Saving faith

Saving faith is a disposition to actively seek after God, constituted by the following three sub-dispositions: first, holy curiosity – a disposition to inquire about God; second, holy concern – a disposition to seek right relationship with God; and third, holy credulity – a disposition to give credence to and pattern one's life in accordance with supposed general or special revelations of God that satisfy one's holy curiosity and concern. Children are paragons of saving faith: they are natively curious about God, tender of conscience, and credulous. Hence Jesus teaches: ‘Truly I tell you, unless you change and become like children, you will never enter the kingdom of heaven’ (Matthew 18:3).Footnote ⁴

That children are paragons of faith indicates that the virtue of faith is odd in comparison with other virtues. Most virtues are not paradigmatically displayed by children, because most virtues take time to develop. The virtue of faith is not like this, a fact that is signalled in the Christian theological tradition by describing the virtue of faith as a gift. One way to understand what it means to call the virtue of faith a gift is that the dispositions constitutive of faith are inborn; persons don't have to work to gain them, although they do have to work to refine them and can diminish and even lose them.Footnote ⁵ That is, God creates us with holy curiosity, holy concern, and holy credulity, dispositions that have God, and God only, as their rightful target. To speak of these sub-dispositions as constitutive of a virtue is to point out at least the following two things. First, these sub-dispositions orient us to a specific good (God) that is crucial for human flourishing. Second, these sub-dispositions are more than instinctual; they are habitual, meaning they can be increased or decreased in response to reason. Like all virtues (but unlike, say, faculties) faith orients us to some good that is crucial to human flourishing. But unlike most other virtues, faith is given to us at birth.

So saving faith is the virtue constituted by the dispositions to inquire about God, to seek right relationship with God, and to give credence to purported revelations of God that appear to satisfy the demands of enquiry about and/or right relationship with God. Holy concern, the disposition to seek right relationship with God, provides a link between holy curiosity and holy credulity. It is because persons care to be in right relationship with God that their curiosity is neither idle nor detached. Such concern lends urgency to the religious quest, making neutral dispassion and objectivity inappropriate. Persons sense – though they can suppress and dull the sensation over time – that their holy curiosity has to do with something that has the power of life and death over them.Footnote ⁶ This sense gives rise to holy credulity; desperation to understand and be rightly related to God leads persons to seek out and believe whatever offers plausible claim to connect them to God.

Saving faith, like all virtues, comes in degrees. A person's degree of faith is constrained by degree of desire for knowledge of and relationship to God. There is great or robust faith at one end of the spectrum.Footnote ⁷ There is little or faltering faith at the other end of the spectrum. People typically think that what makes for great faith is a high degree of confidence in one's particular theistic beliefs. But high credence in specific theological propositions is a common effect, not a necessary condition, of great faith. What really constitutes saving faith is the set of dispositions outlined above: holy curiosity, holy concern, and holy credulity. One's degree of faith increases in proportion to the increase in these dispositions; the more disposed one is to inquire about, seek right relationship with, and give credence to plausible revelations about God, the greater one's saving faith. And a person who lacks these sub-dispositions altogether lacks saving faith.

Saving faith precedes and undergirds propositional faith (faith that God exists) and relational faith (faith in God). One can possess or embody saving faith, in fact, without possessing or embodying these other kinds of faith. One can have the virtue of saving faith without having faith in anything or faith that anything, and furthermore, without subscribing to a particular faith and therefore without being a faithful devotee of any religion. I suspect saving faith rarely persists in isolation from these other kinds of faith because credulity prompts belief except in abnormal circumstances. Nevertheless, it is possible.

The foregoing analysis suggests the following truth-conditions for customary faith-locutions in the theistic domain:

• • Connor has faith that God exists iff Connor has some ‘pro’ cognitive attitude (belief, acceptance, assumption, etc.Footnote ⁸) towards the proposition God exists and that ‘pro’ attitude is partly caused by Connor's saving faith. (Of course, cognitive attitudes can be overdetermined. Connor's ‘pro’ cognitive attitude that God exists might be caused by his saving faith and also by a natural theological argument he's heard about the existence of God.)

• • Cole has faith in God iff Cole has placed his trust in God as a result of the manifestation of his saving faith.

• • Caden believes the gospel by faith iff Caden believes the propositions that constitute the gospel by virtue of their fulfilling the desires and concerns constitutive of Caden's saving faith. (Remember that the causes of belief can be overdetermined.)

• • The righteous live by faith iff their action/way of life is premised on a proposition or propositions towards which they have ‘pro’ cognitive attitudes produced by their saving faith.

• • Many Jews, Christians, and Muslims are people of faith iff many Jews, Christians, and Muslims possess saving faith. (Agnostics, atheists, or practitioners of other religious traditions may also possess saving faith as I've defined it.)

