The affective view of faith as trust

In order to avoid the ‘scoundrel's faith’, Swinburne gives prominence to the affective element of goodwill.Footnote ⁴ Certainly, good character is not enough: in order to be religious, faith requires having those purposes that characterize religion in general, or the main goals of religion. According to Swinburne those goals are: (1) worship of and obedience to God (for theistic religions), (2) salvation for oneself, (3) salvation for others (salvation implies everlasting happiness of a kind that is morally appropriate). One cannot have such goals without some cognitive requirements. Religious faith requires then the minimal set of beliefs that are necessary for the believer to behave in order to reach the goals of religion. In particular, since one would not act for a purpose without at least some guiding beliefs, indicating what to do in order to reach the goal, faith requires some means–end beliefs, beliefs that a certain religious way is one of the best options (at least better than some other, and no worse than any other) for the goals of religion. This belief requirement supposes that there are some theoretical conceptions of the way in question, of what is to be done (including morality and religious practices), which Swinburne calls a Creed. To have religious faith supposes then that one believes that a certain Creed is one of the best options to realize the goals of religion. Such a belief is the belief that the Creed in question is more probable than any rival option (be it theistic or non-theistic), but not that it is more probable than not. So it implies only what Swinburne calls weak belief that the Creed is such a best option, but not the corresponding strong belief.

The weak belief that a certain Creed (defined as above) is the best option to achieve the goals of religion with the strong purpose to achieve those goals requires either that one believe that the Creed is true, or at least that one act as if the Creed were true. To act as if one believed a certain content is what Swinburne calls ‘to act on the assumption of’ that content. In order to assume a proposition or a set of propositions, you do not need to believe them, not even weakly, you just have to adopt them as premises in some practical deliberation. A British soldier might thus act on the assumption that a German told him the truth concerning a way out of a bad situation, without believing him, and even while believing the German might betray him, only because he has no better (or no other) option. To act on the assumption that a Creed is true is to have the purpose of achieving the goals of religion, and the belief that the best chance to do so is by doing such and such actions, and that those actions will achieve those goals only if the Creed is true. What the assumption of p requires in terms of belief is that one does not fully disbelieve p, or that one believe that p has a significant (not too small a) probability. As a consequence, to act on the assumption that the Christian Creed is true does not even require that one believe that there is a God, but only that one assume that there is a God, and act on that assumption. One also needs to assume that God will do for us what we want or need. Such an assumption that someone will do what one needs or wants, when the evidence gives some reason to suppose that he may not, and when bad consequences are to be expected if the assumption is false, defines trust.Footnote ⁵ So to act on the assumption of a certain theistic Creed requires trusting God. It does not require that one strongly believe that the Creed is true, or believe that it expresses God's nature and will, or even that there is a God. One only needs to assume those propositions. Since trust is defined by such assumptions, one might say that trust receives a propositional account (trusting God is trusting that there is a God and that He will do so-and-so . . .).Footnote ⁶

Swinburne can then write: ‘a person has Christian faith if he acts on the assumption that there is a God who has the properties which Christians ascribe to him and seeks to do those good actions which the love of God (if there is a God) would lead him to do’ (Swinburne (Reference Swinburne2005), 148). This conception of faith has quite surprising and unusual consequences, most of them drawn by Swinburne himself.

1. 1. Though faith requires strong belief that a certain Creed defining a certain way is one's best chance to achieve the goals of religion, and thus requires assuming that Creed and following that way (since it requires the strong purpose to reach those goals), it only requires weak belief that assuming that Creed and following that way will reach the goal, and it is compatible with strong belief that it will not.

2. 2. Faith is not only compatible with not strongly believing (not believing that it is more probable than not) that the Creed is true, and even that there is a God, but it is also compatible with strongly believing that the Creed is false, and also that there is no God.

It follows then that one may be said to have the Christian faith while strongly believing the Christian Creed is false, and even while strongly believing that a certain Anti-Nicene Creed is true. In fact, one might have the Christian faith while being an atheist. It is only required that one be not a full atheist, an atheist who gives a zero or very small probability to the existence of God.

