Non-evidential believing and permissivism about evidence: a reply to Dan-Johan Eklund

Abstract

In response to John Bishop's (2007) account of passionally caused believing, Dan-Johan Eklund (2014) argues that conscious non-evidential believing is (conceptually) impossible, that is, it's (conceptually) impossible consciously to believe that p whilst acknowledging that the relevant evidence doesn't support p's being true, for it conflicts with belief being a truth-oriented attitude, or so he argues. In this article, we present Eklund's case against Bishop's account of passionally caused believing, and we argue that it's unpersuasive, at least to those who accept permissivism about evidence, that is, that it's possible for there to be more than one rational response to a given body of evidence. We do this through a novel application of a case of nurtured belief, that is, of a person holding a belief simply because she was caused to do so by her upbringing, and we use it to show exactly where Eklund's argument goes wrong. We conclude by drawing a general lesson drawn from this debate: if permissivism about evidence is true, then belief being truth-oriented is consistent with non-evidential believing being possible.

How could a person believe something without also believing that the evidence conclusively supports it? When the evidence is ambiguous and the belief is passionally caused, or so answers John Bishop (2007). Building on William James's remarks,

Our passional nature not only lawfully may, but must, decide an option between propositions, whenever it is a genuine option that cannot by its nature be decided on intellectual grounds. For to say, under such circumstances, ‘Do not decide, but leave the question open,’ is itself a passional decision – just like deciding yes or no – and is attended with the same risk of losing the truth. (James (1897/2011), 164; emphasis in the original)

Bishop argues that non-evidential believing is psychologically possible because

[t]he responsive attitude of holding a proposition true may be elicited by causes other than the believer's recognition, as such, of evidence for the belief's truth under the evidential practice assumed to be applicable (which, in the limiting case of a belief which that practice counts as properly basic, amounts simply to finding its truth basically evident in experience). (Bishop (2007), 114; emphasis in the original)

Adopting James's terminology, he takes these non-evidential causes to be passional causes, such as emotions, wishes, desires, affections, affiliations, and so on (ibid., 114–115).

To make the James/Bishop account of passionally caused believing in cases of evidential ambiguity vivid, consider the following example.

COMMUNITY: You grow up in a religious community. Everyone you know believes that God exists. You, too, believe that God exists, but it's not a belief you've consciously acquired, that is, you never sat down and asked yourself, ‘Does God exist?’ Rather, believing that God exists is just part of the culture you're a part of, and it's also a big part of your life: you go to church, say your prayers, and help out at the homeless shelter because you think that's what God wants you to do. In school, you study Aquinas's Five Ways and think they're pretty good arguments, and you study the problem of evil and think it's not such a problem after all. You then go to university and find out that a lot of people don't believe that God exists and think Aquinas's Five Ways aren't such good arguments and the problem of evil is a big problem. For the first time, you sit down and ask yourself, ‘Does God really exist?’ You come to think that the only reason you've (unconsciously) believed that he does, and found the arguments for his existence compelling and the arguments against his existence not compelling, is that you were born and raised in a particular community, with a particular culture and set of values; it could easily have been, you think, that you were born into an atheistic community, in which case you wouldn't have believed that God exists, wouldn't have gone to church, and wouldn't have said your prayers, though you think you would have helped out at the homeless shelter because that's still a good thing to do. With this thought in mind, you look at the evidence for and against God's existence, trying to see it from both sides, and find it to be inconclusive, that is, it leaves it open as to whether God exists or not, in the sense that it renders neither God's existence nor non-existence significantly more probable than not. You also find the evidence to be ambiguous, that is, it is open to two viable, competing interpretations, one on which God exists and the other on which he doesn't. In short, the evidence for God's existence is undecidable. What should you (consciously) believe now? ¹

Now Bishop observes that:

lack of evidential support for a proposition's truth does tend, once we become aware of it, to undermine any inclination we may have had to hold the proposition true – and that general tendency is no doubt central to proper cognitive functioning. To take that tendency for a universal psychological law, however, would be a rationalist fantasy – plausibly itself an example of passionally believing that things are as one thinks they ought to be! (ibid., 115; emphasis in the original)

So too, in the case of COMMUNITY, some may come to reduce their confidence in, or even to reject, the proposition that God exists. Others, though, may not. They may, rather, come to sustain consciously their belief that God exists and their confidence-level in that belief. This will be due, no doubt, to their character and temperament, formed in their upbringing, which once influenced their unconsciously acquiring a belief that God exists, and now influence their consciously sustaining that belief. All this, according to Bishop, is right and proper. For, as he writes,

To consciously believe that p, for some proposition p, is indeed to find oneself with the attitude towards p that it is true; but to find oneself with the attitude towards p that it is true is only typically but not necessarily to find p's truth evident or evidentially supported. To believe is, indeed, to believe true; it is not necessarily to believe evident. (ibid.; emphasis in the original)

Thus, because conscious believing does not require believing evident, non-evidential, conscious, passionally caused believing is possible.