• • Christianity is a faith tradition iff there is some relevant collection of Christian beliefs that has been found by many to satisfactorily fulfil the desires and concerns constitutive of saving faith. (This is why it may be wrong, for example, to speak of ‘the Buddhist faith’; some forms of Buddhism simply aren't responsive to the kinds of desires and concerns that are constitutive of saving faith – which is not to say a Buddhist can't have saving faith as I've described it.)

To sum up: saving faith is comprised of an affective disposition (urgent desire for knowledge of and right relationship with God) that gives rise to a cognitive disposition (tendency to believe in plausible revelations) and a behavioural disposition (tendency to orient one's lifestyle around the dictates of plausible revelations). I repeat that one may have saving faith and lack belief (or any ‘pro’ cognitive attitude) and/or religious practice, and this for any number of reasons: because one justifiably deems purported revelations implausible; because one is rationally convinced by atheistic arguments; because one is culturally isolated and has no access to special revelation; because one has been abused by practitioners of a specific faith and is psychologically inhibited from seriously considering their tenets; and so on. In other words, one may possess the virtue of saving faith but, because of circumstances, fail to manifest its customary fruit: religious belief and practice. However, the tendency of saving faith is to prompt theistic belief and practice, in part because the telos of faith is growth in relationship with God, and such growth is deeply aided by theistic belief and practice. So although saving faith does not require theistic beliefs or practice, such are its customary fruit, and a person who had no interest in having theistic beliefs or a religious way of life would thereby be shown to lack saving faith.Footnote ⁹

Saving faith and epistemic justification

My thesis is that saving faith is a virtue that contributes positive epistemic status to agents and to beliefs or other ‘pro’ cognitive attitudes (take ‘beliefs’ in this section as shorthand for ‘beliefs or other “pro” cognitive attitudes’). The thesis includes the following three claims. First, any belief that is caused by an agent's saving faith gains some epistemic justification thereby. A belief gains some epistemic justification if there is some explanation for its being held that increases its probability of being true. Second, sometimes beliefs are epistemically justified by their connection to saving faith. Beliefs are epistemically justified if there is some explanation for their being held that increases their probability of being true and that meets some minimum threshold for establishing that probability. Finally, sometimes agents are epistemically justified by their exercise of saving faith. Agents are epistemically justified with respect to some belief by meeting the standard of intellectual responsibility that rests on anyone who aims at having true beliefs and avoiding false ones. So any belief causally connected to saving faith gains some epistemic justification through the connection, and sometimes that belief gains enough epistemic justification to cross thresholds that secure the belief's epistemic justification and/or the agent's epistemic justification with respect to that belief.

To envision the truth-tracking connection I have in mind between saving faith and theistic belief, consider a child who has holy curiosity and holy concern, stirred up by the starry heavens above and the moral law within. She is told there is a God who created the world, and who has sent Jesus to fix the mess we have made of it. She believes what she is told. She now has several theistic beliefs on the basis of testimony, among them that God sent Jesus to save the world. But it is important to my view that there is more going on here than mere belief on the basis of testimony. This particular testimony satisfies desires the child has antecedently, desires she has to know about God and to be rightly related to God. Compare her newfound theistic belief

to two other beliefs she holds on the basis of testimony,

and

All three beliefs may be accepted on the basis of testimony, but a normal child in the relevant historical and cultural contexts will be more ready to believe (J) and (S) than (M), because (J) and (S) satisfy antecedent desires. (J) satisfies the antecedent desires I've termed holy curiosity and holy concern, (S) satisfies the antecedent desire children have for gifts.Footnote ¹⁰ These antecedent desires aren't evidence for the truth of (J) or (S), but they are partial explanations for why beliefs (J) and (S) are held by the child. These antecedent desires explain, for instance, why a child will more readily believe (J) and (S) than (M). So the difference between (J) and (S) on the one hand, and (M) on the other hand, is that antecedent desire is typically part of the explanation of why a child comes to believe (J) or (S), whereas antecedent desire is not part of the explanation of why a child comes to believe (M).

Now compare (J) and (S). Suppose both beliefs are formed on the basis of testimony and antecedent desire. On my view, (J) may be epistemically justified but (S) is not, because the particular set of antecedent desires operative in moving the child to believe (J) – holy curiosity and holy concern – are designed by God to lead people to true beliefs, whereas the particular antecedent desire operative in moving the child to believe (S) – desire for gifts – is not so designed by God.

The picture is similar in many respects to the one on offer in Alvin Plantinga's version of Reformed Epistemology. According to Plantinga, humans have implanted in them by God a special belief-producing faculty called the sensus divinitatis. Under the right conditions, the sensus divinitatis produces true beliefs about God, true beliefs that aren't based on evidence but are simply produced by the sensus divinitatis faculty functioning properly under the right conditions. This set-up is according to plan; God designed us with this faculty for the purpose of producing in us true beliefs about God. And since, according to Plantinga, a belief has warrant only if it is produced by cognitive faculties functioning properly in an appropriate environment according to a design plan successfully aimed at truth, beliefs produced by the sensus divinitatis have warrant. To say such beliefs have warrant is to say at least that they are epistemically justified.

