to contemplated assent to a revelatory claim.

First, consider a juror who has heard only the prosecution’s side of the case. He may not know that the prosecution’s case is correct and may not be aware of troubling objections to it, but he’s obligated to withhold judgment for a time: he realizes evidence from the defense isn’t yet on the table. The juror has an overriding obligation to adjudicate truth. Now this example doesn’t show that every relevant bit of evidence needs to be on the table before judgment is rendered. That’s impossible. The rules of the courtroom regarding evidence the jury may and must review will allow the juror eventually to view the relevant accessible evidence and render judgment. (At least, this is so in most cases. Very rarely, a juror might, we believe, reasonably refuse to render judgment due to a suspicion that the judge has wrongly excluded evidence—the juror doesn’t know which way the excluded evidence cuts, imagine, but withholds judgment to register protest.)

Similarly, an inquirer into Christianity recognizes that she will never be able to get every relevant bit of evidence on the table before deciding whether to assent.12 But she may also understand that a specific body of evidence not yet in her possession is pertinent to the decision about whether to assent. It’s the body of evidence that comes after one has decided to live a religious, perhaps specifically Christian, life.13 Such evidence is found in commitment to substantive moral principles, not only those that are naturally appealing (“be compassionate”), but also those that are counterintuitive (“never lie, no matter the cost”). It’s found in dedication not merely to classic virtues, but also to what Hume called “monkish virtues” such as self-denial, humility, and fasting. It’s found through prayer, reading Scripture, and receiving sacraments. Just as the juror’s duty to adjudicate the truth requires hearing the defense’s case, the inquirer’s duty to adjudicate the truth may dictate obtaining a body of evidence thus far unexplored.14

Second, consider a situation where you are presented with some evidence that your friend or spouse has betrayed you … it’s troubling evidence, but you may have an overriding obligation to retain your belief in the fidelity of your friend or spouse. Why is it that one is—sometimes—obliged to believe a friend or spouse though there may be disturbing evidence on the other side? Friends and spouses don’t believe in each other’s good intentions only when proved; they don’t seek out evidence of bad intentions and are slow to accept such evidence when it’s presented. Sometimes we have not only the right, but the duty to believe that a friend or spouse is a person of integrity, though we don’t know that they are, and are aware of unanswered and troubling evidence for the contrary claim. It might be thought that the relevant duty is not to believe a proposition, but to trust. But the duty to trust a friend or spouse entails the duty to maintain certain beliefs about their trustworthiness. Management of beliefs about the fidelity of friends and spouses takes place in light of broad obligations. Similarly, management of religious beliefs takes place in light of broad obligations.

In sum: any defensible proportionality precept will need to include an escape clause for overriding obligations, and that clause will render the precept toothless in the attempt to block assent to a putative revelation.

Our discussion of potential violations of epistemic obligations has thus far centered on proportionality precepts. Are there other kinds of epistemic obligations violated by assent to Christianity under the evidential conditions we have described?

The notion of an “epistemic obligation” does not have a widely accepted definition. It might be stipulated to be something you must do to attain truth, or knowledge, or belief, or understanding—either to attain one of these things generally, or to attain it regarding a specific question. If we’re focused on attaining truth or knowledge at the general level, then our list of epistemic obligations could include precepts such as Descartes’s heuristic: “Reduce complicated and obscure propositions to simpler ones.” Specific practices or endeavors come with more specific precepts: “To learn about your opponents’ bridge hands, bid higher than your cards warrant.” Or: “To learn calculus, first learn algebra.”

All these precepts, general and specific, are conditionals, either explicit or implicit. (Though the Cartesian precept doesn’t explicitly rest on a condition, Descartes’s Rules are predicated on the resolve, among other things, to “Never take what is false to be true” [Rule 4].) And they are conditional obligations such that one can always ask: Must I accept the antecedent condition, and if so, why? One may ask, for instance, with a focus on Descartes’s heuristic concerning complicated and obscure propositions: Am I really obligated never to take what is false to be true? The answer, of course, is no. The person resolved to take this precept as controlling will have a paltry and pitiful set of beliefs: to live well we must risk mistakes. The conditions of the more specific precepts can also be questioned. The game of bridge provides means of learning about opponents’ hands, but am I obligated to learn about my bridge opponents’ hands? I have no absolute obligation to play the game, and even if I do play, no absolute obligation to do it well, with an eye solely on winning. Am I obligated to learn calculus? Not unless I have some other, antecedent obligation: to pass a class, or be a good engineer, for instance.

All epistemic obligations are conditional obligations, contingent upon an end being sought. Some ends—though not the ones considered in the preceding paragraph—are given to us, are decided over our heads, so to speak, either by the sorts of beings we are, or the nature of things in general, or perhaps God. In these cases, when one says an epistemic obligation is “something one must do to attain truth or knowledge or belief or understanding”, the “must” involves a non-arbitrary end, an absolute end, an end simpliciter.15 Absolute ends generate absolute obligations that can’t be violated. I have an absolute obligation to help others because I’m a social animal and my end is tied to sociability. And that generates an absolute epistemic obligation: I’m obliged to acquire some knowledge or understanding of how to help others.

