Markan propositional faith

Non-doxasticism might well seem a Johnny-come-lately. At any rate, in analytic philosophical circles, it wasn't until Richard Swinburne's Faith and Reason (1981/2005) and Louis Pojman's ‘Faith without belief’ (1986) that the idea that faith that p does not entail belief that p began to gain traction. While doxasticists insist that faith that p entails belief that p, non-doxasticists allow that some attitude other than belief that p can play the role that doxasticists assign to belief that p. Among the attitudes that some non-doxasticists allow are doxastic attitudes such us belief that p is likely, or belief that p is more likely than not, or belief that p is significantly more likely than each of p’s credible contraries, etc., as well as non-doxastic attitudes such as propositional acceptance (Alston (Reference Alston, Jordan and Howard-Snyder1996), (Reference Alston, Greco, Mele and Timmons2007) ), propositional trust (Audi (Reference Audi2011); McKaughan (Reference McKaughan2013), (Reference McKaughan2016), (Reference McKaughan2017) ), propositional reliance (Rath (Reference Rath2017) ), propositional hope (McKaughan (Reference McKaughan2013); Pojman (Reference Pojman1986) ), imaginative assent (Schellenberg (Reference Schellenberg2005), (Reference Schellenberg2013) ), credence (Buchak (Reference Buchak, Chandler and Harrison2012), (Reference Buchak, Abraham and Aquino2017a), (Reference Buchak2017b); Sliwa (Reference Sliwa, Benton, Hawthorne and Rabinowitzforthcoming) ), and beliefless assuming (Golding (Reference Golding1990), Howard-Snyder (Reference Howard-Snyder2013), (Reference Howard-Snyder2016), (Reference Howard-Snyder2017a), (Reference Howard-Snyder2017b); Swinburne (Reference Swinburne1981/Reference Schellenberg2005) ).

So who's right, doxasticists or non-doxasticists? I argue for non-doxasticism (Howard-Snyder (Reference Howard-Snyder2013) ), and criticize arguments for doxasticism (Howard-Snyder (Reference Howard-Snyder2016) ), and Daniel McKaughan does the same (McKaughan (Reference McKaughan2013), (Reference McKaughan2016), (Reference McKaughan2017), (Reference Howard-Snyderforthcoming) ). Recently, Malcolm and Scott (Reference Malcolm and Scott2016) argue for doxasticism, and criticize arguments for non-doxasticism, to which I reply in Howard-Snyder (Reference Howard-Snyderunpublished). More recently still, Malcolm (Reference Malcolm2017) argues that doxasticism is preferable to non-doxasticism on the grounds indicated above, namely that it alone can unproblematically rule out religious fictionalists as people who have faith. In the context of defending this argument, Malcolm considers whether religious fictionalists can have faith. In order to get a clearer view of the issue, its potential answers, and the related argument for doxasticism, I will consider all three from the vantage of my theory of propositional faith.

Most recently, I have contextualized my theory of propositional faith in a study of the pistis lexicon in the Gospel of Mark, as follows:

Much needs comment. I refer the reader to the cited article; here I briefly clarify the key terms.

Regarding condition (b), I do not define x is a positive conative orientation. Even so, it is clear enough what I have in mind: ‘[a] positive conative orientation towards the object of faith consists in being for its truth, favoring its being the case, wanting it to be so, giving its truth a positive evaluation, regarding it as good or desirable, and the like’ (Howard-Snyder (Reference Howard-Snyder2017a), 48). The earlier expression of my view distinguished ‘a positive evaluation of p’ from ‘a positive conative orientation towards p’, allowing that one might regard the truth of p as good or desirable without having any desire in virtue of which one cared with positive valence that p is true; nevertheless, I argued, both are required (Howard-Snyder (Reference Howard-Snyder2013), 362–363). I now lump these two requirements together in (b), which might be better put as requiring that you have a positive conative-evaluative orientation towards p’s truth.

As for condition (c), the idea is that it is not enough that one simply has the requisite cognitive and conative-evaluative attitudes; they must motivate one's behaviour if one has faith. Of course, on some models of the relations between cognition, conation, and behaviour (e.g. belief-desire-behaviour models), if one has certain cognitive and conative-evaluative attitudes, one will tend to behave in certain ways – in which case (c) might seem redundant. However, if the requisite cognitive and conative-evaluative attitudes can sit idle in a person, or if they can be systemically shut down by, e.g., irrevocable weakness of will or a stray gamma ray, then (c) doesn't seem redundant. Perhaps it's the better part of wisdom to make this behavioural-motivational condition explicit, even if, on some views, it is already implicit.

Condition (d), I argue, is what the author(s) of the GMark regarded as the single-most important feature of faith. In this I am not alone. Christopher Marshall, the foremost expert on faith as a theme in GMark, puts the point well: ‘Without doubt, the leading characteristic of Markan faith is sheer dogged perseverance’ (Marshall (Reference Marshall1989), 237). Condition (d) expands my (2013) condition according to which you have faith that p only if you are resilient in the face of counterevidence to p. The expansion recognizes a variety of non-evidential challenges to living in light of one's faith, e.g. fatigue, emotional dryness, depression, and contrary desires.

That leaves (a), the positive cognitive attitude condition. What, exactly, is it to have a positive cognitive attitude towards a proposition? I offer no definition, but several comments shed enough light to make it sufficiently understandable and theoretically useful.

First, two platitudes among others govern theorizing about propositional faith: (i) it is incompatible with disbelief and, relatedly, (ii) it requires some sort of belief-like attitude. Participants in the debate between doxasticists and non-doxasticists need a neutral way to speak of the attitude in (ii) while abiding by (i). The term ‘positive cognitive attitude’ lends itself to the task.

Second, a positive cognitive attitude, as its name suggests, falls on the cognitive side of the line between the cognitive, on the one hand, and the conative and/or volitional, on the other; the line between thought, on the one hand, and desire and/or will, on the other.

Third, we can distinguish paradigm cases of positive and negative cognitive attitudes. Believing something and disbelieving it are both cognitive attitudes, but only believing it is positive; disbelieving it, as the prefix suggests, is negative.