The view I defend is similar to Plantinga's, except on my view saving faith is not (as it is for Plantinga) the result of a properly functioning faculty. Rather, saving faith is itself the properly functioning faculty. Or something like it. Saving faith is not a cognitive faculty in the way that, say, perception or memory (or possibly the sensus divinitatis) is. It is a virtue, a state of character that triggers responsiveness to a fundamental good of human flourishing. I've described the constitutive dispositions of the virtue of saving faith already. My claim is that the virtue of saving faith – the virtue consisting in holy-credulity-motivated-by-holy curiosity/concern – is designed by God to prompt true beliefs when functioning in an appropriate environment. If this is true, then saving faith is plausibly a source of epistemic justification.

Plantinga envisions something like this as a possibility. ‘Perhaps God has designed us to know that he is present and loves us by way of creating us with a strong desire for him, a desire that leads to the belief that in fact he is there’ (Plantinga (Reference Plantinga2015), 43). Plantinga rightly detects the main objection such a view faces, namely that wish-fulfilment is not an intellectually respectable way of coming to believe. Indeed, as he points out, this is precisely the accusation levelled against Christian belief by the likes of Freud and Nietzsche. If it can be shown that Christian belief is a matter of wish-fulfilment, popular wisdom holds that Christian belief will thereby be debunked. But Plantinga observes that such undermining objections rest on an untested assumption.

Although this is not Plantinga's settled view, it is close to mine. God has created human persons such that it is natural to them to experience holy curiosity and concern about the things of God, such that they are superstitious, credulous, inclined to seek after and cling to claims and practices that purport to satisfy those desires. And God has designed it so that this particular set of desires and dispositions – saving faith – tends over time to track true beliefs, not only beliefs about basic theism such as the existence of a personal, loving, perfect, and holy God, but also more specific beliefs peculiar to Judaism, or Christianity, or Islam. Desire is not usually a source of credible belief, but in the theistic context it can be. Not always, of course. If the desire characteristic of saving faith operates to suppress legitimate defeaters, then it will fail to bring about justified beliefs, even as it pertains to theistic belief. Or if it operates in an environment that contains too many obstacles to its proper functioning, then it will fail to bring about justified beliefs. That is not its design plan; it's not designed to work against the deliverances of evidence and argument, nor is it designed to work well regardless of environmental obstacles. But it is designed to draw human persons beyond evidence to beliefs and practices that put them into better positions to relate to God.

Because saving faith is highly defeasible, it sometimes brings about propositional attitudes thinner than belief. But its design plan is to bring about the strongest propositional attitude consistent with the evidence. Hence theistic belief is the default cognitive result of saving faith. For instance, children are prone to theistic belief; they don't begin by wagering that God loves them, and then accepting that God loves them, and finally believing that God loves them. They just believe that God loves them when they hear it, not simply on the basis of testimony, but because the testimony attests to something that satisfies their antecedent saving faith. This is the way saving faith is supposed to work, and when it does, it can epistemically justify belief.

But saving faith can fail to bring about justified theistic belief and knowledge for a number of reasons. One important possibility is that saving faith itself may be weak, and therefore the tendency to believe a plausible revelation diminished. This is the most common biblical explanation of unbelief. When the cares and concerns of the world interfere with the cares and concerns of saving faith, holy curiosity is dampened, holy concern is quieted, and in turn, holy credulity is diminished. When worldly concerns mute these holy dispositions, firm belief in claims about God is less likely. More likely are weaker propositional attitudes: perhaps acceptance, or assumption, or hope, that God exists.

Another important possibility is that saving faith runs up against counter-evidence. A young girl with childlike Christian faith grows up and goes to college. She is devoted to God, to the works of mercy, and to sharing the gospel, but she encounters arguments in some of her classes that purport to show the religious propositions she believes are false. She can see no way around the arguments, though she begins by believing there are rebuttals. She talks to her friends, her parents, her pastor, and is unsatisfied with their responses (perhaps they are not very adept at natural theology or Christian apologetics). After a while, she finds that she no longer believes everything she once did. She doesn't disbelieve either, but she doesn't have the experience of immediately sensing the truth of these propositions upon calling them to mind. She still patterns her life around these claims, hopes they are true, assumes they are, even desperately prays that they are. She has strong saving faith but, due to its interaction with evidence, it is bringing about ‘pro’ propositional attitudes thinner than belief. Still, I want to say in this case that those thinner propositional attitudes may be epistemically justified by their connection with her strong saving faith. Were it not for her saving faith, she would no longer accept the truths she once believed, or take them to be true in her practical reasoning. Although they fall short of belief, her propositional attitudes about Christian claims are brought about by her saving faith, according to design, and they may be epistemically justified thereby.

