Faith and trusting testimony

The relationship between faith and trusting God's testimony can be found in both historical and recent literature. Typically, such trust is a matter of taking someone's word for it, that is, of believing someone's testimony, and can have as its object either a human or divine speaker. For instance, when exploring the nature of believing another person, G. E. M. Anscombe remarked: ‘At one time there was the following way of speaking: faith was distinguished as human and divine. Human faith was believing a mere human being; divine faith was believing God’ (Anscombe (Reference Anscombe and Delaney1979), 142). Other recent philosophers have made similar remarks. Lara Buchak says that this kind of faith is ‘belief on the basis of testimony: one has faith in a person if one relies on that person's testimony in forming beliefs’ (Buchak (Reference Buchak, Abraham and Aquino2017b), 48).

Other authors who have focused on this kind of faith have been more committed to giving accounts of Christian faith, and so have been more attentive to believing divine testimony. For example, Linda Zagzebski has claimed that ‘religious faith is believing God’ (Zagzebski (Reference Zagzebski2012), 190), and similarly, Anthony Kenny writes that religious faith is ‘belief that a proposition is true because God has revealed it’ (Kenny (Reference Kenny1992), 50). There are other recent authors, too, who have coalesced in defining religious faith as believing God, or maintaining that Christian faith requires believing God's testimony,Footnote ⁵ and this concept also has an impressive historical pedigree in both philosophy and theology.Footnote ⁶

So, what is this kind of faith and how can we give an account of it? A useful place to begin is with testimony itself. Testimony is the primary linguistic means that people have of sharing their knowledge with other people. People tell one another propositions that they themselves (hopefully) take to be knowledge, offering hearers the opportunity to acquire a piece of knowledge on the basis of their testimony. As Elizabeth Fricker puts it:

This gives an idea of the speaker side of a testimonial exchange. But if faith is, or involves, a hearer trusting a speaker for her testimony, then we need to understand what the hearer's response is to the testimony offered to her.

Hearers may respond to testimony in several ways. Predominantly though, they will either trust or distrust what the speaker says. Believing a speaker is sufficient for trusting her, and it is only through forming a belief that one may acquire knowledge from a speaker. But other responses may also be sufficient for trust. One may be unsure about the veracity of the testimony, or about the reliability of the speaker, and so cognitively suspend judgment concerning the truth of the testimony. Nevertheless, one may accept what the speaker says by assenting to the testimony and committing to act as though what the speaker has said is true (Cohen (Reference Cohen1992) ). Some accounts of trust have discussed how trust can be nondoxastic in this kind of way (Frost-Arnold (Reference Frost-Arnold2014) ). In order to leave it an open question whether hearers can trust a speaker by either believing her or accepting her testimony in a nondoxastic sense, I will simply say that hearers can trust speakers for their testimony. Of course, hearers can also distrust testimony by disbelieving it, or, perhaps, by refusing to act as though what one has been told is true.

It is the trusting response on behalf of the hearer to the speaker's testimony that has been identified with faith. This is particularly evident in the quotations from Anscombe and Buchak in the foregoing. And quite possibly faith, in a somewhat mundane sense, can be an instance of trusting another person's testimony. As Daniel Howard-Snyder claims, ‘in so far as this use of “faith” as believing someone involves faith at all, it involves putting one's faith in x, trusting x, to tell the truth about p’ (Howard-Snyder (Reference Howard-Snyder2016), 146). And yet we might feel that religious faith requires more than one-off instances of trusting God's testimony. The reason why is because religious faith, particularly in the Christian tradition, is often taken to involve a relationship between an individual person and God. Someone can trust a particular person on a single piece of testimony with any stranger in the street with whom she has no personal relationship. What seems required for religious faith, though, is for one to have a general disposition to trust God, which involves a disposition to trust what one perceives to be God's testimony, and that partly in virtue of such a disposition one will be in a trusting relationship.Footnote ⁷ From such a disposition, one will trust God for particular instances of perceived testimony.

So, for religious faith, we require an account of faith that can make sense of this richer notion of a relationship of trust. A way into this issue is from recent work on the nature of faith, which has provided two plausible candidates for faith as an attitude with the relevant dispositions towards trust and reliance. First, there is propositional faith, or faith that p, as when someone has faith that her team will win the game, or faith that God exists. Second, there is relational faith, or faith in X, where X is some object, like a government, or some person, like a friend.