Those consequences, and above all that of the ‘faithful atheist’, are paradoxical, even for the person who is open to the idea of implicit faith, and to the idea that one may follow the Christian way (or any other religious way) while being personally in doubt or agnostic about the truths that it professes. They are paradoxical both with regard to common construals of ‘religious faith’, and with regard to the use of the word and other connected ones in the Bible. But it is always possible to build a notion, and to admit some discrepancies both with common uses and with some received and traditional use. As I mentioned at the beginning, the word ‘faith’ is taken in multiple ways both inside and outside its applications to religious matters. In addition, different languages do not allow for the same uses and grammatical constructions concerning the term that is considered as an appropriate translation. Swinburne's account does not aim at keeping together all the modern connotations of the term ‘faith’ (by contrast with Howard-Snyder (Reference Howard-Snyder2013)). But he is no more trying to stick to some received notion or definition, building a notion of faith that should apply widely to different religions, even though he tries to show how his notion fits with most uses of the term for faith (pistis) in the New and the Old Testaments.

The two cognitive views of faith, the rationalist and the fideist, shared a common assumption that faith is a kind of belief. As we see, an affective view such as Swinburne's calls that assumption into question. But the affective and the cognitive views also make a common assumption: that belief is always belief that, propositional belief. An important characteristic traditionally associated with faith is then neglected: that faith is an answer, from the side of human creatures, to divine revelation. To answer a teller is not only to believe what one is told, it is also to believe the one who told.Footnote ⁷ Augustine defines ‘faith’ in a very large sense, covering both human faith (if the teller is human) and divine faith (if the teller is divine), as believing someone on his or her authority: what we know we owe to reason, what we believe we owe to authority (quod intelligimus debemus rationi, quod credimus auctoritati).Footnote ⁸ To believe God is to answer positively to God's telling so-and-so, be it a promise, an order or a simple declaration. It is at least part of the meaning of Abraham's faith, by which he believed God's promises and obeyed God's commands. Such an understanding of faith implies then both that God has told something (we can call it ‘revelation’), and that what He has told is believed because He himself is believed, or that it is not because of what is said, but because of who said it, that it is believed.

This traditional conception of faith, which has roots in Paul's epistle to the Romans (10:17 fides ex auditu), in the Hebrews Epistle, in the early councils, and in many Fathers of the Church, is well captured by the definition given by Vatican Council I: ‘faith is a supernatural virtue by which . . . we believe what God has revealed, not because the intrinsic truth of the things revealed has been perceived by the natural light of reason but because of God's authority who reveals them and can neither deceive nor be deceived’ (DS 3008). Apart from the qualification of faith as a supernatural virtue, the quotation from Vatican I states three features: faith is (1) an answer to a/the divine revelation, (2) because of God's authority (and not because of some natural light), (3) which is guaranteed by the indefectible trustworthiness of God. What follows will be a commentary on that definition. Since it insists on the human answer to divine to Revelation on the basis of divine authority, we need first to elucidate the notion of believing someone, which characterizes personal belief and trust. I will then apply it to God, and consider the two other features of the definition.

Believing (and trusting) someone

The speech act of telling is that of giving an assurance to the audience that what is said is true, by the very act of telling itself and not by way of any other consideration.Footnote ⁹ The teller certainly has authority over the propositional content and the nature of his speech act. It depends on him to say that p, and to present his saying that p as a telling that p. By this, he invites the audience to believe that p, and to believe it just because it is he who says that p. The teller endorses the epistemic responsibility concerning the truth of p; he takes on his shoulders the burden of being justified, and presents his own telling to the hearer (or reader) as sufficient for her to have a justified or warranted belief – that p. The teller entitles, by his own telling, the audience to believe that p. And he demands that what he says be believed in recognition of his own epistemic authority over the audience; he demands to be believed or trusted for the truth of his saying.Footnote ¹⁰ The hearer is not only invited to believe what the teller said, but to believe it because the teller told it. She is invited to accept the teller's commitment in abandoning her responsibility in the teller's hands. Any other response would frustrate him.

Of course the refusal to believe what is told would frustrate the teller. But, even if the hearer finally believed that p, she would equally frustrate him if she looked for some further reasons or evidences that p is true. She would believe what the teller said, but she would not believe the teller – which is what the telling is demanding. It would not be sufficient to add that she must believe the teller that p because he said it: he might just have awakened her attention; nor to add that she must rely on his saying: she might know he is ‘double bluffing’ and so believe what he says though he intends to fail her; nor even to add that she must believe what he believes and have reasons herself to think the teller is right: he might be right by chance.Footnote ¹¹ In that case she would not believe him. To believe someone that p, you must not only believe that p, but also believe it because he is telling you that p, and in an adequate response to the telling: by trusting the teller.