Yet another important possibility is that saving faith's orientation towards true theistic belief is thwarted by an inhospitable environment. Suppose, for example, that a child is raised in an overtly white supremacist community. Because conscience is deeply susceptible to environmental influence, the child grows up convinced that non-white people are morally inferior, and he casts about for revelation of the divine author of his strong moral sense of racial hierarchy. Such a child may have saving faith, even strong saving faith, and if he does, such saving faith is liable to bring about distorted, damaging, false, and unjustified beliefs about God. This child is liable to believe the wrong purported revelations about God, due to operating in a severely inappropriate environment.Footnote ¹¹

And, of course, there are many causes of waning faith, many kinds of evidential challenge, and many environmental obstacles. Scripture is strict about these matters, suggesting that most of us are ‘without excuse’ (Romans 1:20). I take this to mean that most of us operate in environments in which, properly heeded and nourished, saving faith can lead us to many true beliefs about God.

Objections

Of several objections that could be raised against the theory of saving faith I have proposed, three are most threatening. Objection one: if saving faith confers epistemic justification in the way I suggest, it justifies too much. For presumably the kind of desire-motivated credulity that I have placed at the centre of the account leads people to conflicting religious views. And a belief-producing mechanism that consistently brings about conflicting beliefs cannot be a source of epistemic justification for those beliefs. Suppose perception or memory were consistently producing conflicting beliefs between people. Wouldn't that lessen our trust in perception or memory as reliable belief-producing mechanisms? Surely one criterion a mechanism must meet in order to make a claim to reliability is that of producing widely shared beliefs. But people don't just occasionally disagree about religious claims; they disagree consistently and dramatically. Therefore there is no reliable virtue of saving faith that prompts theistic beliefs.

In response, it bears repeating that a virtue differs from a faculty. Epistemic faculties are shared by all normal epistemic agents, but this is not so for virtues. So the mere existence of widespread disagreement isn't a reason to deny that the virtue of saving faith can generate justified beliefs. Rather, it would be the existence of widespread disagreement among those who possess the virtue of saving faith that would pose a challenge to the view I am defending.

How widespread is disagreement among those who possess the virtue? It is difficult to say. We can't easily know who possesses the virtue, and to what degree. But suppose, as seems entirely reasonable, that there are Jews, Christians, and Muslims with the virtue of saving faith.Footnote ¹² How widespread is the disagreement among them? There is a great deal of overlap. They will have a positive cognitive attitude towards the claims that God exists; that God rewards those who earnestly seek God; that God intends to redeem this world that has fallen into disrepair; that God is good, powerful, and just; that God is displeased with human greed, envy, hatred, malice, lust, and pride; that God is pleased with human generosity, love, compassion, loyalty, and humility; and the list could be extended at great length. But there are disagreements, of course. Christians with saving faith have a positive cognitive attitude towards the claim that Jesus is God whereas Jews and Muslims with saving faith do not. Does the existence of disagreement at this level among those who possess the virtue of saving faith undercut the claim that saving faith can confer justification on theistic beliefs? I don't think so, for the following reason.

The more specific the belief in question, the more finely tuned the virtue needs to be and the more dependent it is on ideal conditions. In this sense there is an analogy between the reliability of the virtue of saving faith and the reliability of faculties like perception and memory. For these latter faculties, too, some highly specific beliefs admit of much broader disagreement even among those who possess well-functioning faculties. For highly specific perceptual and memory beliefs, the relevant faculty has to be very strong and it has to be operating in ideal circumstances. Visual perception, for example, confers varying degrees of epistemic justification depending on these two metrics – the condition of the faculty and the propriety of the circumstances. A man with 20/20 vision sees a street sign fifty feet away and immediately forms the belief that Commonwealth Avenue is ahead. Another man with 20/40 vision sees the same sign from 150 feet and forms the belief that Cannondale Avenue is ahead. Both beliefs may possess some epistemic justification, but probably not the same amount, since both the circumstances and the functioning of the visual faculty are so much closer to ideal in the first instance. Maybe only the first possesses enough epistemic justification to count as a justified belief; nevertheless, both beliefs gain some positive epistemic status from their causal connection to the faculty of sight.

The same is true for the way in which saving faith confers epistemic justification on theistic beliefs. Strong or great saving faith exercised in ideal circumstances is likely to lead to true theistic beliefs; that is the design plan of the virtue of saving faith. But few have strong saving faith, and even those who do exercise it in conditions of varying sub-optimality. Thus disagreement on highly specific theistic claims is unsurprising. I've already said what makes for strong saving faith, what makes for a 20/20 virtue of saving faith, so to speak. I haven't said anything systematic about the conditions under which it is designed to operate, and I won't try to do that here, either. But the claim is that strong saving faith exercised in circumstances free of the most debilitating obstacles (subjection to propaganda, familial censure, extreme cultural prejudice, crushing poverty or pain, etc.) tends, over time, to narrow the range of viable theistic beliefs and to increase the justification of those theistic beliefs that are held. If something like this is true, my thesis won't be easily rebutted by the observation that people hold conflicting theistic beliefs.