Propositional faith is widely thought to require holding a cognitive attitude towards a proposition p of either belief or nondoxastic assent, such as acceptance or assumption. This can be characterized in terms of cognitively taking p to be true by, for instance, representing p as true, and being disposed to act as though p.Footnote ⁸ In addition, propositional faith is also thought to require one to be in favour of the truth of p, by, for example, desiring that p, or believing p to be good or desirable (Schellenberg (Reference Schellenberg2005), 108; Audi (Reference Audi2008), 92; Howard-Snyder (Reference Howard-Snyder2013), 367). Thus construed, propositional faith is an attitude or mental state that a person has, concerning the truth of some proposition.

Relational faith is often thought to be constituted by trust (Swinburne (Reference Swinburne2005), 142–147; Audi (Reference Audi2011), 55–56) or a disposition to rely on something (Howard-Snyder (Reference Howard-Snyder2017), 56). For example, faith in one's spouse will dispose one to trust or rely on one's spouse. Relational faith also appears to require taking on a cognitive attitude towards the truth of certain propositions. For instance, when someone has faith in a friend to tell the truth, she accepts or believes that her friend will tell the truth. So, faith in X can be seen as a complex attitude in which one holds a cognitive attitude toward the truth of p, while also requiring the dispositions towards trust or reliance.

Religious faith, in the sense of trusting God and being disposed to trust God's testimony, may be understood as a complex of propositional and relational faith. For instance, as when a person S has faith that God is trustworthy and what God tells S is true, and when S has faith in God to be trustworthy and to speak truly. Such a complex of faith attitudes will dispose a person in various ways to trust God, including trusting God for his testimony, to rely on it in thought and decision-making, and to act as though the perceived testimony is true.Footnote ⁹

Note that there is a distinction here between attitudes with dispositions and actions exhibiting those dispositions. This can be put in terms of having faith in God and putting faith in God. Consider this in relation to faith in God's testimony. Someone can have faith in God to tell the truth in virtue of a disposition to rely on God's testimony. But she can also put faith in God's testimony by trusting what she perceives God to have said. Having faith in X provides one with a complex mental judgment involving a psychological disposition to trust or rely on X. Even though this state motivates someone to action, it is not itself an action. But putting faith in X is to exhibit the dispositions given by one's faith, and when one's faith concerns the reliability of someone as a speaker, then these dispositions are exhibited by trusting what the person says. This can be done through belief, acceptance, and the commitment to act as though the testimony is true.

When someone's faith in another is exhibited through action – when she puts her faith in someone – then we can say that she performs an act of faith in the sense that she acts on her faith. When she acts on her faith, she performs the actions that she is disposed towards in virtue of her faith. Such actions are exhibitions of her faith in much the same way as with other kinds of traits or attitudes. For instance, an intellectually generous researcher will be disposed to share her findings as widely as possible, both to academics and members of the public (cf. Roberts & Wood (Reference Roberts and Wood2007), 286–304). When she performs these actions, we can say that she performs acts of intellectual generosity. An intellectually closed-minded student will be disposed to confirmation bias, and hence to look only for evidence that supports his existing views. When he performs these actions, we can say that he performs acts of intellectual closed-mindedness.

Of course, any of these actions can be performed without having the corresponding trait or attitude. Someone can share research findings beyond academia without being intellectually generous, and someone can search only for confirmatory evidence for a belief without being intellectually closed-minded. And similar claims can be made about faith. Someone could perform an act of faith without having faith in X, or faith that p. But since we're interested in a deep kind of relational faith that disposes someone to trust God – a kind required by religious, and particularly Christian faith – then I'm focussing on acts of faith in the sense that they express or exhibit a person's existing faith, rather than acts distinct from it.

We can see the distinction between faith and acts exhibiting one's faith in both general and religious cases. For instance, when Jane has faith that her friend will pick her up from the airport on time, she is disposed to rely on her friend in this respect, and can act in this faith by relying on her to collect her rather than calling a taxi. When Mark has faith in the Grand Canyon Skywalk to hold his weight, he is disposed to rely on it in this respect, and he can act in his faith by walking out onto it. And when Kate has faith in God to speak truly, and faith that God is a reliable and credible testifier, then she is disposed to defer to, trust, and rely on what she takes to be God's testimony. Kate can then perform an act of faith that expresses or exhibits her faith in God in these respects by trusting the purported testimony, deferring to it, committing to live as though it is true, etc.