Looking for evidences that have nothing to do with the teller and his telling is obviously refusing to take his telling as a sufficient reason to believe what he says, and so to believe and trust him.Footnote ¹² But what about the search for evidences concerning the reliability of the teller, his sincerity and his competence? Sincerity might be ensured by the circumstances and human psychology, as one can suppose that under threat, or under torture, the prisoner will be sincere in giving information. His declarations would be good reasons for their truth. One could rely on them or on him, but that still would not be to believe or to trust him. The prisoner's declarations would be evidences for their truth, manifest expressions of his beliefs, in a way his behaviour could also manifest his beliefs. But they would not be a telling: he would not endorse any responsibility, he would not commit himself, but just behave as a natural gauge of truth.Footnote ¹³ Telling someone that p is not to present one's declaration as evidence for p: the teller would be frustrated and maybe offended if his audience wanted to check his sincerity by a truth-serum, a threat, or any other procedure, including a checking of the evidences for what he says. On the contrary, if the hearer wanted to check the teller's sincerity, she would show thereby that she does not care about his commitment and does not abandon into his hands her epistemic responsibility.

Being believed is the adequate response required by telling. If the hearer treats the teller as a gauge of truth, she might be disappointed in case the information were wrong (as one would be disappointed by a failing machine), or even angry with the teller (as one would not be against a machine), but she would not feel betrayed by the teller, as she would if she had accepted his commitment and abandoned her responsibility into his hands. Proofs of sincerity might confirm the trust given to someone's word, but they cannot ground it, since the more one relies on such proofs, the less one is trusting the teller. This is not to say that in trusting someone who tells her that p, the hearer is not thinking that the teller is trustworthy, nor to deny that, if she thought him not to be trustworthy, this would play against her trusting him. What is incompatible with trusting someone is to ensure his trustworthiness from certain conditions (threat, torture). And it is also incompatible with trusting someone to ground the belief that he is trustworthy on evidences that are independent of his telling (truth-serum).

This leaves room for other proofs or evidences of trustworthiness. A person is usually considered trustworthy because she has been sincere and competent in the past. Thus, trustworthiness increases with time: the more one has given sincere and competent declarations in the past, the more one is supposed to give others in the future. This is of course an inductive, Humean, justification of belief acquired by testimony. And now the question is: is it possible to trust someone while grounding one's belief on the (past) evidences of his trustworthiness? I think not, and for the same reasons as before. From the point of view of the teller, if the demand to be believed were satisfied in case the hearer considered past evidences concerning his sincerity and competence, he would have no point in committing himself and assuming epistemic responsibility. He would be offering his declaration as evidence for its truth, having to be relied on for reasons independent of any actual commitment, freedom, and choice. But this is not what he is demanding in telling someone that p, and once again I think he would be frustrated and maybe offended if his audience wanted to check his actual telling against his past ones. On the side of the hearer, to rely in that way on past evidences of sincerity and competence of the teller would not be to abandon her responsibility; on the contrary, it would be to rely on evidences that are independent of the actual telling, and to treat the teller as a gauge of truth in the long run. Acting so, the hearer would be disappointed if the teller were insincere in a particular case (induction would be defeated), but she would have no right to complain of being betrayed. So to rely on evidential proofs of trustworthiness is not to trust or believe the teller.

Why would one believe the teller and accept his authority, if trust is not reducible to an evidential belief concerning the teller's trustworthiness? I have no other answer than to say that it depends on a personal relationship, a relation from the hearer to the teller, and no other word than a relation of trust or of confidence, or of faith. That relation may have evidentialist grounds and confirmations, but it cannot be reduced to them, as the act of trusting cannot be reduced to an evidentially based judgement of trustworthiness. It is a relation (also) built out of personal relationships of esteem, moral appraisal, love, and friendship, which creates a feeling of dependence over the trustee. This is what I call personal trust.

Believing (and trusting) God

If to have Christian faith is to believe God, then what has been described as believing someone applies similarly to human and to divine faith. Or does it? Doesn't the fact that God is now the teller change something? The preceding paragraphs are an account of the authority of the teller, any teller, and so of the second element of the Vatican's definition of faith. I will now comment on the two other elements: the fact that it is God who reveals, and the insistence on his full competence and full sincerity. I will end with considerations on the rationality of faith.