Fourth, some cognitive attitudes are neither positive nor negative, e.g. entertaining the proposition that all platypi are monotremes, or wondering whether the Diamond Creek Fire will burn until snowfall.

Fifth, taking the paradigms of belief and disbelief as clues, we can say that, unlike entertaining or wondering, both positive and negative cognitive attitudes have a mind-to-world direction of fit, in this sense: a positive cognitive attitude towards p is how it ought to be only if p is true, and a negative cognitive attitude towards p is how it ought to be only if p is false.

Sixth, a cognitive attitude towards p is positive in that one who has it takes a stand in favour of p's truth, whereas a cognitive attitude towards p is negative in that one who has it takes a stand against p's truth. How might we understand this metaphor? I suggest that we think of taking a stand in favour of p as partly consisting in a complex dispositional state, one we would antecedently expect to be congruent with taking a stand in favour of p (Howard-Snyder (Reference Howard-Snyder2013), 361, 366; cf. Howard-Snyder (Reference Howard-Snyder2017b), sect. 5). (Mutatis mutandis, the same goes for having a negative cognitive attitude towards p and taking a stand against the truth of p.) More on this to follow.

Seventh, positive and negative cognitive attitudes are evidence-sensitive in this sense: no one can have a positive or negative attitude towards p that does not reflect their estimation of the truth or falsehood of p. Their estimation need not be conscious or deliberative; it might only be implicit and automatic. Their estimation need not be right; indeed, it may well be wrong, wildly wrong. But unless the stand they take in favour of or against the truth of p is at least partly due to their estimation of p’s truth or falsehood, they have neither a positive nor a negative cognitive attitude towards p.

Eighth, we might be tempted to think that in order for a cognitive attitude towards p to involve taking a stand in favour of the truth of p, its owner must have a tendency to outwardly affirm p when asked whether p and a tendency to inwardly assent to p when considering whether p. To be sure, that's how things go for the paradigmatic positive cognitive attitude towards p, i.e. belief that p. But there are ways to take a stand in favour of the truth of p that do not involve those two tendencies. For example, consider someone who lacks those two tendencies but who has a tendency to outwardly affirm and inwardly assent to the proposition that p is likely, or the proposition that p is five times as likely as each of its five credible contraries, among other possibilities. Moreover, they have a tendency to use p as a premise in practical and theoretical reasoning when appropriate and, more generally, they have a tendency to perform enough actions the performance of each of which is aptly described as a way to take a stand on behalf of the truth of p. Furthermore, suppose these tendencies are not grounded in self-deception, or an intention to deceive others, or anything like that; rather, suppose they are, in no small part, grounded in an estimation of the truth of p. Given that this person satisfies the other conditions for having a positive cognitive attitude towards p, their psychology constitutes taking a stand in favour of the truth of p, although tendencies to outwardly affirm p and inwardly assent to p are absent. (Mutatis mutandis, the same goes for taking a stand against the truth of p.)

Ninth, when Malcolm articulates what ‘most or all of the various formulations’ of non-doxasticism share in common, he writes that they are committed to the claim that, necessarily, you have faith that p only if you have ‘a positive cognitive attitude toward p’ (Malcolm (Reference Malcolm2017), 3). Now, I introduced the notion of a positive cognitive attitude towards p into the literature (Howard-Snyder (Reference Howard-Snyder2013), 361). Unfortunately, what Malcolm says about it can easily lead to misunderstanding. For the sake of clarity, I mention five corrections.

(i) Malcolm says that the sense in which positive cognitive attitudes are ‘positive’ is that ‘they involve a positive orientation towards the object of the proposition’ (Malcolm (Reference Malcolm2017), 3). This is false, at least if by ‘positive orientation’ Malcolm means ‘positive conative-evaluative orientation’. You can have a positive cognitive attitude towards the proposition that you are dying of cancer; even so, you do not regard its truth as good or desirable and you do not care (with positive valence) that it is true. The positivity of a positive cognitive attitude towards p is not the positivity of a positive conative-evaluative orientation towards p’s truth.

This is a good place to point out that I am responsible for any misunderstanding here. After all, I expressed the positivity of a positive cognitive attitude towards p with taking a stand in favour of p, while at the same time illustrating the positivity of a positive conative-evaluative attitude towards p’s truth with favouring p’s truth. The result is that, quite confusingly, you can take a stand in favour of the proposition that you are dying of cancer without favouring its truth. Clearly enough, I would not mean the same thing by the two uses of ‘favour’ in that sentence. Still, as it stands, it is confusing. Perhaps it would be better to express the positivity of a positive cognitive attitude towards p as taking a stand on behalf of p, which is less evaluatively loaded than in favour of. At any rate, it seems less confusing to say that you can take a stand on behalf of the proposition that you are dying of cancer even when you do not favour its truth; after all, your evidence for it might be unimpeachable.

(ii) Malcolm says that the variety of non-doxastic positive cognitive attitudes offered as alternatives to belief that p ‘all share’ a common feature: ‘that they can be adopted voluntarily, or at will’; that they are ‘ “taken on” by the agent voluntarily’; that they are ‘formed’ with ‘our (direct) voluntary control’ (ibid.). This is false. While some non-doxasticists take themselves to posit attitudes that they think are under our voluntary control (Alston (Reference Alston, Jordan and Howard-Snyder1996), McKaughan (Reference McKaughan2016) ), others do not, or leave the matter open (Audi (Reference Audi2011); Howard-Snyder (Reference Howard-Snyder2013) ). At any rate, it is not definitive of a non-doxastic positive cognitive attitude that it be under our voluntary control.

(iii) When speaking of positive cognitive attitudes, Malcolm emphasizes the contrast between non-doxastic attitudes and belief simpliciter. This is misleading. For, on doxasticism, many doxastic attitudes are incompatible with faith that p, a point that I underscored (Howard-Snyder (Reference Howard-Snyder2013), 361; Idem (Reference Howard-Snyder2016), 153–154). On doxasticism, faith that p requires belief that p. No other proposition will do. Thus, on doxasticism, you can't have faith that p if instead of believing p you only believe that p is likely, or you only believe that p is more likely than not, or you only believe that p is five times more likely than each of the five contraries to p you find the least bit credible, and so on. It is misleading to omit this fact and its correlate that, on non-doxasticism, these doxastic attitudes may well be fit to play the requisite role, along with the non-doxastic attitudes mentioned earlier.