Now, although the model of faith discussed thus far began with a comparison between trusting human testimony and trusting God's testimony, there is an important disanalogy between the two that moves us towards a distinctive account of religious faith. For, while there is generally no problem identifying a speaker in typical cases of human testimony, it is not straightforward identifying a purported instance of God's testimony as God's testimony. This is generally the case because testimony from human speakers comes to us immediately and directly: speakers just assert to us, and we either trust or don't trust them. Divine testimony, however, if indeed it is spoken, typically comes to us indirectly, mediated via various sources, including scripture, prophets, sermons, tradition, and perhaps even major life events or advice from friends. There is a separate task for the believer of divine testimony, over the believer in human testimony, namely, to adjudicate whether a piece of purported divine testimony is indeed God's testimony. Subsequent to that, a person may simply believe the testimony itself, as with general cases of testimony.

Observation of this disanalogy prompts an account of religious faith on three levels. First, one trusts God in the sense described earlier in this section by having faith in God in various respects, and faith that relevant propositions concerning God are the case. Second, such trust will dispose someone to trust God has spoken:

Someone with religious faith will be disposed in this way because she will have adopted a religious world-view in which there is a God who ‘speaks’ by various means, and will trust this God to speak truly. Hence, she will exhibit these dispositions by trusting God has spoken (TGS). But third, she will also be disposed to trust God's testimony:

That is, once one believes God has told her something, she will also have the disposition to trust what she takes herself to have been told. And trusting God has spoken (TGS) and trusting this purported divine testimony (TGT), will be to perform acts of one's faith – actions which exhibit one's religious faith in God. Someone who has faith in this three-tiered sense I will say has faith as trusting God (FTG).

From a theistic perspective, then, we can see faith as someone taking herself to be ‘hearing’ and trusting God's testimony, where God is the ‘speaker’ of the testimony.Footnote ¹⁰ However, FTG does not specify what the source of the mediated testimony must be: the actual mode of the purported divine testimony can vary to a large extent (e.g. scripture, prophets, visions, the Holy Spirit, quasi-perceptual sensus divinitatis (Plantinga (Reference Plantinga2000) ), etc.). What it therefore is for God to perform acts of testimony can manifest in various ways, and it is up to the individual account to fill in the details. However, what I have to say about intellectual humility will not be affected by the way one construes what it is for God to speak, nor what it is to hear from God.

Nevertheless, it may be insisted that FTG requires one to ‘go beyond the evidence’ in some sense (see Malcolm (Reference Malcolmforthcoming) ). The reason for this is because at the stage of TGS, it is unlikely that anyone could acquire the relevant evidence to make such a move epistemically justifiable. Sometimes this idea is built into the very concept of faith itself, albeit not always in contexts of theistic faith. For instance, it has been argued that faith involves, implies, or even requires:

I won't take a stand on which of these accounts is correct. But I accept, first, that the concept of propositional faith does imply that one is in non-ideal epistemic circumstances, although these circumstances do not need to require that one lacks epistemic justification.Footnote ¹¹ And second, in the sense described here, faith will probably require one to trust that God has spoken without being in ideal epistemic circumstances for believing that God has spoken. Nevertheless, trusting in this way could yet be satisfied through nondoxastic acceptance rather than belief if one is not justified in believing God has spoken. This issue, and related concerns surrounding one's appropriate levels of confidence, are revisited in later sections. But with this account of faith (FTG) in hand, I want to move now to consider intellectual humility, and how it must accompany someone's faith if her FTG is to be virtuous.