1. The first element of the definition states that faith answers to God's revelation, or telling. It implies that God has revealed something and so that God exists. Faith, then, does not bear upon, but presupposes, those two propositions. It would not be absurd to believe a revelation concerning God's existence, or concerning the fact that God has revealed something, or that he has revealed so-and-so, if that revelation were not taken as coming from God. But it would be absurd to think the revelation is true because it comes from God, or that it is divine and that God exists because it says so. It may come as a surprise, or as an objection, that faith excludes from its content the very existence of God, since we often call ‘believers’ precisely those who believe that God exists. We can answer to this, first, that it might sound paradoxical if we look at contemporary uses of ‘faith’, but not if we look at the uses of the term in the Bible: to believe is to believe the word of God, of the prophet, of Jesus, and the sin of unbelief is not the negation of God's existence, but rather the lack of confidence in His word or in Jesus's word.Footnote ¹⁴ Second, note that to say that God's existence is presupposed by faith does not mean that one must already believe in God's existence before believing his supposed revelation. The point is logical, and it might be that one happens to believe both at the same time. Theistic (religious) belief and faith can begin and grow together.

What is the basis for both beliefs that God exists and has spoken? It can be evidences and it can also be human faith. Natural theology offers arguments for the existence of God. One can then argue in favour of the probability of a Revelation. And some evidences (from miracles or from prophecies) can favour the belief that some writer or speaker is writing or speaking in the name of God, so that such and such a discourse is originally from God. So many evidences can be adduced to sustain the preambles of faith.Footnote ¹⁵ One can wonder whether those arguments are good and persuasive. One can also wonder whether the preambles are usually believed on such a basis alone. I guess not. And since my goal is to give an account of the nature or faith, it would be a deficiency if it were limited only to the case of people with a philosophical mind. We must also have an account of the ‘faith of the simple’. Most probably it begins with human faith in people telling that there is a God, and that He revealed so-and-so. Those people may be parents, pastors, prophets, in a word the Church, who are believed (that p) on the basis of human faith (to be distinguished with believing the Church as speaking for God, which is then divine faith).

Because of the nature of the teaching and the presentation of it as ‘coming from God’, simple human faith may not be enough to give rise to a full attitude of trust. But, once one believes that God is speaking through the prophet or through the Church, human faith becomes inextricably mixed with divine faith: it is now because one believes God that one believes the Church. This seems to be an empirical fact, and to fit with the model proposed of the good of trusting someone. It has also a theological rationale: if God finds some good in having human beings believing (trusting) him, He might also find some good in having them trusting each other, and specially trusting the Church. And finally this mixing of divine and human faith makes room for the idea that religious faith bears not only on the divine word, but on the fact that some writer or speaker is divinely inspired and a messenger of God (when he is not partly divine . . .), and also on the fact that there is a God. To believe by faith that there is a God would be to believe it while accepting a divine revelation. This is the way most people believe that there is a God: as part (presupposition) of their acceptance of a religious revelation.

2. Let us now come to the third feature of the definition of faith: that God can neither deceive nor be deceived. Doesn't this imply that one relies on God's competence and sincerity, and so uses Him as a (divine) gauge of truth? How could there be room left for believing God, and so for (religious or divine) faith as previously defined?

One may believe someone to be trustworthy and so believe what he says, without believing him. So an expert in psychology who believes what the examinee is saying, because she knows him to be sincere and competent. One might also have sufficient reasons to believe the teller to be trustworthy but resist believing what he says on a particular occasion, because of some reasons for the opposite (the expert might resist believing the examinee saying something extraordinary, for example a revolutionary account of Napoleon's death). We could say the hearer believes the teller to be trustworthy in general but does not believe what he says in that case, so that the teller is not trustworthy in that case. Finally, we can imagine that, in spite of those contrary reasons, and not because of her belief that the teller is trustworthy, the hearer nonetheless comes to believe what he says because she trusts or believes him. Those are human cases, in which the trustworthiness had evidences rationally sufficient to ground the belief, but psychologically insufficient to produce it, so that room is still left for belief and disbelief, for trust and distrust.

But what if the teller were known to be absolutely competent and sincere as God is known to be by nature? We suppose there are no evidences strong enough to overcome, nor even to counterbalance, the evidence for God's telling the truth. It is an a priori evidence, not arising from experience and grounding a defeasible inductive inference in the teller's trustworthiness, but arising from the true concept of God's nature, and grounding a deductive inference: ‘since He is telling that p and is absolutely sincere and infallible, then p’. There seems to be no room for trust or distrust on purely theoretical and unimportant matters (as there was with the preceding example). But it is not obvious that, concerning what one cares about, such as one's own life, the deductive inference from indisputable premises is always sufficient for believing the teller. Think of the episode presenting Jesus walking on the sea and calling Peter to do the same. After some steps, Peter sinks. And Jesus says to him: ‘Man of little faith, why have you doubted?’ One could think Peter was able to infer: ‘since He tells me to walk, I can’. But that was not sufficient to overcome . . . what? The belief, naturally well grounded, that one cannot walk on the water? The fear of sinking? Jesus reproaches Peter for his lack of faith (his doubt). He should have believed Jesus, he should have trusted him. In fact, he did so at the beginning, and we might say that he lost his faith on the way. How to interpret that loss?