(iv) Malcolm tells us that, on non-doxasticism, when it comes to the cognitive component of propositional faith, it ‘merely requires a “positive cognitive attitude”. This broad condition, however, can be satisfied by several pragmatic approaches to a domain, including fictionalism’ (Malcolm (Reference Malcolm2017), 1). Later we learn what it is for someone to take a ‘pragmatic approach to a domain’. They ‘positively affirm’ the propositions expressed in that domain ‘for pragmatic reasons’ (ibid., 5; cf. 13, 15), ‘for purely instrumental purposes’ (ibid., 7, emphasis added), ‘in order to attain certain goods’ (ibid., 12), for no reason ‘other than to accomplish one's aims’ (ibid., 16, emphasis added). Note: they do not affirm them, not even partly, on the basis of ‘evidential, or any other epistemic considerations’ (ibid., 5; cf. 7); not even partly because of their estimation of ‘the possible truth of the proposition’ (ibid., 16). So, on Malcolm's view, the positive cognitive attitude towards p required by faith that p either must or at least can be had for pragmatic reasons alone. This is false. As I made clear when I discussed examples, conditions, and objections (Howard-Snyder (Reference Howard-Snyder2013), 362, 366–367; Idem (Reference Howard-Snyder2017b), 157–166), and as I made clear above, you have a positive cognitive attitude towards p only if you take a stand on behalf of its truth, and that stand is at least partly grounded in your estimation of p’s truth. Consequently, on my view, the positive cognitive attitude towards p required by faith that p is not an attitude that is grounded only in ‘pragmatic reasons’ and not at all in ‘evidential, or any other epistemic considerations’.

(v) Malcolm includes Kvanvig (Reference Kvanvig2013) and Buchak (Reference Buchak, Chandler and Harrison2012) among those who offer theories of faith that are to be ‘included under the conditions for propositional faith’ that he discusses (Malcolm (Reference Malcolm2017), 2–3). As we've seen, one of those conditions, on Malcolm's view, is that you have faith that p only if you have ‘a positive cognitive attitude toward p’ (ibid., 3). In that case, you would expect that Kvanvig offers a theory of propositional faith, or at least a theory of faith that allows us ‘to determine what this attitude would be on the theory’ (ibid., 2). But Kvanvig offers no theory of propositional faith; moreover, insofar as we might try to determine what this attitude would be on his theory, it does not require a positive cognitive attitude towards a proposition. That's because, on Kvanvig's theory, faith is a disposition to act in the service of an ideal, and that disposition, he says, is compatible with disbelief that the ideal is realized or true. Unlike Kvanvig, Buchak (Reference Buchak, Chandler and Harrison2012) offers a theory of propositional faith. More accurately, she gives a decision-theoretic analysis of an act of faith that p, one that implies that you can have faith that p so long as, on the cognitive side, you have a non-zero credence that p. Thus, on Buchak's view, you can have faith that p even if your credence that p is vanishingly close to zero. That's a far cry from saying that you have faith that p only if you have ‘a positive cognitive attitude toward p’. According to Malcolm, these two theories require that you have a positive cognitive attitude towards p in order to have faith that p, which suggests a misunderstanding of the idea of a positive cognitive attitude, a misunderstanding to be avoided.

On MPF, you can't have faith that p unless you have a positive cognitive attitude towards p – in the sense specified here. Now let's turn to religious fictionalism.

Revolutionary religious fictionalism

Presumably, religious fictionalism is like other sorts of fictionalism. Mark Balaguer (Reference Balaguer and Zalta2015) tells us that mathematical fictionalism

Daniel Nolan (Reference Nolan and Zalta2016) says that modal fictionalists are fictionalists about possible worlds. They

As for fictionalism in ontology, Jason Stanley writes:

Richard Joyce (Reference Joyce2017) tells us that the moral fictionalist holds, among other things, to an error theory about moral discourse, according to which

A common theme here is that, according to the fictionalist, although we might for various reasons engage in a discourse that appears to commit us to the existence of certain entities, there are no such things; consequently, theories that invoke them are false or untrue, and the same goes for any claims that presuppose the existence of those entities or the truth of those theories.

Revolutionary fictionalists are commonly distinguished from hermeneutic fictionalists. Applied to ontology, Stanley says that

To generalize, while revolutionary and hermeneutic fictionalists agree that the relevant discourse refers to entities that don't exist, takes certain claims to be true when they are not, etc., they disagree over whether normal use of the discourse is a pretence. Hermeneutic fictionalists say it is, and so, e.g., normal reference to abstract objects is already reference within the pretence of abstracta. Revolutionary fictionalists says it isn't, and so, e.g., normal reference to abstract objects is not already reference within the pretence of abstracta . . . but, even so, the revolutionary fictionalist continues, we should use the discourse as if it did refer to abstract objects, and we should use the discourse as if it expressed truths about abstract objects, and so on – after all, it's quite useful to do so.

Malcolm is interested in revolutionary religious fictionalists (Malcolm (Reference Malcolm2017), 5ff.). So am I. Our tour through these terms of art suggests that revolutionary religious fictionalists make two claims: (i) there is no God, and religious theories and claims are false or untrue; nevertheless, (ii) just for practical reasons, we should employ religious discourse as if there were a God, we should use religious discourse as if it expressed truths about God, and so on. Just to be clear, let's call this view revolutionary atheistic religious fictionalism. (I'll leave ‘revolutionary’ and ‘religious’ implicit henceforth.)

We are now in a position to answer a disambiguation of our title question: can atheistic fictionalists have faith that, e.g., there is a God? On doxasticism, the answer is arguably no. Atheistic fictionalists disbelieve that there is a God, and so they lack belief of that proposition, and so they lack faith that there is a God. On MPF, the answer is also arguably no. Atheistic fictionalists disbelieve that there is a God, and so they have a negative cognitive attitude towards that proposition, and so they lack a positive cognitive attitude towards it, and so they lack faith that there is a God. Let me comment on each of these two arguments, starting with the second.