According to Dan-Johan Eklund (2014), this account is conceptually confused. It's impossible, he thinks, to sustain consciously a passionally caused belief whilst at the same time believing that evidence to be undecidable, for it conflicts with belief being a truth-oriented attitude, or so he argues. In what follows, we present Eklund's case against Bishop's account of passionally caused believing, and we argue that it's unpersuasive, at least to those who accept permissivism about evidence, that is, that it's possible for there to be more than one rational response to a given body of evidence. We conclude by drawing a general lesson from this debate: if permissivism about evidence is true, then belief being truth-oriented is consistent with non-evidential believing being possible.

Eklund's argument that non-evidential believing is impossible

Now, Eklund doesn't doubt that beliefs can be acquired unconsciously by passional causes, as in the COMMUNITY case. However, he does doubt that once that belief becomes conscious, it can be sustained. So, for Eklund, in the COMMUNITY case, you should, or even must, as a conceptual point, be (consciously) agnostic, since you (consciously) think that the evidence doesn't tell in favour of either theism or atheism.

Building on a discussion between Bishop (2005) and Andrei Buckareff (2005), Eklund argues that the central problem for Bishop's account is that if (conscious) believing is believing true, as Bishop maintains, then ‘consciously believing that p for non-truth-related, passional reasons is conceptually troubling’ (Eklund (2014), 314). Thus, according to Eklund, consciously believing requires believing evidence, in which case sustaining a non-evidential, passionally caused belief consciously isn't possible. He sums up his argument for this claim as follows:

Suppose that my evidence for p is inconclusive and that I have in less than full consciousness acquired the belief that p by a passional cause. Suppose then that I happen to reflect consciously [on] this passionally caused belief. But now I see that I believe irrespective of what seems to me to be the truth of p, since I acknowledge that passional causes do not indicate that p is true and that my evidence for p's truth is inconclusive. But believing irrespective of what seems to be the case is inconsistent with belief's truth-oriented nature, and this is something I also realize. Thus, I do not and cannot believe that p anymore. So, consciously to sustain passionally caused belief is not only psychologically peculiar, but it involves a conceptual inconsistency. (ibid., 315–316)

In order to evaluate the argument, we'll take a concrete example of it, reconstructing it and applying it to the COMMUNITY case:

1. 1. For reductio, say that, in the COMMUNITY case, after reflection, you come to believe consciously that God exists, sustaining that belief acquired unconsciously during your upbringing, even though you now think that the evidence for God's existence is undecidable, that is, it is both inconclusive and ambiguous.

2. 2. If you consciously believe that God exists and, at the same time, believe that the evidence for God's existence is undecidable, then you consciously believe that God exists irrespective of whether it seems to you that God exists.

3. 3. But if you believe that God exists irrespective of whether it seems to you that God exists, then you're believing that God exists irrespective of whether you believe it's true that God exists.

4. 4. Since, belief is truth-oriented, that is, believing requires believing true, it's impossible to believe that God exists irrespective of whether you believe it's true that God exists.

   So, in the COMMUNITY case, you can't come to believe consciously that God exists and, at the same time, think the evidence for God's existence is undecidable.

In what follows, we argue that premise (2) is false. ² To do that, we have to discuss first permissivism about evidence, in general, and Bishop's particular instance of it, namely, his ‘thesis of the evidential ambiguity of theism’, or ‘the ambiguity thesis’, for short (Bishop (2007), 70).

Permissivism about evidence and non-evidential believing

Permissivism about evidence is the view that there is more than one way to respond rationally to a given body of evidence. ³ Bishop assumes a particular version of this view in his ambiguity thesis, as we term it, according to which

the evidence [for the classical theistic God] is ‘open’ in the sense that it neither shows the truth of the claim that God exists nor the truth of its denial to be significantly more probable than not. The thesis further describes this situation of open evidence as ‘ambiguity’ by making the claim that the total available evidence is systematically open to two viable competing interpretations – in a sense of ‘viable’ that is hard to make fully precise, but may be compared by analogy to the sense in which the drawing of the duck-rabbit is open to two viable perceptual Gestalts. (ibid., 71)

As evidence for the ambiguity thesis, Bishop observes that equally intelligent, well-informed people continue to debate, after many centuries, whether the evidence does or doesn't support God's (non-)existence, and he also proposes that the familiar arguments for and against God's existence suffer from epistemic circularity. Now, we take permissivism about evidence, in general, and the ambiguity thesis, in particular, to be plausible. But we won't argue for these claims, for that would take us too far afield. Rather, we will argue that, if they are true, premise (2) is false.

