Profitable Discourse

The Pursuit of Theological Clarity



    Countless battles have been waged on a particular expanse in the vast landscape of religious thought—within this expanse lies the question of determining the truth or falsity of theological beliefs. And though this has a long and complex history, I want to draw our attention to a modern proposal whose influence has expanded dramatically in our present age. According to this proposal, theological beliefs ought to be judged on the basis of their moral impact. This proposal emerges from a deeper philosophical assumption, which is that the truth or falsity of beliefs is less important than the impact beliefs have on someone's life.

     Now while it's clear that this proposal has a number of obvious benefits, I am nevertheless convinced that it's deeply flawed. But before I present my arguments, let's look at this position more closely. The philosophical assumption behind this proposal is essentially part of a pragmatist epistemology, which essentially means a theory of knowledge whose highest values are usefulness and utility. The primary argument is that one ought to be less concerned with the correctness of a belief and more concerned with the practical ramifications of a belief—and this is actually very helpful. It has the benefit of immediately eliminating the philosophical partition between theoria and praxis, between orthodoxy and orthopraxy, between beliefs and practices. Both ideas, values, and convictions on the one hand, and habits, actions, and practices on the other ought to be conjoined into a holistic vision of epistemology and ethics. 

     The problem, however, is that this position ends up eliminating the partition not by bringing epistemology and ethics together, but by absorbing epistemology into ethics. Yet for many Christians, especially post-evangelicals, progressives, and liberals, who all rightly oppose fundamentalism and value intellectual humility, this seems almost automatically true. I'll admit it isn't too difficult to criticize some parts of rigorous theological traditions for all too often manifesting in platitudes instead of hard action. And if you've grown up in such a tradition, the path to the pragmatist framework can practically be traversed in a single step. Therefore, this proposal has proven very attractive to many—yet I’m not sure it’s a helpful way forward.


     The proposal with which we're concerned is that beliefs ought to be judged on the basis of their moral impact, which is to say that the impact a belief has on someone's life is more important than the belief itself. It seems clear to me that this proposal faces three insurmountable problems: first, it's self-referentially incoherent and circular; second, it's impossible to demonstrate; third, it forces one into the fallacy of affirming the consequent.

1. It is self-referentially incoherent and circular

     The first problem is that this position fails on its own terms. This becomes clear once one realizes that this position, “that beliefs ought to be judged on the basis of their moral impact,” is itself a particular belief. And if the claim “all beliefs ought to be judged on the basis of their moral impact” is itself a belief, then it ought to be judged on the basis of its moral impact—but it can't be. See, actively believing only those beliefs which produce the greatest moral impact is one ordeal, but believing that those beliefs which produce the greatest moral impact are the correct beliefs is an entirely different ordeal. One is concerned with immediate moral pragmatism, the other with general truth. One can’t move to the latter from the former, one can only assume the latter to move forth into the former. At some point one has to provide some sort of justification for why the claim “all beliefs ought to be judged on the basis of their moral impact” is even worthwhile, and this is where the problem emerges. This is because if its justification is based on its moral impact, then the argument assumes its conclusion and thus engages in fallacious circular reasoning. But if its justification is based on its being true, then the argument’s very thesis itself is denied.

     Look at it like this. Let's say it’s false that the impact a belief has on someone's life is more important than the belief itself. If it's false, then we shouldn't believe it, we shouldn’t judge other beliefs accordingly, and we shouldn’t live accordingly. But let's alternatively say it's true that the impact a belief has on someone's life is more important than the belief itself. If it's true, then we should believe it, we should judge other beliefs accordingly, and we should live accordingly. But do you see the problem? If this is granted, then the primary motivation and justification of this very belief would stand in direct contradiction to its conclusion—namely, one would believe it not on the basis of its moral impact but on the basis of its truth. Yet to proceed from this foundation (beliefs are judged by their truth value) and then turn in the opposite direction (beliefs are judged by their moral impact) is like cutting off the wings of a jet plane while it’s still rushing down the runway—it'll never get off the ground. Therefore, this proposal rests on the inescapable horns of a dilemma between circular reasoning and self-referential incoherency. And this only intensifies as we address its further problems.

2. It is impossible to demonstrate

     The second problem for this proposal is that it’s impossible to demonstrate, meaning that while it seems possibly true in general that all beliefs ought to be judged by their moral impact, this becomes increasingly difficult when analyzing specific beliefs and practices. This is essentially an epistemological form of a long-standing and potent critique against pragmatist ethics.