Atheistic fictionalists disbelieve that there's a God. As such, they take a stand against the truth of that proposition. Four things follow.

First, when they are asked whether there is a God, they have a tendency to outwardly deny it. To be sure, the contours of their tendency to outward denial will differ from the contours of the tendency to outward denial had by the atheistic non-fictionalist. For example, in religious settings, we expect that the atheistic fictionalist will not outwardly deny that there is a God when asked but rather deceptively or insincerely affirm it (or perhaps ‘quasi-outwardly-affirm’ it, if there is such a thing as quasi-outward-affirmation that avoids deception and insincerity). But in the same settings, we expect that the atheistic non-fictionalist will outwardly deny that there is a God when asked. In the sort of case imagined here, the atheistic fictionalist's moral commitment to outwardly act as if they believe that there is a God inhibits the manifestation of their disbelief, whereas the atheistic non-fictionalist lacks any such inhibition. Even so, they both have a tendency to outwardly deny that there is a God.

In this connection, it is useful to compare the atheistic fictionalist with an atheistic non-fictionalist who, e.g., due to fear of reprisal, commits to outwardly acting as if they believe that there is a God. (Imagine that they live in Iran, Saudi Arabia, Mauritania, or Sudan, where the death penalty can be legally imposed for outwardly affirming atheism.) Imagine further that their fear and commitment inhibits their behaviour in all the ways that the atheistic fictionalist's fictionalism inhibits theirs. In that case, does the atheistic non-fictionalist lack a tendency to outwardly deny that there is a God? I think not. Rather, they have a tendency to outwardly deny that there is a God but its manifestation is inhibited. The same goes for the atheistic fictionalist. The manifestation of a tendency can be inhibited – even severely – without eliminating the tendency.

Second, just as the atheist who becomes a fictionalist retains a tendency to outwardly deny that there is a God when asked, so they also retain a tendency to inwardly deny that there is a God when they consider the matter.

Third, when the atheistic fictionalist is asked whether there is a God in a court of law or any other context where the penalties for failing to disclose one's disbelief are much more severe than non-disclosure, and they are averse to such penalties, they will outwardly deny it. Moreover, when the atheistic fictionalist is asked whether there is a God while attending an assembly of atheistic fictionalists – e.g. a conference with a pro-atheistic-fictionalist theme, or an atheistic fictionalist support group – they will outwardly deny it. More generally, when the atheistic fictionalist is asked whether there is a God and all inhibitions are absent, they will outwardly deny it.

Fourth, the direction of fit of their cognitive attitude towards the proposition that there is a God has the following property: being how it ought to be only if it is false.

On MPF, none of these four things is the case if you have faith that there is a God. If you have faith that there is a God, you have a positive cognitive attitude towards that proposition, not a negative one. As such, you take a stand on behalf of the truth of the proposition that there is a God. Thus, when asked whether there's a God, you lack a tendency to outwardly deny it; and when you consider whether it's true, you lack a tendency to inwardly deny it. Moreover, it is false that, when someone who has faith that there is a God is asked whether there is a God and all inhibitions are absent, they will outwardly deny it. Furthermore, the direction of fit of your cognitive attitude towards the proposition that there is a God lacks the property of being how it ought to be only if it is false.

In passing, note that Malcolm expresses doubt that, on my view, faith that p precludes disbelief that p (Malcolm (Reference Malcolm2017), 3–4, 6, 7; Malcolm & Scott (Reference Malcolm and Scott2016), 13). I disagree for reasons indicated above and earlier (Howard-Snyder (Reference Howard-Snyder2013), 361; Idem (Reference Howard-Snyder2016), 149).

Now consider the first argument above – the argument that, on doxasticism, atheists disbelieve that there is a God, and so they lack belief of that proposition, and so they lack faith that there is a God. Interestingly, Malcolm never addresses the question of whether, on doxasticism, faith that p precludes disbelief that p. He appears content to observe that, on doxasticism, faith that p requires belief that p (Malcolm (Reference Malcolm2017), 11–12). That's unacceptable, however. Instances of S believes that p contradict instances of It's false that S believes that p, but they don't contradict instances of S disbelieves p. So why suppose that, on doxasticism, faith that p precludes disbelief that p? Perhaps a satisfying answer lies in the neighbourhood of an argument analogous to the one I gave for the conclusion that, on MPF, faith that p precludes disbelief that p.

So, to return to our question: can atheistic fictionalists have propositional faith? Answer: not if faith is what MPF says it is.

As it turns out, some people posit agnostic fictionalists (Nolan et al. (Reference Nolan, Restal and West2005); van Fraassen Reference van Fraassen1980). This is how Malcolm approaches our title question, in the main. He grants, just for the sake of argument, that non-doxasticism generally and, presumably, MPF in particular, preclude the atheistic fictionalist from having faith (2017, 6–7), but he argues that agnostic fictionalists pose a problem for these views as well. In this connection, he describes a woman, Sam, who grows up in a Christian home, believes the BCS, engages in the practices and discourse of her community, and is committed to following Jesus. However, at an older age, she begins to feel the pull of a variety of considerations, both evidential and not, against her stance towards the BCS. Nevertheless, she is

Sam is Malcolm's paradigmatic agnostic fictionalist.

Malcolm does not tell us what he means by ‘agnostic’. Etymologically, in its religious use, the agnostic lacks knowledge of God, or they believe that people generally lack knowledge of God. Presumably this is not what Malcolm has in mind for Sam. Perhaps he means that the agnostic is someone who neither believes nor disbelieves that there is a God. That would allow agnostics to believe it's likely that there is a God, however. Presumably this is not what Malcolm has in mind. Perhaps this is closer to Malcolm's intentions: to be an agnostic is to withhold judgement on whether or not there is a God, including judgements about how likely or plausible it is – except, perhaps, for judgements in the neighbourhood of it is as likely as not that there is a God or the probability of the proposition that there is a God falls in the interval <0,1 >. In what follows, this is how I will think of the agnosticism of agnostic fictionalists.