Imagine a Muslim and a Christian, both of whom have robust saving faith. Imagine, furthermore, that they are in excellent religious communities. What makes for an ideal religious community will depend on what is true. However, suppose that excellent religious communities share in common openness to theological enquiry, an emphasis on deep moral formation, caution about the untoward effects of excessive family or caste pressure, and vigilance about the corrosive temptations to use religion for manipulative or ideological purposes. So these two have robust saving faith, and they spend their lives in excellent religious communities, one Christian and one Muslim. The view of saving faith I am defending predicts that, over time, there will be increasing overlap between the beliefs of these two and, furthermore, the justification that each has for her beliefs will increase. Overlap will increase because the range of viable theistic beliefs for each will narrow; those religious claims in each tradition that are primarily the result of internecine conflict or ideological control will recede from the centre of each's outlook, with those beliefs that are most able to satisfy the longings of saving faith taking centre stage. And because saving faith is doing the work here, the epistemic justification that each has for these central beliefs will grow. Does that mean that all of their respective theistic beliefs are justified? Not necessarily. Some of the claims made by each tradition are so specific that they may require not only robust saving faith, but an ideal epistemic environment for justification. And since Islam and Christianity can't both be true, at least one of their epistemic environments is less than ideal. So, for example, if Islam is true, the Christian's belief in the Trinity may not be justified. Nevertheless, the Christian may be personally justified in believing in the Trinity, and her belief in the Trinity will have gained significant epistemic justification from her saving faith, though perhaps not enough to make the belief justified.

Objection two: if saving faith is so important, why isn't it more epistemically robust? Why would God allow it to be so fallible if it is meant to direct people to the truth about God? First of all, it is not clear to me that the track record is as spotty as is often suggested. And Christian scripture, at least, supports the view that those who carefully attend to their conscience, preserving it and allowing it to shape their actions and outlook, will come to know many true things about God (Romans 1:19–20). But even if saving faith, over time, tracks the truth about God to a high degree, it is far from perfect. It remains fallible. So, the objection runs, if salvation depends on saving faith, why wouldn't God make it more robust?

God need not make it more robust because salvation depends upon saving faith, and not on the beliefs to which saving faith customarily gives rise. Indeed, I would argue that it is a feature (not a bug) of the virtue of saving faith that the beliefs that purport to satisfy holy curiosity are highly defeasible. Because of this, we never reach a state of perfect theological knowledge and must therefore continually labour to grow in wisdom. Sharpening the virtue of saving faith, by deepening its constitutive dispositions, is one of our best ways of finding our way to more adequate theological formulations. Exactly because the virtue must be so refined to work optimally, our quest for knowledge of God demands moral rigour. In this way, the epistemic fallibility of the virtue provides incentive for our growth in holiness, which is, after all, what is most important for our eternal flourishing.Footnote ¹³

Objection three: even if saving faith is truth-tracking, the agent is not in a position to determine whether her theistic belief is a consequence of saving faith or some other more commonplace form of wishful thinking, and given that wishful thinking is normally a bad way of arriving at true belief, the agent thus has no reason to trust the beliefs prompted by saving faith.

A staunch externalist will reply that the agent need not have reason to think she is justified in order to be justified. Granted, she may be in the dark about whether her belief is connected to saving faith or some untrustworthy form of wishful thinking. No matter. It seems to her that God exists, and this seeming is caused by saving faith. Therefore the belief has justification regardless of her lack of reasons for thinking her belief is connected to saving faith.

But there is more to say here, which may quiet what seem to me to be appropriate internalist worries. For although saving faith can be thought of broadly as a type of wishful thinking, it is not typical. Saving faith is not just any wish that any god exist. It is a desire for God that is rooted in holy concern. One of the most customary ways we have of talking about this holy concern points to our experience of ‘the voice of conscience’. As Newman points out, ‘Conscience does not repose on itself, but vaguely reaches forward to something beyond self, and dimly discerns a sanction higher than self for its decisions, as is evidenced in that keen sense of obligation and responsibility which informs them’ (Newman (Reference Newman1903), 107).Footnote ¹⁴ It is the experience of conscience giving rise to the sense of compunction and obligation that most commonly generates the holy concern constitutive of saving faith.Footnote ¹⁵ And therefore, an agent generally is in a position to identify differences between the desires constitutive of saving faith and more customary modes of wishful thinking.

Evidentialism, fideism, and Reformed Epistemology

Like Reformed Epistemology, the Newman-inspired view I am developing rejects both evidentialism and fideism on their most straightforward interpretations. Evidentialism is the view that belief should be proportioned to the evidence, but the position I'm elaborating says that the design plan of saving faith is precisely to take us beyond the evidence, to draw us by desire and a corresponding credulity to supra-evidential belief. So the ‘justification by faith’ position is a non-evidentialist one.

It's also non-fideist if fideism is defined as the suspension of the epistemic for some other mode of justification. Pascal might be taken as an instance of fideism so understood, with his suspension of the epistemic in favour of a prudential justification of religious belief. But the term is contested, and there are other ways of defining fideism than this. Some would define as fideist any position that denies evidentialism. On this definition, Reformed epistemology is a variety of fideism, as would be the view I am defending.