     Here’s how the critique works: in order to determine whether a belief is virtue-producing or positively-impactful, one must demonstrate that the belief is what philosophers call optimific. To claim that a belief is optimific is to claim that it produces actions whose benefits outweigh their drawbacks. So to determine whether a belief is optimific, one must 1) discover its benefits, 2) discover its drawbacks, 3) determine the balance between benefits and drawbacks, and 4) determine whether the benefit/drawback balance is greater than any other similar belief. Thus if one wants to prove the conclusion of this proposal with regard to any actual belief, they must know 1) all possible and actual actions proceeding from the belief, 2) all possible benefits and drawbacks of these actions, 3) all the associated options to be considered, and 4) the overall value of all options in light of all circumstances. Only then can they compare these differently-valued options to each other in order to decide which action is optimific and therefore whether or not the belief in question is optimific. This is an incredibly difficult task to say the least. And therefore, since endlessly existent criteria make it nearly impossible to establish that any belief is satisfactorily virtue-producing, this proposal provides an unworkable epistemological and ethical theory. 

     To see how see how this might be the case, let me show what one would need to accomplish in order to successfully determine the correctness of a given belief on the basis of its moral impact. It’s a three-part problem:

A. One must obtain and defend detailed and accurate statistics that are nearly impossible to obtain.

     Say you think penal substitutionary atonement ought be rejected because of the lack of positive moral impact it produces. Maybe you grew up in a morally toxic environment at a church who held this belief. But while your story might be emotionally resonant, it might also be a gross mischaracterization of Christians at large. Many who hold to PSA might be radically virtuous individuals, and PSA statistically might thus be proven so virtue-producing as to be judged correct and in need of acceptance. Thus many beliefs which you currently reject might actually produce the greatest positive impact in their category, and thus ought to be accepted. See, you need to know what kind of people PSA produces all over the world, all throughout history, with all things considered. In fact, what if recent statistics demonstrate that PSA has ceased to be produce positive moral impact but also that its surplus of historical moral impact will outweigh all its drawbacks, all things considered, until we reach 2200 AD? Unfortunately, you’d lose your justification for rejecting it (and would only be justified in 2200, should it persist in having a negative impact). 

      Also, which examples of moral impact should be cited as flowing from rejecting PSA (in exchange for say, the Moral Example Theory of Atonement)? Let’s say this decreases anxiety, decreases one’s inclination towards violence, and increases one’s positive perception of those unlike them. In order to justify the rejection, one has to prove, with actual statistical data, that these actions and dispositions flow from rejecting PSA and not from anything else (or not predominantly anything else). What if someone shows a decrease in their inclination towards violence, but because of their belief that God will bring about true justice and not because of their rejection of PSA? Or what if someone shows an increase in their positive perception of others, but because of their belief that Jesus calls them to love others as he loved them and not because of their rejection of PSA? I discuss this further below, but the data and calculus required for this decision making would be overwhelming.

     There’s one more possible move here: perhaps, finding this proposal near impossible to demonstrate on the general level, you decide to turn it inwards so as to make it a theory that only prescribes one’s personal epistemology and ethics. This seems like a strategic move, since you’d both escape the need for required data and only prescribe beliefs according to your personal inclinations (to which you have immediate and generally accurate access). But this move comes at a cost–most seriously, you lose your purchase on contributing to any ethical discourse and you provide full moral justification for any belief (it actually begins to resemble forms of relativism). In other words, those who ascribe to PSA, or who think all homosexuals should right now go to hell, or who believe that America is God’s only chosen nation, all gain full epistemological justification and clearance so long as they cite these beliefs as producing positive moral impact. 

     Now maybe you could refute this. Sure, there are certainly a few individuals for whom these above beliefs produce positive moral impact, but by and large there isn't any positive moral impact being produced. But who says the ethical values by which you judge are themselves correct? Who says what “positive moral impact” or “virtue production” must be in order that a belief might attain correctness? See, if you argue that love, kindness, tolerance, &c., are the positively-impacting moral values because the belief that they are itself produces positive moral impact, then we’re back to the fallacious circular reasoning

     Let’s say someone who ascribes to the above beliefs argues that speaking the truth (for them, warning homosexuals of impending hellfire) is a far richer moral value than social tolerance? Or worse, what if they think not warning homosexuals itself must necessarily flow from a belief that doesn’t produce positive moral impact? You’re only way out is to move the theory back into the general and objective sphere, but then you return to facing the impossible task of demonstrating that any belief is actually correct. You couldn’t even claim that Paul’s Fruit of the Spirit is the guide to determining which beliefs produce positive moral impact, since someone could always claim that believing in a hermeneutic that dismisses some of Paul’s theology itself produces the most positive moral impact (and any Christian who rejects Paul views of women wouldn’t have much of a response). 

B. One must demonstrate that these beliefs will always be the best beliefs over time with their total moral impact measured and accounted for. 