So, then: with Sam as our paradigm, can agnostic fictionalists, so understood, have faith? On doxasticism, the answer is arguably no. Agnostics lack belief that there is a God, and so they lack faith that there is a God, etc. On MPF, the answer is arguably yes, however. After all, even though Sam is an agnostic, in the specified sense, Malcolm says that she accepts the BCS, which some non-doxasticists regard as one of the many ways in which you might have a positive cognitive attitude towards a proposition. Given the rest of Malcolm's description, isn't it just obvious that Sam satisfies the other conditions for faith laid down by MPF? Malcolm thinks so (ibid., 7).

The argument from fictionalist faith

We have here a challenge to the proponent of non-doxasticism in general and MPF in particular. I will focus on MPF. The challenge can be put in the form of an argument.

The Argument from Fictionalist Faith (MPF version)

1. 1. If MPF is correct, then faith that p requires a positive cognitive attitude towards p but it does not require belief that p.

2. 2. If faith that p requires a positive cognitive attitude towards p but not belief that p, then some agnostic fictionalists (notably, Sam) can have faith that there is a God.

3. 3. No agnostic fictionalist can have faith that there is a God.

4. 4. So, MPF is incorrect. (1–3, logic)

What should we make of this argument? (1) is true by definition, and (2) seems to be supported by the reflections at the end of the previous section. That leaves (3). Malcolm asks: ‘why think that fictionalists do not have faith?’ (ibid., 9). Good question. Indeed, that's my question. Why think that no agnostic fictionalist can have faith? Why think that (3) is true? Malcolm affirms ‘a few reasons for why this might seem to be the case’ (ibid., 9–11). Let's consider them, one by one.

‘The first reason [to ‘think that fictionalists do not have faith’] might be theoretical’, says Malcolm. Agnostic fictionalism, on the one hand, and non-doxastic theories of faith (‘NDT’ in Malcolm's nomenclature), on the other,

This is confused. Obviously agnostic fictionalism is not the same theory as any version of non-doxasticism; the explananda and the explanans are distinct. But that fact is wholly irrelevant to whether or not Sam – Malcolm's model agnostic fictionalist – and others of her ilk engage with the practices and discourse of Christianity in a way that is ‘quite different’ from the way in which someone who has faith engages them. That is, the fact that the theories are distinct is wholly irrelevant to the proposition that fictionalists do not have faith.

The second reason Malcolm gives for thinking that premise (3) is true immediately follows the first:

Once again: confusion. To repeat: non-doxastic theories of faith are distinct from theories of fictionalism, but that fact is wholly irrelevant to whether or not, on non-doxasticism in general or MPF in particular, fictionalists can have faith. That is, the distinctness of the theories is wholly irrelevant to (3).

Malcolm's third reason to ‘think that fictionalists do not have faith’ raises an important issue, one that leads to another disambiguation in our title question. He asks: ‘What guides the intuition that fictionalists do not have faith?’ He answers:

So agnostic fictionalists are ‘engaged in a pretense’, they are ‘merely pretending in their attitudes’ towards religious claims; for this reason, asserts Malcolm, no agnostic fictionalist can have faith.

This is surprising. Pretence was not explicit in Malcolm's description of Sam, his paradigm agnostic fictionalist. Nowhere did he describe her as ‘merely pretending in [her] attitudes’ to the BCS and the like. Where does pretence show up? Does she ‘merely pretend’ to be in doubt about the BCS, when she really disbelieves it? Does she ‘merely pretend’ to accept it, when in fact she doesn't? Does she ‘merely pretend’ to retain her affection for the BCS, when she couldn't care less? Does she ‘merely pretend’ to inwardly acknowledge her sins, to feel contrition for them, and to be grateful for the Lord's forgiveness, when she really never acknowledges them, never feels contrition, and is never grateful? Does she ‘merely pretend’ to think of moral, social, political, and personal issues from a Christian perspective, when she really never thinks of them from any such perspective? Does she merely pretend to direct her soul to God when she prays, ‘Our Father, who art in heaven, hallowed be thy name’? Does she merely pretend in these attitudes and hundreds like them only because she wants to keep her job, retain her friendships, keep her family relations peaceful, look legitimate in the eyes of her community, and so on? Is she really engaged in a colossal deception?

If agnostic fictionalists merely pretend in their attitudes towards the BCS and the like, then they don't satisfy the conditions that MPF lays down for faith; they ‘merely pretend’ to do so. Thus, on MPF, (3) is surely true and, just as surely, (2) is false. For even if faith that p requires a positive cognitive attitude towards p but not belief that p, it does not follow that agnostic fictionalists can have faith that p. That's because agnostic fictionalists – as Malcolm describes them – are mere pretenders in their attitudes towards the BCS, and so they do not really satisfy MPF's conditions on faith.

Before assessing Malcolm's other reasons to ‘think that fictionalists do not have faith’, let's pause over two things. First, Malcolm asserts that proponents of non-doxastic theories of faith ‘cannot individuate’ those who ‘pretend to have faith’ from those who really have faith (ibid., 10; cf. Malcolm & Scott (Reference Malcolm and Scott2016), 11–13). This is bizarre. Everybody can ‘individuate’ those who ‘pretend to have faith’ from those who really have faith, just as everybody can ‘individuate’ those who pretend to have cancer from those who really have cancer. Of course, not everybody can distinguish (in the sense of tell the difference between) those who pretend to have faith and those who really have faith, just as not everybody can distinguish those who pretend to have cancer from those who really have cancer. But the inability to distinguish does not entail the inability to individuate. Second, we now have an answer to another disambiguation of our title question: can agnostic pretending fictionalists have faith that there is a God, etc.? Answer: not if faith is what MPF says it is.

Here an important question arises: must agnostic fictionalists pretend to have attitudes that can constitute faith? Is it definitive? I don't see why we should say that. At any rate, for the sake of thoroughness, I propose that we continue our inquiry into the answer to our title question while allowing that they need not be pretenders. Now back to Malcolm's reasons to ‘think that fictionalists do not have faith’.

Malcolm's fourth reason for premise (3) of the Argument from Fictionalist Faith has to do with ‘saving faith’.