Intellectual humility, servility, and arrogance

Recent work on intellectual humility has focused on the role of humility in giving someone a proper appreciation of both her intellectual strengths and weaknesses. For instance, according to Kidd (Reference Kidd and Baehr2015, 54), ‘humility is a virtue for the management of confidence’, and in particular, he focuses on the role of humility in regulating someone's proper confidence and recognition of her intellectual capacities. Take, for example, a student coming to the end of her studies at high school, and deciding what further education she would like to pursue. If she has virtuous intellectual humility, she will have the correct level of confidence in her capacities with respect to, for example, processing numbers, reading and writing, arguing and debating, computers and programming, compassion and tolerance, timekeeping, manual labouring, and other cognitive, emotional, and physical capacities. This will give her the dispositions needed to select wisely the best future study for her, and the most appropriate career-path given her intellectual abilities. If she has excellent intellectual capability with reading and writing, but limited abilities with mathematics, then, her intellectual humility will give her the appropriate level of confidence in these capacities.

Remarks along a similar line can be found in two further accounts (Whitcomb et al. (Reference Whitcomb, Battaly, Baehr and Howard-Snyder2017); Tanesini (Reference Tanesini2018) ), which understand intellectual humility as the acknowledgement of one's intellectual limitations, being properly attentive to them, and working them appropriately into one's practical reasoning. According to Whitcomb et al., intellectual limitations include

Whitcomb et al. then go on to define intellectual humility as ‘having the right stance toward one's intellectual limitations’ where this is a matter of being ‘appropriately attentive to them and to own them’ (ibid.). The colloquial expression of ‘owning’ something is a matter of embracing or endorsing that thing. Suppose, for instance, that someone has been recently promoted in her job and is intimidated by the prospect of the added responsibility. One could ‘own’ the new job role by acting as though one feels ready to take it on, talking confidently about it, making assertive decisions with one's new authority, and so on. In this way, the person ‘owns’ the new job by embracing and endorsing it.

Consider the same concept when someone ‘owns’ an intellectual limitation. Suppose someone has poor eyesight but spends a fair amount of time driving. This person's intellectual limitations include an unreliable cognitive process – poor eyesight. What does it mean for someone to be ‘appropriately attentive’ to the fact of one's poor eyesight, and to ‘own’ it? Whitcomb et al. identify four characteristic dispositions that attentiveness and owning will yield concerning one's intellectual limitations:

For our driver with poor eyesight, then, he will be disposed to (1) believe he has poor eyesight and believe that this can lead to poor or dangerous driving; (2) admit that he has poor eyesight; (3) care that he has poor eyesight and take this seriously; and (4) feel dismay but not hostility towards his poor eyesight. These dispositions will manifest themselves in certain behaviours for the driver. He will be careful to make sure he always wears his glasses when driving, and to drive slowly to offset any potential risks caused by his condition. He will also tell others about his eyesight when asked, or when required to admit that he has the condition, for instance, by declaring it on his driver's licence. This is what it would mean to be attentive to, and to ‘own’ one's poor eyesight, and hence to exhibit intellectual humility concerning it.

It should be noted that both the accounts by Kidd and Whitcomb et al. (as well as Tanesini) focus on deficits in one's own intellectual limitations. However, we could broaden this to encompass deficits in humans generally. All humans have an extremely limited epistemic vantage point, and lack the ability to know of all of the different experiences and points of view of other humans in other cultures at different times in history. Having intellectual humility requires taking these more general limitations into account, as well as one's own particular limitations. These general limitations are particularly relevant at points for religious epistemology.

With this in mind, the way that someone is required to form her beliefs in light of intellectual humility depends on the evidence at her disposal, and the proper conclusions she draws concerning her intellectual strengths and limitations. For instance, if someone has poor navigational abilities, humility requires that she believe she has poor navigational abilities, but if they are excellent, then she is required by humility to believe they are excellent. These dispositions would follow on each of the accounts outlined thus far.

Another clear example of how intellectual humility regulates our beliefs is in the social sphere, with testimony providing the most obvious example. Whitcomb et al. argue that intellectual humility ‘increases a person's propensity to defer to others who don't have her intellectual limitations’ (ibid., 522). Although such a propensity will require one to believe someone else who lacks one's own limitations, it will not require this for all cases. For instance, if I know I have poor knowledge of current affairs and speak to a friend about current affairs who I know is highly knowledgeable about them, then my intellectual humility will give me a propensity to defer to her testimony. And if I do defer to it, then that is an act exhibiting my humility. But suppose we are talking about a particular issue, say, the sale of arms from western countries to Saudi Arabia. This is an issue I know a little about having watched a reliable documentary on it. She tells me something on this issue that I disagree with, but I have good reason to suspect that she is just speculating and lacks any credible evidence to back up her claims. Well, in this case it is such that I needn't defer to her even given my general limitations in the area of public affairs. It is entirely proper of me to resist deference to her consistently with a correct appreciation of my confidence in my intellectual capacities, and while acknowledging and ‘owning’ my intellectual limitations. This is because I have superior epistemic capacity in this case, and have reasons for doubting the veracity of her testimony.