     This is also impossible since no one can reasonably predict the causal relationship between and resultant consequences of beliefs and actions centuries or even generations into the future. This goes back to my discussion about how the proposal would work should statistics show that PSA’s overall benefits would outweigh its drawbacks until 2200. But furthermore, what if after a couple centuries of drawbacks, believing in PSA proved again to have produced the greatest amount of moral impact in the greatest amount of Christians, all things considered, a thousand years from now? One cannot determine whether or not this is the case, but the proposal requires such deliberation (unless one turns to the personal-relativistic version, which has its own problems).

C. One must demonstrate which beliefs work for which people at which time. 

     Identical beliefs will produce different actions and have different impacts on different people in different cultures, both at the same time and at different times. Let’s say the belief that God punishes sinners with some form of eschatological punishment is one which you think ought to be rejected on account of its lack of positive moral impact. Could it really never produce any moral impact? On the contrary, for someone who has been submerged in war and violence most of their life and has seen atrocities committed even against their own friends and family, the belief that God will finally achieve justice (through punishing sinners) might be the most powerful force in their life leading them towards hope and love and away from despair, anxiety, and revenge (see Miroslav Volf’s work for instance). 

     For some, this belief might have a negative impact and produce no virtue, while for others it might have a dramatically positive impact and produce much virtue–one has to determine when the belief is correct and incorrect, when it is to be accepted and rejected. Thus one must both be able to determine which individuals are properly disposed for which beliefs to produce which actions, and must be constantly updating their catalogue of which beliefs produce positive moral impact—since beliefs that produce positive moral impact right now might not in a generation and vice versa.

3. It forces one into the fallacy of affirming the consequent

     The last problem this proposal faces is that it necessarily forces one into the fallacy of affirming the consequent. Affirming the consequent is a form of reasoning in reverse—like reasoning that since you have a sore throat, then you must have the flu. This is fallacious since the flu is not the only cause of a sore throat–you might actually have a common cold. In the same way, one cannot reason that because someone performed some virtuous action, that they did so because of one specific belief–they could have performed it for any number of reasons.

     So in order to justify the claim that a belief ought to be judged by the merit of its moral impact, one must be able to draw a straight line from subsequent behavior to antecedent belief, but this isn’t possible. In order to do so, one would need to be able to demonstrate that the positive impacts on and virtuous actions of a person's life are either 1) only or 2) predominantly caused by a particular belief or by a set of beliefs while holding everything else in their life in consideration. Let’s say the belief that God as father loves all and punishes none is a belief considered to produce positive moral impact, and therefore to be correct. Let’s further say that many wonderful parents, whose virtue cannot be denied, are revealed as holding to this belief. Now what if in the course of conversing with one of these parents, we are surprised to hear them explain that the reason they love their children so well is not because of this belief at all. Rather, they claim, they love their children because their own parents were abusive and distant, and so the tremendous, ever-present force of reversing the trauma of their own upbringing is responsible for their virtuous actions. 

     Essentially, then, we’d need not only statistics but detailed polls and interviews in order to properly judge the relationship between beliefs and practices. And worse, we’d almost always have less virtuous actions than reasons why they came about. This would mean both that generalizations would forever be suspect and that it would prove tremendously difficult to determine which beliefs produced which actions (do they work at the soup kitchen because of belief A, B, C, or D?).

     Virtuous behavior can be produced by many other things than only a specific belief or set of beliefs. For instance, virtuous behavior can emerge from hypocrisy, or it can emerge more on the basis of emotion and intuition than on the basis of a belief. It could emerge more on the basis of random circumstance and implicit reaction than on reasoned, conscious, intellectual conviction. It could emerge from a complex arrangement of beliefs differing from person to person or else containing some beliefs that could be judged good and others poor. I could go on, but these all strike at the heart of the proposal.

     Therefore the proposal, in order to be of any practice guidance, must start somewhere in naming some beliefs as producing positive moral impact and others as not. But this project fails from the outset by affirming the consequent, in that the good behavior picked out in the process of determination could always have been produced by a different cause than the belief identified as its cause. This proposal, then, is forced into fallacious logic before it can even begin.

     With these arguments in mind, I think the first likely provides the clearest way forward. One must return yet again to a process of judging beliefs on the basis of their truth or falsity rather than on the basis of their moral impact. But as I mentioned at the outset, this needn’t cause anyone to establish yet again a separation between epistemology and ethics, between beliefs and practices. My larger argument is simply that the initial benefit of this pragmatic proposal should be taken seriously, so that our beliefs and practices receive balanced attention without one being absorbed into the other. 

     John Raines is a Master of Arts in Theology student at Fuller Theological Seminary, co-host of the podcast Reconstruct, and blogs at Profitable Discourse. He lives with his wife in Seattle, Washington.

© John Raines and Profitable Discourse, 2016. Unauthorized use and/or duplication of this material without express and written permission from this site’s author is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to John Raines and Profitable Discourse with appropriate and specific direction to the original content.

Site artwork by John Martin (1789-1854)

Design and Branding by John Raines