What, exactly, is Malcolm's reason for (3) here? It's hard to say.

Here's a stab at making his words relevant: if agnostic fictionalists can have faith, then, if Christianity is true, they can have saving faith. But it's false that, if Christianity is true, they can have saving faith. After all, on Christianity, saving faith is ‘sincere and genuine’ but the (so-called) faith of a fictionalist is neither. Premise (3) follows.

What should we make of the premise that the faith of a fictionalist is neither sincere nor genuine? The charge of insincerity does not differ from the charge of pretending, and so advances no reason for (3) distinct from Malcolm's third reason. As for the charge of being ingenuine, it is nothing more than the claim that the faith of a fictionalist is not faith, which is a mere stylistic variant on (3) itself. So, at bottom, the fourth reason advances nothing over the third reason, or it is a mere stylistic variant on (3), the premise on whose behalf it is offered.

Malcolm's fifth reason to think that no fictionalist can have faith is that ‘fictionalism is pursued for the wrong kinds of reasons, reasons relating to one's own interests, rather than being properly guided towards a possible truth that one values highly’, which is required for faith (ibid., 11). The argument seems to be that, if a fictionalist can have faith that God exists, etc., then one can have faith for reasons related to one's own interests rather than for reasons related to the possible truth of p. But one can't have faith for reasons related to one's own interests rather than for reasons related to the possible truth of p. So, no fictionalist can have faith that God exists, etc.

This strikes me as an excellent reason to think that premise (3) is true, given how Malcolm describes the agnostic fictionalist here. Unfortunately, if it is a sound argument, premise (2) of the Argument from Fictionalist Faith is false. For even if faith that God exists does not require belief that God exists, it does require a positive cognitive attitude towards that proposition, and given Malcolm's description here, agnostic fictionalists lack a positive cognitive attitude towards it. That's because, on Malcolm's description, their reasons for having faith that God exists are not related to the possible truth that God exists and so they do not take a stand on behalf of the truth of that proposition, a stand that is at least partly due to their estimation of its truth. Thus, they do not satisfy MPF's positive cognitive attitude condition on faith.

Malcolm's sixth reason to ‘think that fictionalists do not have faith’ – what he calls ‘a third reason’ in addition to the ‘theoretical and practical’ reasons we have already canvassed – is found in these words:

How exactly are we supposed to connect these words with ‘fictionalists do not have faith’?

The best I can do by way of charity is this: if agnostic fictionalists can have faith that God exists, etc., then some of them will have faith ‘imposed’ on them, and that will displease them since they want to be distinguished from those who have faith. Faith should not be imposed on fictionalists; we should not displease fictionalists in this way. So, fictionalists can't have faith that God exists, etc.

What should we make of this argument? For my part, I propose that even charity rightly has its limits.

To sum up, Malcolm gives ‘several reasons’ to think that ‘fictionalists do not have faith’. Naturally enough, those reasons are relevant to premise (3) of the Argument from Fictionalist Faith, according to which no fictionalist can have faith. So far as I can tell, however, only the third and fifth are worthy of consideration and, if they are sound, then premise (2) of the argument is false.

Can non-pretending agnostic fictionalists have faith?

Let's return to our title question. So far we have learned that non-doxasticists need not grant that atheistic fictionalists or pretending agnostic fictionalists can have faith. At any rate, on MPF, there are no such people. But perhaps Sam, Malcolm's model agnostic fictionalist, reveals how, on non-doxastic theories of faith, non-pretending agnostic fictionalists can have faith. Is that right?

I suspect not. For consider the peculiar motivations that Malcolm attributes to Sam. In particular, consider what motivates her acceptance of the BCS. Upon becoming an agnostic, she ‘determines that she will pursue her religion purely because it brings her great spiritual comfort’ (ibid., 7, emphasis added). Presumably that pursuit partly involves the acceptance Malcolm attributes to her. In the same vein, he writes that ‘all fictionalists’ ‘accept’ claims ‘because they regard them as good or beneficial in some way – because they evaluate them positively’ (ibid.). Regarding Sam in particular, he says that in ‘accepting’ the BCS, she ‘only take[s] into account practical considerations’ (ibid., 8, emphasis added).

It appears, then, that Sam's acceptance of the BCS is ‘purely’ due to her estimation of what practical benefits she can get out of accepting it; ‘only’ practical considerations motivate her acceptance. In that case, Sam fails to satisfy MPF's positive cognitive attitude condition, as indicated earlier. That's because Sam does not accept the BCS at least partly because of her estimation of the BCS's truth, and so it is not evidence-sensitive in the requisite way.

If evidence-insensitivity is partly definitive of the acceptance of Malcolm's agnostic fictionalist, then premise (3) of the Argument from Fictionalist Faith is surely true; and (2) is surely false. For even if faith that p requires a positive cognitive attitude towards p but not belief that p, it does not follow that agnostic fictionalists have faith that p. That's because the acceptance of agnostic fictionalists – as Malcolm describes Sam, at least initially – is evidence-insensitive, and so they fail to satisfy MPF's positive cognitive attitude condition on faith.

So we have another disambiguation of our title question: can non-pretending, evidence-insensitive, agnostic fictionalists have faith? Answer: not if faith is what MPF says it is.

In what follows, I will only consider the evidence-sensitive. In this I am encouraged by Malcolm, who reflects on the point that Sam's acceptance is evidence-insensitive and reports that this feature is not ‘essential’ to the agnostic fictionalist (ibid.). Suppose that's right. Furthermore – and this does not follow from evidence-sensitivity – let's grant that Sam's acceptance is a card-carrying positive cognitive attitude in every way. Under that description, does Sam show that non-doxasticists must grant that agnostic fictionalists can have faith?

Hardly. For Sam, as we've seen, ‘determines that she will pursue her religion purely because it brings her great spiritual comfort’ (ibid., 7 emphasis added). Since her pursuit includes her engaging her community's practices and discourse, and since, according to Malcolm, she engages them ‘purely because’ they bring her great spiritual comfort, it is false that she engages in them even partly because she accepts the BCS. But if it is false that she engages in them even partly because she accepts the BCS, then she fails to satisfy MPF's behaviour-motivational condition, according to which you have faith that p only if you are disposed to live in light of your positive cognitive attitude towards p. Thus, on MPF, the agnostic fictionalist, modelled on Malcolm's Sam, does not have faith, i.e. premise (3) of the Argument from Fictionalist Faith is true; and (2) is false for the expected reason.