This example also provides us with a way of seeing how humility as a virtue is a mean between the vices of servility and arrogance. For, if I had no reason to doubt the trustworthiness of my friend, nor the veracity of her testimony, and had no knowledge of the issue for myself, then humility would require me to defer to her. But if I simply took myself to have superior knowledge of political affairs (which I lack) and refused to defer to her testimony on these grounds, then this would indicate that I lack humility, and instead have intellectual arrogance, at least with respect to this issue in political affairs. This would particularly be the case if I systematically failed to appreciate my limitations and so held onto beliefs about political matters when I should have let go of them in the face of people with greater knowledge than myself. This idea is captured in the account by Whitcomb et al. who say that ‘the paradigmatically arrogant person is often oblivious to his limitations; they don't show up on his radar’ (ibid., 516). When I don't see my own limitations, particularly with respect to the strengths of others, then I will tend towards arrogance rather than humility.

Just as we can use the deficit of attentiveness to limitations to understand intellectual arrogance, we can see the over-attentiveness to limitations as a way of understanding intellectual servility. The arrogant person is hardly aware of her own intellectual limitations, but ‘[t]he paradigmatically servile person . . . hardly sees anything else; his radar is perpetually peppered with his limitations’ (ibid., 517). With servility, someone would be overly keen to defer to the testimony of others, even those who have significant intellectual limitations beyond our own. So, suppose I am told things concerning the sale of weapons to Saudi Arabia by someone whom I judge correctly to lack knowledge on the subject when I myself have knowledge on the subject. Well, if I defer to the speaker's testimony because I think to myself, ‘I must be wrong about my views because I'm too stupid to understand public affairs’, then this indicates servility, rather than humility: I am servile to my limitations. In this case, humility would require that I give proper appreciation to my strengths and limitations, rather than over-appreciation to my limitations, and resist deference to my friend's testimony.

With this picture of the nature of intellectual humility, servility, and arrogance in hand, we are in a position to determine what humility would require of someone who has faith in the sense of FTG. As we'll see, FTG must be accompanied by humility if it is to be virtuous.

Humble, servile, and arrogant faith

Given what we have said about intellectual humility, one might assume that the humble response to God's testimony is simply to trust it. After all, on a traditional theistic view, God is all knowing and cannot lie, whereas humans are substantially cognitively limited in ways that God is not. But this view faces a problem, which is evident in the account of faith developed earlier: trusting God's testimony (TGT) requires one first to trust that God has spoken (TGS). It seems reasonable that trusting God's testimony is required by humility when one has clear evidence that God has spoken, but it's not so clear what humility requires concerning TGS. John Lamont phrases the problem in this way:

Lamont proposes that the reason for the difficulty in identifying divine speaking is the means of the speaking. If we take God to speak by appropriating various communicative acts, including sermons, written scripture, and songs, then we must determine whether a purported piece of divine speaking is in fact divine speaking in a way we do not with human speaking.

Now, suppose we have someone called Sarah who has faith in the sense of FTG. Sarah has recently begun engaging with the Christian faith, forming an attitude of trust towards God. This has been cultivated within a Christian community of others who also trust God, and where it is common to be reliant on God by, for instance, petitioning God in prayer, committing to live as though God's plan for salvation is true, and basing one's decisions on Christian ethical values. During her engagement with Christianity, Sarah has been attending Christian church services where she sings hymns, hears sermons, and reads scripture. Sarah has had the impression that God has been speaking to her through these religious activities. Among the things she takes God to have told her are that he is a loving God, that she will continue to live spiritually after her bodily death, and that God was literally born into human history in the life of Jesus Christ. These are not propositions that Sarah had previously accepted, endorsed, or believed. However, she has responded to this perceived divine revelation by acting on the supposition that their source is God's own speech to her, trusting God for this testimony, and committing herself to living as though such propositions are true.