On one interpretation, Peter keeps his belief that Jesus is trustworthy, but stops believing Jesus and so stops believing what Jesus says (‘you can walk’). Believing that Jesus is trustworthy is then not enough to believe what he says, one needs also to believe him. On that interpretation, the lack of faith breaks the step, otherwise natural, from belief in one's trustworthiness to belief in what one says. On another interpretation of the case, Peter did not stop believing what Jesus said, but stopped from accepting it, from acting on the basis of what he believed. To believe someone that p, implies not only that one believe that p, on the authority of the teller, but also that one accept that belief and act on that basis. On both readings, faith supposes a decision to abandon full security, to risk oneself, in action (walking on the water).

If I am right here, this shows that something must be added to the preceding analysis of faith as believing someone. Not only does faith require abandoning one's epistemic responsibility into the hands of the teller, it also requires acting on that basis. Trust as reliance on the authority of the teller might be required for action, even when it is not required for belief. Some trusting may just emerge naturally, involuntarily or passively, on a personal relationship built out of affective bonds with the teller. This is the way we often believe what we are told by other people, by friends, by teachers, or even by the man in the street from whom we ask the time and believe his answer just on the basis of our shared humanity.Footnote ¹⁶ But an act of trusting does not always naturally follow from such a relationship. It might also be voluntary, as Jesus's reproach, in the Gospel's episode, supposes that the lack of trust is voluntary or that the lacking act of trust would have been voluntary.

I would like to compare voluntary trusting to acceptance, and passive trusting to belief.Footnote ¹⁷ Acceptance is of another kind than belief and other states (like desire, hope, etc.), and compatible with them. As we have seen that one can believe a certain proposition without accepting it or acting on it, it might be said that one can also act as if a proposition were true, or accept it, without holding it true, or believing it.Footnote ¹⁸ The profession of faith, ‘I believe’, might express active acceptance without passive belief (as when one doubts, but wants nonetheless to profess one's faith). Whereas passive trust seems to imply belief (if I passively trust my friend out of my relationship with him, I believe what he tells me), active or voluntary trust does not. Of course belief and acceptance usually go together, and so do passive and active trust, but they need not and can come apart. I might trust a friend who tells me to jump over a precipice though I fear and believe I may well fall. Peter might have (and maybe first has) trusted Jesus's word, and acted on it, while naturally believing this was dangerous (even if he had reasons to think otherwise: Jesus's perfect trustworthiness).

Acceptance seems close to Swinburne's assumption. But to assume a proposition (or a belief) is to act on it, to act as if it were true, because one calculates this is one's best chance to reach a certain objective, in the absence of good and sufficient information. However, to trust (believe) someone actively is not decided on the basis of a calculus, but on the basis of the same kind of relationship that prompted passive trust. It is an act that corresponds to a tendency to go with the teller, even while not believing fully what she says.Footnote ¹⁹ This is why actively trusting someone, even without believing what he says, and only accepting it, is still believing him. The teller might expect the hearer to believe what he says on his own authority. And there might be some disappointment if the hearer only accepts without believing it. But if acceptance is based on a voluntary act of trusting, it still is a recognition of the teller's commitment, of his authority. And it still is a renunciation of one's own epistemic responsibility into the hands of the teller, and an appropriate response to his telling. So the teller should not be frustrated if the hearer only accepts what he tells her. He might even appreciate, admire, and feel some gratitude for such a risky and gratuitous act of trust. There is a good in trusting someone and in being trusted in general: it improves the personal relationship. There is a great good in trusting and being trusted actively: it has the form of a meritorious action, arousing a feeling, a reactive attitude of gratitude and of moral appraisal.