So we have another disambiguation of our title question: can non-pretending, agnostic fictionalists who are not disposed to live in light of their positive cognitive attitudes nevertheless have faith? Answer: not if faith is what MPF says it is.

Let's grant that agnostic fictionalists can be non-pretenders; moreover, let's grant that they can be disposed to live in light of their positive cognitive attitudes. Under that description, does Sam show that non-doxasticists must grant that agnostic fictionalists can have faith?

Not at all. For notice that prior to becoming an agnostic, Sam had ‘a deep affection for God’. After becoming an agnostic, however, that seems to have gone the way of her ‘worship of God’: out the window (ibid., 8). Even so, Malcolm says, she ‘retain[s] her affection for the Christian story’ (ibid., 7). But notice what he compares that affection to: ‘her love for . . . stories that she enjoys reading, but knows are not true’. Suppose that is all there is to her conative-evaluative orientation towards the BCS. It's no different from her ‘love’ of Brothers Karamazov, Harry Potter and the Philosopher's Stone, The Lord of the Rings, Cutting for Stone, and the like.

In that case, she fails to satisfy MPF's second condition, according to which Sam has faith that the BCS is true only if she has a positive conative-evaluative orientation towards its truth. As we've seen, this consists in regarding the truth of the BCS as good or desirable and caring that it is true. Sam bears no such relations to the truth of the BCS. Just as she doesn't regard the truth of Brothers Karamazov as good or desirable, so she doesn't regard the truth of the BCS as good or desirable. Just as she doesn't care that Brothers Karamazov is true, so she doesn't care that the BCS is true. Moreover, if she lacks a positive conative-evaluative orientation towards the truth of the BCS, then, naturally enough, she won't be disposed to live in light of it. In that case, she also fails to satisfy MPF's behavioural-motivational condition. Once again, on MPF, the agnostic fictionalist, modelled on Sam, does not have faith, i.e. premise (3) of the Argument from Fictionalist Faith is true and, obviously enough, (2) is false.

So then, we have another disambiguation of our title question: can non-pretending, agnostic fictionalists who are disposed to live in light of their positive cognitive attitude towards the relevant propositions, but who lack a positive conative-evaluative orientation towards their truth, nevertheless have faith? Answer: not if faith is what MPF says it is.

Let us, therefore, grant that agnostic fictionalists can be non-pretenders, possess a positive cognitive attitude towards the relevant propositions and a positive conative-evaluative orientation towards their truth, and be disposed to live in light of that attitude and orientation. Under that description, must non-doxasticists grant that agnostic fictionalists can have faith?

Not at all. For even agnostic fictionalists under this description – people who are disposed to live in light of their positive cognitive attitude and positive conative-evaluative orientation – might be so disposed only when the going is easy. When the going gets tough, or even a wee bit sketchy, they have a tendency to throw in the towel. They lack any stamina, grit, tenacity, steadfastness, perseverance, or resilience in the face of even the smallest challenge to living in light of their attitude and orientation. They have no stomach. In that case, they fail to satisfy MPF's resiliency condition. Faith requires stomach: not necessarily ripped six-pack abs, but at least some abdominal tension that can provide minimal pushback. So: on MPF, agnostic fictionalists such as this cannot have faith, i.e. premise (3) of the Argument from Fictionalist Faith is true and, unsurprisingly, (2) is false.

So we have another disambiguation of our title question: can non-pretending, agnostic fictionalists who have positive cognitive attitudes towards the relevant propositions, as well as a positive conative-evaluative orientation towards their truth, and who are disposed to live in light of that attitude and orientation, but who lack resilience in the face challenges to living in light of them, nevertheless have faith? Answer: not if faith is what MPF says it is.

Let us grant that agnostic fictionalists can be non-pretenders; further, let's grant that they can have (i) a positive cognitive attitude towards the proposition that there is a God, etc., as well as (ii) a positive conative-evaluative orientation towards their truth; in addition, let's grant that they can be (iii) disposed to live in light of that attitude and orientation, and that they can possess (iv) at least some measure of resilience in the face of challenges to living in light of them. Under that description, must non-doxasticists grant that agnostic fictionalists can have faith?

Not at all. That's because some non-doxasticists may think that other features are constitutive of faith. These non-doxasticists will sing a familiar tune: although premise (3) is true, (2) is false.

But what about me? Must I concede that some agnostic fictionalists – notably, those who fall under the last description – can have faith? After all, they satisfy all of MPF's conditions.

By way of reply, I note that ‘agnostic fictionalist’ is a term of art that can be defined just about any way you wish, as we've seen. Suppose it is defined in such a way that the term applies to the people who fall under the last description, those who satisfy the conditions for faith, on MPF. In that case, we might well wonder: what exactly is the problem these ‘agnostic fictionalists’ are supposed to pose for non-doxasticism in general or MPF in particular? After all, Christian paradigms of faith – e.g. most recently Saint Teresa of Calcutta, but also Bartimaeus, the Syrophoenician woman, the woman with a haemorrhage, Jairus, the father of the demon-possessed son, and Jesus himself, in Gethsemane and Golgotha – arguably turn out to be agnostic fictionalists in the sense identified in the last description, as both McKaughan and I have proposed (Howard-Snyder (Reference Howard-Snyder2017a); McKaughan (Reference McKaughanforthcoming) ). But, even if St Teresa, Bartimaeus, Jesus, and so on aren't Christian paradigms of faith, from the point of view of at least some authors in the New Testament, they seem to be faithers nonetheless (Morgan (Reference Morgan2015), (Reference Morganforthcoming) ).