Trying to defend a particular account of fideism is not necessary here. At one extreme there are evidentialist views of the relationship between faith and rationality, and at the other extreme are views that insist on an epistemic gap between faith and rationality. In between are views that deny evidentialism but insist on some other source of epistemic connection between faith and rationality. My view, along with Reformed epistemology, belongs in the middle here,Footnote ¹⁶ because I think the virtue of saving faith is designed by God to be truth-tracking. Recall the way saving faith functions in bringing the child to believe that God sent Jesus to save the world. Saving faith does not provide her evidence, but neither is it epistemically neutral. So, the antecedent desires that are constitutive of saving faith are not evidence for theistic belief; rather, they prompt theistic belief. Sometimes, they prompt theistic belief in circumstances where the evidence itself is compelling. In such cases, theistic belief is overdetermined. But at other times, they prompt theistic belief in circumstances where the evidence is ambiguous, where the theistic beliefs in question are underdetermined by the evidence available to the agent. When this happens, the antecedent desires constitutive of saving faith partly explain how the agent came to the theistic belief in question in the absence of compelling evidence. But because those antecedent desires constitutive of saving faith are designed by God to play just this role, they are epistemically relevant to the probability that the belief is true. So there are cases in which saving faith may epistemically justify theistic belief in the absence of compelling evidence.

In sum, the ‘justification by faith’ view is neither evidentialist nor fideist because it holds that supra-evidential theistic beliefs can nevertheless be epistemically justified. In this respect, it is in agreement with Reformed Epistemology. But although there are structural parallels between Plantinga's Reformed Epistemology and the view developed here, there are important differences. I'll mention four.

First, a virtue is under the control of an agent in a way that a faculty or a process is not. For instance, the ‘internal instigation of the Holy Spirit’ (IIHS) – Plantinga's name for the process whereby specifically Christian beliefs are brought about – is a process over which the agent has little control; it is given to some and withheld from others (apparently). Saving faith, on the other hand, is a virtue, a disposition that an agent may strengthen over time and regenerate if it has been lost. Agents therefore have a great deal of control over what they believe: not ‘directly’, as though they are autonomously free at any one moment to believe anything they choose, but also not merely by ‘indirectly’ putting themselves in environments or circumstances that conduce to the beliefs they are interested in holding. Rather, the primary way in which humans control what they believe about God is by controlling their character. By honouring the dictates of conscience, pondering the good and the beautiful, and steadfastly cultivating inner purity and quiet, a person grows in the virtue of saving faith. And by such growth a person typically comes to believe in ever more robust ways true things about God.

Second, and relatedly, a virtue, because it is under the control of an agent, is subject to increase and decrease over time. And, moreover, this increase and decrease is relatively accessible to consciousness. A person can know when she is growing or flagging in saving faith, for instance, when she notices her conscience has become more or less active. Contrast this with the Reformed Epistemology view that focuses on faculties (sensus divinitatis) and external processes (the IIHS). There is little reason to think the reliability of these faculties or processes might change over time. In any event, if they were to change, one would not be in a position to know it (a feature of Plantinga's strong externalism). Thus there is a misfit between the story told by Reformed Epistemology about the way in which theistic and specifically Christian belief is produced on the one hand, and the actual phenomenological experience of most theists, including Christians, on the other. Reliable faculties like perception and memory bring about beliefs that are relatively consistent in firmness over time; we don't find ourselves confidently believing there is a tree outside the kitchen window one week, but waffling about it the next. But most religious believers do experience swings, often dramatic ones, in the degrees to which they hold various theistic beliefs to be true. On Plantinga's view this is mysterious, but not on my view. Variability is to be explained primarily by the waxing and waning of the virtue of saving faith, and persons have a variety of ways of gauging the strength of the virtue.

Third, unlike Plantinga's Reformed Epistemology, I'm proposing a view according to which the same entity – the virtue of saving faith – confers epistemic justification on both restricted and expanded theistic belief. I don't deny that there is a sensus divinitatis or IIHS. That might all be true. My claim is simply that saving faith is a virtue designed to bring about both kinds of belief – restricted beliefs about the existence of a personal God and expanded beliefs pertaining to the plan God has revealed for humans’ reconciliation and growth in intimacy with God. Of course, this means my view must face challenges that Plantinga can avoid, challenges I have attempted to overcome. The advantage of my view, however, is that it is less ad hoc and better explains the intuition that thoughtful and devoted Muslims may be as justified in holding their religious beliefs as thoughtful and devoted Christians.Footnote ¹⁷

Fourth, on my view saving faith is not a kind of belief but a source of belief. For Plantinga faith is a cognitive state of some kind. Of course it is more than this, but it is at least this. And since what one believes are propositions, ‘to have faith, therefore, is (at least) to believe some propositions’. Which ones? ‘To have faith is to know that and how God has made it possible for us human beings to escape the ravages of sin and be restored to a right relationship with him; it is therefore a knowledge of the main lines of the Christian gospel’ (Plantinga (Reference Plantinga2015), 58–59). Insofar as Plantinga is talking about saving faith, he seems to me to be mistaken. He has confused Christian faith with saving faith. Saving faith is not constituted by belief at all, although it is partly constituted by a disposition to believe, a disposition that I have called holy credulity. So saving faith is not a type of belief. It's a virtue. According to Plantinga, ‘faith is the belief in the great things of the gospel that results from the internal instigation of the Holy Spirit’ (ibid., 62). But on my view, faith is the virtue that (paradigmatically) prompts belief in the great things of the gospel.