Does Sarah's response exhibit humility? Or does it exhibit servility or arrogance? This depends on two factors:

1. (1) the evidence E she has for the proposition p that God has testified to her; and

2. (2) the cognitive attitude Sarah forms concerning p given E.

Suppose that, concerning (1), her evidence for God speaking or testifying to her is limited and not sufficient to warrant belief. She therefore has a limitation: a gap in her evidence required for justified belief or knowledge. Now, concerning (2), suppose that Sarah believes that God has spoken to her, and that her confidence in this belief is high such that she takes it to constitute knowledge – she is confident that she knows that God has spoken to her. Well, in this kind of case, we could then charge Sarah with either intellectual arrogance or servility, depending on what her reasons are for believing as she does. Let's see why.

Suppose that Sarah takes herself to have greater epistemic abilities than she does have for discerning whether or not God is speaking to her. She might take herself to hear a clear voice from God when she prays, which in fact she does not hear. In fact, this belief is formed due simply to a desire to hear God's voice, and so she has convinced herself that she can hear him. What is evident is that she now has another limitation: an unreliable cognitive process. So, Sarah has two important epistemic limitations: insufficient evidence and an unreliable belief-forming mechanism. Now, in this case she can be accused of intellectual arrogance because she is being under-attentive to her limitations (on Whitcomb et al.’s (Reference Whitcomb, Battaly, Baehr and Howard-Snyder2017) account; cf. Tanesini (Reference Tanesini2018) ) and overly confident with respect to her intellectual capacities (on Kidd's (Reference Kidd and Baehr2015) account). And because of this, Sarah's response to what she perceives to be God's testimony is accompanied by arrogance – she is arrogant with respect to her limitations – and hence her FTG is a vice.

Consider Sarah's situation under a different response. Suppose that Sarah is initially unsure about whether God has spoken to her. She recognizes, correctly, that she has a limited epistemic vantage point and is a new Christian and has new intuitions. Nevertheless, she also acknowledges, also correctly, that she has perfectly good judgment of whether to trust her senses, and so initially resists believing that God has spoken to her. However, because of the pressures of being in a new religious environment where most people claim to hear from God, she begins to assume that her epistemic limitations – being a new Christian and having a limited vantage point – must be preventing her from hearing from God. She thinks to herself, ‘I must have this wrong. I am just a limited human being and I should just believe that God is speaking to me.’ Now, certainly Sarah has limitations. But here she is being over-attentive to them, and under confident with respect to her intellectual strengths. She actually possesses the correct level of discernment, but has ceased to be disposed to follow it in this religious context. So, in this case, Sarah's response to what she perceives to be God's testimony is accompanied by servility – she is being servile to her limitations – and hence her FTG is a vice.

So, we have seen where Sarah's faith can be arrogant or servile. But despite this, it can also be humble. However, in order to see this, we must change the setup of the example. Note that in the above example, her response exhibits an intellectual vice because, with respect to the proposition that God has testified to her, Sarah (1) lacks evidence to warrant belief, but (2) believes anyway, and even presumes her belief to constitute knowledge. We can change the example by, first, assuming that Sarah has evidence sufficient to warrant belief and believes, or lacks evidence sufficient to believe, but does not believe. Let's consider these options in order.

It might be assumed that, given our intellectual limitations – our limited epistemic vantage point, inability to discern human from divine testimony, etc. – it is not possible for us to have evidence sufficient to justify the belief that God has spoken. But we could deny this. Even John Locke, who holds a similar view of faith to FTG,Footnote ¹² and who was highly sceptical of people claiming to have received divine revelation, thought that it is possible to justify one's belief that one has been spoken to by God. Locke's view on the matter was that confirmation of the revelation by clear miracles would be sufficient to justify belief that the purported revelation was genuine (see Locke (Reference Locke and Pringle-Pattison1924), ch. XIX, §15). This will be the case regardless of the intellectual limitations of the recipient of the revelation. The example Locke uses is Moses, who was spoken to by God through a burning bush, and whose staff repeatedly turned into a serpent. Perhaps Moses has clear evidence that he was spoken to by God via clear miracles.