If permissivism about evidence is true, then there are permissive cases, that is, cases where there's more than one way to respond rationally to a given body of evidence. If the ambiguity thesis is true, the COMMUNITY case is one such case. Now, in the COMMUNITY case, when you were growing up, not only did you acquire unconsciously your belief that God exists, but also you acquired your standards for reasoning, weighing evidence, and making judgements, where (i) belief that God exists is consistent with those standards and (ii) finding the arguments for God's existence compelling and objections to his existence not compelling are consistent with that way of reasoning, weighing evidence, and making judgements. Consequently, your unconscious judgement that God exists is rational relative to those standards for reasoning and weighing evidence. ⁴

In metaphorical terms, you have a way of seeing the world, given to you by these standards of reasoning, weighing evidence, and making judgements, which you acquired unconsciously, and the (unconscious) judgement that results is how the world (unconsciously) seems to you, in this case, that God exists. You then encounter others who don't see the world in the same way that you do, that is, they have different standards for reasoning, weighing evidence, and making judgements, and on that way of seeing the world, it seems to them that God doesn't exist. You now see that there are two ways of seeing the world and two ways the world could seem, two perceptual Gestalts, as it were. Considering the others' way of seeing the world and how the world seems to them alongside your way of seeing the world and how the world seems to you, it seems to you that these two ways of seeing and how the world could seem, seem equally reasonable. Now, following this encounter, you will either continue to see the world in the same way as you did before the encounter, and the world will seem the same to you as it did then, that is, that God exists, or you will come to see the world in a new way where the world seems different to you than it did before the encounter, that is, that God doesn't exist. In either case, premise (2) of Eklund's argument is false. For convenience, say that you continue to see the world in the same way as you did before the encounter:

• • You consciously believe that God exists, and, at the same time, you believe that the evidence for God's existence is undecidable.

• • But it's not the case that you consciously believe that God exists irrespective of whether it seems to you that God exists, for it does seem to you that God exists based on how you see the world, that is, your standards of reasoning, weighing evidence, and making judgements about the world.

A general moral can be drawn from the above debate. To summarize: both Bishop and Eklund agree that belief is truth-oriented. Where they disagree is over whether non-evidential believing is possible. Our reconstruction of Eklund's argument against Bishop and our reply on behalf of Bishop shows that Bishop can maintain both belief being truth-oriented and the possibility of non-evidential believing by relying on his ambiguity thesis, or, more generally, that permissivism about evidence is true. Thus, generalizing from the Bishop/Eklund debate, we can see that, if permissivism about evidence is true, then belief being truth-oriented is consistent with non-evidential believing being possible.

Passionally caused believing and doxastic voluntarism

Perhaps the strongest objection to our way of reconciling belief being truth-oriented with non-evidential believing is that it relies on doxastic voluntarism. In essence, that permissivism about evidence entails that it's possible consciously to decide to believe, or not to believe, some proposition. Since it's plainly obvious to some, including both Bishop and Eklund, that beliefs can't be directly willed, it's important to show how our account isn't committed to doxastic voluntarism.

To begin, it's worth noting that some of the arguments against doxastic voluntarism turn on a conceptual connection between something being under some sort of rational control and its being caused by factors that logically connect the world and the content of the belief. However, this causal analysis will be resisted by some, especially those inclined to think that reasons aren't causes (see, for example, Alvarez (2010)). Commitment to this connection will thus rule some forms of doxastic voluntarism out of court from the outset, and exploration of non-causal analyses of action (especially decision) in the context of belief formation may open up possibilities for doxastic voluntarism, wherein willing is not a causal notion, that are immune to the kinds of attack proposed by Eklund.

Nevertheless, there are some reasons to suspect that James's thesis, as it's postulated in ‘The will to believe’ (1897/2011), assumes the possibility of some form of doxastic voluntarism. What begins to generate the problem in James's account is the ambiguity of the word ‘decide’ in his defence of passionally caused beliefs. In one sense, ‘decide’ means the involuntaristic ‘determine’ and, in another sense, it means the voluntaristic ‘choose’. On our reading of James, our passional natures operate involuntarily and determine, say, our faith-beliefs; following this determination, we then choose whether or not to act on these beliefs in our practical reason. One reading of what James means by ‘deciding to believe’, then, is that we decide (voluntarily) to use our faith-beliefs, determined (involuntarily), in part, by our passional natures, in practical reason. Now, it's clear that James's account, at least on our reading, isn't committed to doxastic voluntarism, since, on his account, we're not choosing to believe anything, just choosing whether or not to act on our belief, which may have been determined by our passional natures. We may have some control over our passional natures, though, if we choose to develop them in one way or other, e.g. by going to church, but that sort of indirect doxastic voluntarism isn't what's up for discussion here. Bishop then goes on to develop James's account in his own work, but we now argue that it, too, doesn't depend on doxastic voluntarism.