Faith, humility, and salvation

Christians claim that persons enter into right relationship with God only by way of saving faith. But why should this be? Why should they not enter into right relationship with God by way of right belief and right action? One way of testing an account of saving faith is by asking whether it sheds light on this important question, and the theory of saving faith I have proposed does.

Consider the scriptural prerequisite to the virtue of saving faith. Earlier, I quoted Jesus’ words instructing his followers to be like children if they want to enter heaven. ‘Truly I tell you, unless you change and become like children, you will never enter the kingdom of heaven’ (Matthew 18:3). And I suggested it is their faith – their holy curiosity, holy concern, and holy credulity – that makes children so pleasing to God and thus fit for relationship with God. In the next verse, Jesus reveals the path to this virtue of childlike faith. ‘Whoever becomes humble like this child is the greatest in the kingdom of heaven’ (Matthew 18:4). Humility is the scriptural prerequisite to saving faith. The two are so closely linked in scripture as to be inseparable (which is not to say indistinguishable). There are no biblical instances of humble persons who are said to be faithless, nor are there biblical instances of humble persons who are unacceptable in God's sight. ‘God opposes the proud, but gives grace to the humble’ (James 4:6 and 1 Peter 5:5, citing Proverbs 3:34). There are, by contrast, many biblical instances of right-believing and right-acting people who are deemed unacceptable to God.

The Christian scriptures report at least three occasions on which Jesus connects degree of faith to degree of humility (Matthew 8:5–13, Matthew 15:21–28, John 5:44). And the Fathers and Doctors of the church followed suit in connecting humility and faith. St Augustine, responsible more than any other Church Father for the elevation of humility as a central Christian virtue, depicts in the Confessions how his own lack of humility prevented him from embracing the Christian faith: ‘To possess my God, the humble Jesus, I was not yet humble enough’ (Augustine (Reference Chadwick1991), 128). Aquinas also sees a crucial role for humility in relation to faith: ‘Certain virtues may be said to precede faith accidentally, in so far as they remove obstacles to belief. Thus . . . humility removes pride, whereby a man refuses to submit himself to the truth of faith’ (Aquinas (Reference Aquinas1991), 1189). The Reformers, too, connect humility with faith. Calvin claims ‘there is no access to salvation unless all pride is laid aside and true humility embraced’ (Calvin (Reference Calvin and Beveridge2016), 331), and Luther, the great champion of justification by faith alone, also wrote, ‘humility alone saves’ (Luther (Reference Luther and Bouman1974), 237).

Thinking through the connection between humility and saving faith should help us understand why it is faith, rather than belief or action, that allows us to enter into right relationship with God. But what is humility in the first place? There are many accounts of humility, some better than others. The most prevalent contemporary view is that humility is a cognitive disposition to accurately estimate one's worth, accomplishment, and skills, or at least to avoid overestimation of those things.Footnote ¹⁸ From a biblical perspective, this is surely wrong: children are rarely so disposed.

Another influential view, the so-called low-concern view, asserts that humility is ‘an unusually low concern for status coordinated with an intense concern for some apparent good’ (Roberts & Wood (Reference Roberts and Wood2007), 241). This is closer to a biblical view, since it treats humility as a feature of one's affective life. But the definition may be too negative. There appear to be ways of meeting the low-concern requirement apart from having humility. For example, obsessives tend to lack self-concern due to their intense concern for some apparent good, but they aren't thereby humble. Certainly a kind of self-forgetfulness is typical of the humble, but it appears to be a symptom of something more essential.

A third view seems closer to the mark still. According to this view, humility is ‘proper attentiveness to, and owning of, one's . . . limitations’ (Whitcomb et al. (Reference Whitcomb, Battaly, Baehr and Howard-Snyder2015), 12).Footnote ¹⁹ On this view, the humble person is someone who is aware of her limitations (though she need not be exceptionally accurate in appraising them) and who exhibits a range of cognitive, behavioural, motivational, and affective responses: in appropriate contexts she calls her limitations to mind, accepts them, admits them, seeks to overcome them when possible and appropriately compensate for them when not, and feels dismay and regret about them. This is close to the biblical view, but still wide of the mark. What seems right about the view is that it locates humility at the point of response to limitations; the humble person is indeed the one who appropriately responds to her limitations. But from a biblical perspective the affective response adumbrated by the limitations-owning account of humility is insufficient. For there are some limitations – indeed those most fundamental to us – about which human persons should experience neither dismay nor regret.