Malcolm's methodology

I want to close with a word or two about Malcolm's methodology. Doxasticists and non-doxasticists disagree over whether, necessarily, S has faith that p only if S believes that p. Who's right? Malcolm thinks doxasticists are right, as do many others. On behalf of his view, he offers the Argument from Fictionalist Faith. However, in the process of developing that argument, he tries to find a common core to non-doxastic theories of propositional faith. What he arrives at is this:

According to NDT, then, a person S has faith that p if and only if S has:

1. (1) a positive cognitive attitude towards p;

2. (2) no outright disbelief that p;

3. (3) a positive evaluative/affective attitude towards p. (Malcolm (Reference Malcolm2017), 4)

He then argues that, since these three conditions are sufficient for faith that p, and since a person can be an agnostic fictionalist while satisfying them (that's where Sam enters the picture), on non-doxastic theories of propositional faith, an agnostic fictionalist can have faith that, e.g., there is a God or that the BCS is true and the like.

I want to make two observations about Malcolm's methodology.

First a word about what he might have done, but did not do. Malcolm might have examined an exemplar of non-doxastic theories faith and explained how, on the conditions it laid down, something objectionable follows: an agnostic fictionalist can have faith. Then he might have argued that, like the exemplar, other extant non-doxastic theories of faith share those features in virtue of which that objectionable thing follows. If my theory had been the exemplar, then this alternative methodology would have had Malcolm actually quote my theory of faith as expressed in Howard-Snyder (Reference Howard-Snyder2013), (Reference Howard-Snyder2016), (Reference Howard-Snyder2017a), and (Reference Howard-Snyder2017b), all of which Malcolm had access to, explaining its key terms carefully, exhibiting all of the illustrations offered on its behalf, followed by a description of someone who (i) clearly satisfied all of the conditions laid down by MPF, where the terms of those conditions were understood with the meaning that I gave them, and (ii) clearly satisfied all of the conditions Malcolm laid down for being an agnostic fictionalist, where the terms of those conditions were understood with the meaning that he gave them. In effect, he would have begun with a narrative in which a fictionalist who – unlike his paradigm Sam – satisfied our last description above (‘Let us grant . . . ’). Why didn't Malcolm do that?

I don't know. But notice that if he had done it, it would hardly have been obvious that the result was a ‘fictionalist-style counterexample’ to non-doxastic theories of faith generally (Malcolm (Reference Malcolm2017), 2), or MPF in particular. Nor would it have been obvious that non-doxastic theories of faith, in general, or MPF in particular, faced something properly called ‘the problem of distinguishing faith from fictionalism’ (Reference Malcolm2017, 11). That's because it would hardly have been obvious that the person in his narrative lacked faith and, perhaps more tellingly, because it would hardly have been obvious that the person was a fictionalist in anything like the sense of the term used elsewhere in the philosophical literature. After all, when was the last time you saw a mathematical, modal, ontological, or moral fictionalist described as not ‘engaging in pretence’, and as (i) possessed of a positive cognitive attitude towards the proposition that, e.g., there are possible worlds or there are moral truths, taking a stand on behalf of their truth, (ii) possessed of a positive conative-evaluative orientation towards the truth of those propositions, i.e. regarding their truth as good or desirable and caring with positive valence that they are true, (iii) being disposed to live in light of that attitude and orientation, and (iv) being resilient in the face of challenges to living in that way, whether those challenges are evidential or non-evidential?

In this connection, consider Malcolm's take-home point: ‘fictionalism presents itself as a counterexample to NDT [non-doxastic theories of propositional faith] when NDT does not specify why someone adopts a positive cognitive attitude’ (Reference Malcolm2017, 16). Hold on! MPF does specify why: if you have faith that p, then you adopt a positive cognitive attitude towards p at least in part because of your estimation of the truth of p, a feature of MPF shared by other non-doxastic theories of faith, as a fair reading of Alston (Reference Alston, Jordan and Howard-Snyder1996) and Swinburne (Reference Swinburne1981/Reference Schellenberg2005) will reveal, contrary to what Malcolm says about those two theorists (Malcolm (Reference Malcolm2017), 13).

Second, a word or two about Malcolm's engagement with my theory. Malcolm says that, on my version of non-doxasticism, agnostic fictionalists do not have propositional faith. ‘But’, he says, ‘if [Howard-Snyder's] version of NDT is encapsulated by (1)–(3), and the fictionalist appears, prima facie, to satisfy (1)–(3), then the fictionalist does meet Howard-Snyder's NDT’ (ibid., 9). While Malcolm does not here flatly assert that my version of NDT is encapsulated by (1)–(3), in context he clearly insinuates that it is; moreover, the point of the Sam narrative is to exhibit how that is the case. However, what I hope to have shown is that, when it comes to faithfully representing real, live non-doxastic theories of faith, the details matter.

Not any old understanding of ‘a positive cognitive attitude towards p’ will do. Not any old understanding of ‘a positive evaluative/affective attitude towards p’ will do. If you want to engage real, live non-doxastic theories of faith, you need to attend carefully to the meaning of the terms of the theory assigned to them by its author, and not the meaning you attribute to those terms. Otherwise, you'll misrepresent the theory.

But that's not all. You also need to ensure that you do not omit conditions that a theorist specifies. In this connection, note that I explicitly look Malcolm's biconditional above in the eyes and say no: the right side is insufficient for the left (Howard-Snyder (Reference Howard-Snyder2013), 367–368, The Obama Objection; Idem (Reference Howard-Snyder2017a), 35–48). That's because you have faith that p only if you are resilient in the face of counterevidence to p or, more generally, only if you are resilient in the face of challenges to living in light of your overall positive stance towards p’s truth. Moreover, you have faith that p only if you are disposed to live in light of that stance (Howard-Snyder (Reference Howard-Snyder2017a), 57), and so in the case of faith that the BCS is true, your engagement with Christian practices and discourse will be at least partly grounded in your estimation of the truth of the BCS, contrary to Malcolm's description of the agnostic fictionalist.

To sum up: if you want to argue against non-doxasticism, focus on particular versions of it. And be sure to faithfully represent what a theorist means by the terms they employ; likewise, be sure to faithfully represent all of the conditions a theorist lays down. Anything less and you will fail in your noble goal to ‘faithfully represent this class of theories’ (2017, 3), and so you will mislead and misinform your readers.Footnote ¹