So, when presented with any precept specifying an epistemic obligation—whether it’s a proportionality precept or some other sort of rule—one may, and often should, ask: Why am I obligated to act in accord with this precept? If there’s an overriding obligation, the answer will be: You’re not. In some cases, involving absolute obligations, the answer will be something like: Your end, decided above your head, demands it.

The point may be illustrated through reflection on a putative epistemic obligation not (at least on the face of it) involving a proportionality precept, an obligation that might be proffered as undercutting our principle G. Consider the claim: The contemplated act of assent to Christianity violates the epistemic obligation to remain always open to the evidence.

Why would it be thought that assenting to Christianity breaches an obligation to remain open to the evidence? Perhaps because some popular forms of Christianity16 demand whole-hearted and resolute assent, which involves fixing the will consciously to set aside some worries or attacks or objections. Of course, the question that titles this paper is “Can I be obliged to believe?”, not “Can I be obliged to believe resolutely?” Still, given the prevalence of forms of Christianity that appear to require resolute assent, the objection is worth taking seriously. Furthermore, it might be thought that assenting to a religious doctrine without a case better than fifty-fifty (if not considerably stronger) itself violates an obligation of openness. So, for more than one reason we might consider whether assent to Christianity flouts an epistemic obligation to be open to the evidence.

But brief consideration turns up situations where it appears there’s an obligation to cease searching for new evidence and even cease being willing to examine new evidence when it’s put in front of your nose. People who succeed in business take chances: they can’t keep rethinking decisions, looking for new evidence and second-guessing themselves whenever new evidence is put before them. The medievals thought astrology was a dishonorable study. They didn’t deny it worked but thought it might work via the devil. Many of them held that, however it worked (if it did work), it wasn’t good to try to look into the future astrologically: the evidence wasn’t to be sought or examined. And of course, an enormous amount of the knowledge or information people seek these days isn’t honorable. “If you want to learn a lot about pornography, get on the internet late at night.” That’s an epistemic precept with a questionable antecedent.

Further, it’s permissible to take a course of action that one can confidently predict will produce new evidence for a position, without seeking new evidence against the position. For instance: One can join a group of people committed to a certain cause, perhaps nuclear disarmament, or criminal justice reform, knowing one’s beliefs about the issue will get stronger and the sort of evidence one will be able to access will shift. We don’t criticize people for this general sort of activity, though we might criticize the specific cause or ideology. It all depends on the end: Is it a good end?

In thinking about how assent to Christianity—or any religion—can be reconciled with being reasonably open to new evidence, it may help to circle back to a point we made towards the beginning of this paper, concerning the indefiniteness of the object of belief. What is it that one assents to when one assents to a putative revelation? Assenting to Christianity doesn’t require taking every sentence in the Christian Bible as literally true. One assents to the narrative or message or doctrine, however one defines it. And one understands that a rider is attached to the content: we say we assent to “that, or something like that”, allowing room for variation in insignificant details, room for growth in understanding the ramifications of new evidence. In the case of an inquirer with the motivations we are imagining, the essential elements of the Christian message, the ones that can’t be scuttled (without abandoning the initial aim), are the elements that, if true, would enable one to pay one’s debts.

There are limits, of course, as to how far a doctrine can be interpreted or stretched to accommodate new evidence without its turning into an altogether different doctrine. Sometimes an inquirer may judge that evidence dictates a very different path in pursuit of the final goal, the final good. Recall one of the alternative scenarios we presented involving our cave explorer, in which he’s halfway up what he hopes is an exit route and sees for the first time a different route that looks significantly better than the one he initially chose, a track accessible from his present position. It may be incumbent on him to access the newly recognized path. And likewise for the religious inquirer who, having committed to a particular religion, comes to think, perhaps based on new evidence, that there’s a preferable path. Belief in a revelatory claim, even resolute belief, doesn’t mean joining a cult that forbids acting on conscience and changing direction. For both the cave-explorer and the religious inquirer, questions about whether and when these major changes are necessary are prudential: algorithms don’t settle matters. But an obligation to assent—even an obligation to assent resolutely—can coexist with the epistemic obligation to remain open to competing evidence.

Stepping back, then, from details of specific epistemic obligations, where have we arrived? In sum, no matter what specific epistemic obligation is cited as an objection to the argument we defended in the preceding section, it can always be asked whether there is another, overriding obligation. Epistemic obligations are subordinate to obligations simpliciter, those “decided over our head”. And we know of no obligation simpliciter, epistemic or otherwise, that renders the contemplated assent to Christianity impermissible.

The agnostic inquirer’s will to secure the good allows her to assent to Christianity though she has but a realistically possible case for its truth. The good, payment of her debts, demands the action—and the reality she reaches towards is the source and apex of the good.