In a modern context, Wolterstorff (Reference Wolterstorff1995, 261–280) considers some instances in which people can be entitled to believe that God has spoken to them through confirmation of the experiences of others. In particular, where one person is considering whether God has spoken to her on a particular matter, and another person tells her, without knowing her situation, that God has spoken to her about the very same thing. Under some conditions, such interactions could plausibly constitute evidence sufficient to permit belief that God has spoken.

These responses from Locke and Wolterstorff face two problems. First, in line with some of the views about faith mentioned earlier, if Sarah has clear evidence that God has spoken to her then she wouldn't need faith that he has, and hence Sarah wouldn't exhibit FTG, which is the main concern of this article. This point is important, and I want to offer two possible ways of responding to it.

First, we could say that even if one can be justified in believing God has spoken, this doesn't imply that the person can have clear evidence or certainty that he has, and hence her evidence would be sub-optimal, even if it isn't sub-par. Her belief that God has spoken may still require some small faith step, even if this is minimal, particularly given how amazing it would be that God has spoken to her. Believing this will still require some measure of faith, even if only a small amount. Moreover, recall that in the earlier iteration, Sarah not only believes, but has an extremely high level of confidence that her belief is justified and true – so high that she takes herself to have knowledge. But in the case we have been considering, she could have just enough evidence to permit belief, and hence to be rational, but not have enough evidence to require belief, and hence to be fully rational. Given this, her confidence in her belief could be far more tentative. She could be unsure about whether her belief is justified and true, and hence be disposed to caution when it comes to acting on the basis that it is true. In this instance, Sarah hardly takes her evidence to be clear or conclusive. There is plenty of room here for being in a weak epistemic position (Alston (Reference Alston, Jordan and Howard-Snyder1996), 12), not taking her evidence to support certainty (Buchak (Reference Buchak, Callahan and O'Connor2014), 53), faith-venturing inadequately supported by her evidence (Bishop (Reference Bishop2007), 106), and faith being sub-optimal (Howard-Snyder (Reference Howard-Snyder2013), 370).

A second (and possibly complementary) way of addressing this issue is by claiming that her belief is fully justified in relation to some contexts but not others, and hence her faith is in a weak epistemic position with respect to some contexts but not all. Note that it could be reasonable for Sarah to believe certain propositions given her Christian context. Suppose she believes that God has told her that he is Triune. We might suppose that in Sarah's community, there is widespread consensus that God is Triune, and that this belief is adopted and endorsed among all of the other members of her religious community. So, she could be warranted in believing this proposition, partly because of the consensus in her community that it is true, but only in relation to her Christian context: it is the norm to believe it. Outside that context her belief may not be warranted. And hence, what we could propose is that Sarah's faith ‘goes beyond the evidence’ in some way with respect to the wider context but not with respect to her Christian context.

Now, even if one, or perhaps both, of these responses is tenable, a second problem arises that human experiences of miracles and divine interaction are notoriously controversial. After all, even occasions where one person feels that her circumstances lend strong evidence to the claim that God has testified to her, there are still issues concerning interpretation of the ‘evidence’, religious disagreement from other traditions, and biases and partialities to take into account. Nevertheless, we should permit that in principle people can have the experiences required to justify their belief that God has spoken, even if such experiences seem rare or unlikely.

So, suppose that, somehow, Sarah has evidence sufficient to warrant belief that God has spoken to her – albeit weak and only justifying belief partially – and she believes accordingly. Under these circumstances, she can exhibit intellectual humility. Sarah can acknowledge her limitations, her gaps in knowledge, her limited time spent as a Christian. She can even fairly and correctly identify that she has epistemic strengths that lead her to believe, perhaps rightly, that hearing God speaking is unlikely or rare. And yet when presented with good evidence that God has spoken, and without being overly attentive to her limitations or under-attentive to her strengths, Sarah can form the belief that God has spoken. Intellectual humility certainly permits this belief, if it does not require it, depending on the strength of the evidence.

But she may also be required by humility to have the appropriate level of confidence in her belief. She must acknowledge her limitations, including the limitations of her evidence base, and her confidence in her belief vis-à-vis knowledge must be measured appropriately. What seems most consistent with humility is that she ought to hold her belief tentatively, and not presume to know that God has spoken to her. This might dispose her to be particularly sensitive to new evidence disconfirming her belief, and to require clearer evidence to support the belief. She might also be disposed towards caution with respect to acting as though it is true, and in telling others that God has spoken to her, particularly with respect to norms of assertion.