There are some limitations – for example creaturely neediness – that persons must learn to rest in and even embrace if they are to be humble. This is why scripture maintains the child as the paradigm of humility, for the child is neither regretful nor dismayed about his neediness or his limitations. Rather, his neediness becomes the occasion for his reliance upon and delight in his parents’ care. So at ease are children with their neediness that parents must train them out of it, must teach their children to come to experience regret and dismay about their limitations. But cosmically we are forever children, for whom regret and dismay are not appropriate responses to our deepest dependencies. Satan was deeply dismayed by his dependence, and this dismay was, in fact, the root of his pride. So on my view (Dunnington (Reference Dunnington2016) ), humility is something like an embrace of our radical dependence, a view of humility supported in the Christian tradition by the depiction of Jesus in the ‘Christ hymn’ of Philippians 2.

As such, true humility is the seed of saving faith. Saving faith just is, to quote Psalm 10, ‘the desire of the humble’. To recognize our creatureliness, our inescapable neediness and fragility, to acknowledge and honour what Schleiermacher called the sense of absolute dependence (Schleiermacher (Reference Schleiermacher, Mackintosh and Stewart1963), 8), and to do so without resentment, regret, frustration, or dismay, automatically orients us beyond ourselves to the source upon which our lives depend. Such a posture fosters a desire for God, just as the hungry infant desires his mother's breast. And such is the posture that befits human persons for relationship with God. Relationship with God is the path to salvation, and yearning for such relationship is a sufficient condition for its obtaining: ‘When you search for me, you will find me; if you seek me with all your heart’ (Jeremiah 29:13). And what could possibly prevent a person from seeking? Pride. So long as pride is resisted by humility, the desire for God wells up within, constituted by holy curiosity and holy concern, expressed through holy credulity. And this, as I've contended, is saving faith. Saving faith is pleasing to God because it is reaching towards God from a position of appropriate creaturely humility. Whereas belief or action is compatible with pride and a refusal of relationship with God, saving faith is not. Neither right belief nor right action can guarantee right posture for relationship with God; saving faith can.

Of course many Christians (and probably Jews and Muslims) think of faith as a species of belief. But I have argued that the focus on belief should be secondary. It is really faith that God wants from us, but faith so commonly prompts belief that the two can easily be conflated. They are distinct, however, and this is good news to those who find it hard to believe.

This brings me to one final objection. If what matters for salvation is saving faith, and if saving faith does not require belief, why think belief matters at all? Why spend time analysing theistic claims, weighing their evidences, and defending them? Why do theology? Why catechize, preach, or evangelize? Especially considering how much violence and oppression has been wrought in the name of theistic beliefs, why not dispense with ‘propositional faith’ as far as possible? In other words, the objection runs, the position I'm developing here ends up making the actual contents of specific religions otiose.

I don't think so, but there is something right about the objection. Having the virtue of saving faith really is more valuable than possessing the right theistic beliefs. Christian scripture is clear about this. Even the demons believe (James 2:19). In fact, the demons presumably believe many more true propositions (and fewer false ones) about God than any human being except Jesus. The idea that what God wants from us most is proper belief is surely confused.

Right theistic belief is still important, however, even if it has little to do with whether or not we are found acceptable by God. If, as Jews, Christians, and Muslims believe, union with God is the human telos, then there is no more intellectually significant endeavour than the attempt to understand who God is and how to be rightly related to God. Take the relationship between a parent and a child. Normally, what the child believes about the parent is of little consequence to whether or not the parent loves the child. The child is loved unconditionally by the parent. But what the child believes about the parent is crucial for the child's ability to enter into that love and to be nurtured by it. If the child cannot really believe in the parent's love, or if the child is tragically confused about the nature of that love, the child cannot flourish.

The same is true in our relationship with God. And we human beings, living as we do after sin, live in a cosmic broken home. We are the products of abuse, estrangement, betrayal, deceit, envy, and hatred. And, like children of dysfunction, we perpetuate it. Theology, the attempt to understand the nature of God's love for us, is the most important exercise of human reason because it is the attempt, for the sake of our own future flourishing, to expose patterns of dysfunction and tragedy that threaten to rend us from God. God gives us revelation, Jews, Christians, and Muslims believe, precisely to rescue us from the kinds of distorted, confused, and self-deceptive ways of understanding God that we inevitably incline towards without God's gracious intervention. The contents of specific religions are therefore of urgent importance. Their differences matter, precisely because they are differences about the nature of God's love for us and about the best way into that love. We cannot always understand how they are to be interpreted as differences about these matters, but unless they are, they are not differences that matter. Indeed, the task of theology is to show how specific doctrines allow us to more perfectly display that God loves us and that our destiny is union with God.Footnote ²⁰