According to Bishop, in explaining the nature of belief, it's important to make the distinction between ‘holding true’ and ‘taking to be true’ (Bishop (2007), 33). Beliefs, Bishop maintains, are responsive attitudes towards propositions which ‘are neither formed nor revised under the direct control of the will’ (ibid., 30); or in other words, Bishop explicitly rejects doxastic voluntarism. He does, however, admit that there is some level of doxastic control that a believer can have over her beliefs. Believing involves, Bishop thinks, both ‘holding true and taking to be true in reasoning’ (ibid., 34, emphasis in the original). To consider the difference between these two aspects of belief, consider his example:

Mary believes her pet tortoise is liable to roam, so, while she shows it off to her guests during tea on the lawn, she keeps a wary eye on it so as to avoid lengthy searching in the undergrowth at the bottom of the garden . . . Here Mary holds true the proposition that her tortoise can hide itself surprisingly quickly, and – now that she has set her tortoise at large – this belief becomes salient given her intention not to lose it. Through an effortless piece of practical reasoning in which, inter alia, she takes this proposition about the tortoise to be true there results Mary's action in keeping a close eye on it. (ibid.; emphasis in the original)

Taking to be true and holding true are closely connected – Mary only takes her belief to be true in her practical reasoning because she holds the belief to be true. On the one hand, holding to be true is a state towards a relevant proposition which is not under the direct control of the will – Mary cannot force herself to believe that tortoises are not liable to roam, for example. On the other hand, however, taking a proposition to be true is a mental event which requires deliberative reasoning. It's possible in some circumstances, Bishop thinks, voluntarily to do otherwise than to take a belief to be true even when we already hold it true. For example, I might be convinced that a certain piece of medical advice is correct, whilst in practice, I refuse to follow this advice, or I might simply ‘hedge my bets’ regarding it (ibid., 37). So whilst it's impossible to will to hold a proposition to be true (the rejection of doxastic voluntarism), in some cases, it's up to us what we do with those beliefs which we hold to be true.

To return to our argument, it might seem that we invoke some form of doxastic voluntarism when we say,

Considering the others' way of seeing the world and how the world seems to them alongside your way of seeing the world and how the world seems to you, it seems to you that these two ways of seeing and how the world could seem, seem equally reasonable. Now, following this encounter, you will either continue to see the world in the same way as you did before the encounter, and the world will seem the same to you as it did then, that is, that God exists, or you will come to see the world in a new way where the world seems different to you than it did before the encounter, that is, that God doesn't exist.

But this isn't the case. Neither Bishop nor we are committed to the view that continuing to see the world in the same way as you did before the encounter or coming to see the world in a new way is a voluntary matter, though you may be aware that you have non-voluntarily come to see the world in a particular way. Rather, in order to take a proposition to be true, it's necessary that you already have the appropriate motivations (i.e. that you already hold the proposition true). From your motivations, you will maintain your current standards for reasoning weighing evidence, and making judgements that sustain your belief, or you will come to have different ones that reject it. And, since we already know that this is a case in which the evidence is ambiguous, these motivations can only be passional. Taking a belief to be true, then, is not a ‘wilful leap’, Bishop argues, but ‘a motivated choice to take to be true what one holds through causes that one recognizes oneself to be non-evidential’ (Bishop (2007), 117).

Given that you already believe, based on passional causes, adopting standards that would sustain your belief emphatically doesn't then amount to self-induced believing, for you already do believe (ibid.). Rather, the question is: given that the evidence is ambiguous, are you motivated to sustain your belief or not? The important point, then, is that it's perfectly possible to continue operating, now consciously, with the standards of reasoning, weighing evidence, and making judgements that you acquired unconsciously when you were growing up on which God exists. And so it's perfectly possible to sustain consciously (or take to be true, to use Bishop's terminology) your unconsciously acquired, passionally caused belief that God exists, whilst at the same time believing the evidence to be undecidable. In such a case, it does seem to you that God exists, because that's the way you're motivated to see the world, metaphorically speaking. That is, you are motivated to sustain your belief. To use Bishop's duck-rabbit example, your motivations have resulted in your seeing a rabbit (i.e. believing that God exists), even though you're perfectly aware that you could be seeing a duck (i.e. believing that God doesn't exist), and that others do so, as a result of their own motivations. Thus, if the ambiguity thesis is true, premise (2) in the reconstructed form of Eklund's argument is false, since it does seem to you that God exists, even though you acknowledge that the evidence is undecidable. ⁵