3. This account of personal trust seems to defy rationality. Trusting belief is prompted by, and trusting acceptance is decided on the basis of, a personal relationship with the teller. It is not grounded on some evidences in favour of the truth of her saying nor in favour of her trustworthiness. Nor is it the rational conclusion of any utilitarian calculus of one's best chances to reach a certain goal. In that sense, both attitudes are not grounded on a rational process. That does not mean they cannot be rationally justified, since they can be backed by evidences. In fact, Richard Swinburne's work shows that one can argue for all religious beliefs: for the existence and nature of God, for the credibility of the prophets, for the contents of Revelation (for the truth of the doctrine of the Trinity and that of the Incarnation). And once God is recognized as the teller of the revelation (in the Scriptures), then belief that the revelation is true is grounded on divine infallibility and sincerity. But, if a belief is sufficiently grounded (or taken by the believer to be sufficiently grounded) on such evidences, then there is no more room for trust and so for faith. In the same way as Paul says faith should disappear with the full vision of the divine essence, faith disappears when the arguments in its favour are considered sufficient evidence for it. Now, if the Revelation is a telling from God to humanity, since a telling requires faith from its audience, it is probable that God would not make such a requirement if faith could be replaced by sufficient evidence.

A coexistence of faith with full evidence seems to be at stake in the Gospel's insistence that the disciples had faith in the Resurrection, even though they had full (sensory) evidence for it. ‘Because you have seen, you believe’ says Jesus to Thomas, and the ‘believe’ here is certainly the verb for faith. It seems that, according to John, in order to believe that Jesus was risen it was not enough to see him. One had to recognize him, and sensory evidence was not sufficient for this. What is the telling and what is the believing in response to a telling here? Well, Jesus did announce or promise his resurrection. It might well be that the faith shown by the disciples when they saw the empty tomb (John), or when they heard the women (the nine others), or when they saw Jesus (Thomas), or when they saw him doing a certain thing (the disciples from Emmaus) was a trusting and late answer to Jesus's promise. It might be argued that the same is true for later believers: they might have a certain amount of evidence in favour of the Resurrection (the empty tomb, the number and trustworthiness of the testimonies, the implausibility of any rival explanation), but might also need in addition (human) faith in the tellers (the ultimate tellers, and also the chain of tellers, up to the first tellers: the evangelists and the disciples), and (divine) faith in Jesus himself who announced his resurrection. The evidences would not suppress the element of faith even if they were sufficient for the belief to be evidentially grounded.

Concerning human faith, nonetheless, even if the teller could be confirmed by evidences (for the truth of her telling, or for her trustworthiness), it is often not so, and beliefs and acceptances are then not rationally (fully) justified, at least from the point of view of the hearer. The hearer might be externally justified (there are enough reasons to believe the teller tells the truth), but not internally (those reasons are not fully available to the hearer). It is then the personal relationship with the teller that grounds her belief and/or her acceptance, her trust in the teller. What is the protection against gullibility within such a view? Well, any reason to disbelieve what the teller says, or to put in doubt his trustworthiness, should require at least an epistemic reaction on the part of the believer. This reaction might go to the point of grounding evidentially the belief, and so to eliminate trust, or to check the objections. Maybe trust can remain in place only by default, when it is not held in check, but any reason received as opposed to trust should at least be examined.

When the teller is not perfectly trustworthy, it is always possible to think that he is not sincere, or not competent. Trusting the teller, even after examination, may seem akin to being gullible. But it is being gullible with some good in view: the improvement or the maintaining of the relationship with the teller. There is then some rationality in trusting someone, when there is not enough evidence to ground one's trust. Whereas propositional trust is rational on the basis of a rational estimation of the evidence for truth or for the best, personal trust might be rational in itself. If there is some good in trusting someone, a good that concerns human relationships, love, esteem, in addition to the good of giving us access (often) to true beliefs we would not have got otherwise, then we should not (it would be irrational to) adopt as a strategy the evacuation of all our beliefs that are based on personal trust. One who thinks he should do so, says Augustine, seems unable to have any friend (nullum mihi habere posse amicum videtur Footnote ²⁰ ). It would be a pity and irrational to let trust out of our lives.

The leap of faith can then touch divine as well as human faith. Concerning divine faith, since God cannot deceive nor be deceived, there is no prejudice in this lack of rationality, no gullibility to fear. The leap of faith is due not to insufficient evidence for belief (that), but to the very nature of trust, active or passive, in believing someone. Concerning human faith, which is often at the root of divine faith, there might be a leap over rationality, when evidence is insufficient to justify belief, and one relies on the authority of the teller. But, though this is not a reason to admit all kinds of leaps, and not a justification of any particular leap, we have good reasons to admit the existence of (some) leaps of faith in our lives.