When Sarah's response to what she perceives to be God's testimony is as described here, her FTG is accompanied by humility. She is being humble concerning her strengths and limitations, reflected in the tentativeness of her belief that God has spoken. Given this, her faith can be construed as a virtue, provided it is not subject to other vicious limitations or shortfalls.

Although this kind of situation is possible in principle, it may be more likely that Sarah lacks evidence sufficient to warrant belief that God has spoken. But we might imagine that, rather than believing that God has spoken, as she did when exhibiting arrogance or servility, Sarah does not believe that God has spoken. So, suppose that Sarah has the impression that God is speaking to her through various appropriated means. However, it's not clear to her that he is in fact speaking. She has a fair and reasonable awareness of her limitations. For instance, she is new to this experience and is still trying to discern what to make of it. She appreciates that she has a limited set of experiences and a restricted epistemic vantage point. But part of this assessment of her limitations is also the recognition that she may seriously want to feel that God is speaking to her, and so she needs to temper her confidence with respect to believing that he has, or else she could end up, with false or delusory beliefs. She also has an awareness of her intellectual strengths – she has a good sense of her abilities to discern whether someone is trustworthy, and plenty of life experience, which instructs her on the biases and interests of others. She understands that religious people can often have strong partialities, and is wary of not becoming someone who lacks proper discernment.

So, rather than coming to belief that God has spoken, Sarah suspends judgement concerning this proposition. Moreover, she holds her levels of confidence fairly low, and is disposed to require strong evidence before coming to believe that God has spoken to her. Nevertheless, she both hopes and accepts that he has spoken to her. Since she is tentative about the truth of God speaking to her, such hope and acceptance disposes her to act in measured ways; for instance, by praying to God when she feels that he is communicating to her, doing good actions that may have been commanded by God, such as showing love to her work colleagues, and committing herself to live as though Christ died for her, which she takes God to have told her. She wouldn't be disposed, for instance, to change her life completely by quitting her job because she simply has an impression of God telling her to do this. Her acceptance is cautious because her levels of confidence are measured and restrained.

Sarah's response appears to be paradigmatically humble. She doesn't rush off, believing and having full confidence because she takes herself to have excellent abilities in discerning God's voice, which she lacks, as the arrogant person would. Nor is she overly sensitive to her limitations, and so believes God has spoken because she thinks to herself that she must have her judgements wrong, when she doesn't, as the servile person would. Her response exhibits humility because she ‘owns’ her limitations in a way that appreciates she has them, and so should be open to God speaking to her. But she is also properly confident with respect to her strengths, and so measures her cognitive response to the possibility of God speaking by resisting outright belief, and has a restrained level of confidence. In such circumstances, FTG is accompanied by intellectual humility and hence can be a virtue.

Now, we might agree that this response is one of humility, but is it also one of faith? Doesn't faith require Sarah to believe, rather than merely hope or accept, that God has spoken? Well, as was discussed earlier, there is widespread (though not universal) support for the view that faith does not require belief, and hence, when Sarah trusts that God has spoken (TGS), she could simply act on the supposition that God has spoken. In terms of cognitive attitudes, this can be satisfied by either belief or acceptance, and so Sarah's response, when she has sufficient evidence, can be one of belief, and when she lacks it, it can be one of acceptance. But of course, in order for her response to be one of faith in the sense of FTG, Sarah must also trust what she takes God to have said to her. Again, this can be a nondoxastic response, but will also involve relying on God in certain ways, as Sarah does when, for instance, she commits to living as though the Christian narrative of salvation is true.

In summary, in order that Sarah's FTG be virtuous, it must be accompanied by intellectual humility. This requires her to make the appropriate cognitive and epistemic response to putative divine testimony, which is determined by both Sarah's intellectual strengths and limitations, and the evidence available to her concerning whether or not God is speaking to her. At times, belief and high levels of confidence may be consistent with an intellectually humble response, while in others, a nondoxastic response may be the intellectually humble attitude to take. When she fails to take the appropriately humble response, her faith can be accompanied by the vices of servility or arrogance.