Comments on Daniel Whiting’s the range of reasons

Appeals to reason are central in human interaction. We ask for reasons, we provide reasons, and we respond (or fail to respond) to reasons. Reasons can explain, justify, and guide us. And they can do so regarding our actions, thoughts, and feelings. The weather report telling me a hurricane is touching down close by gives me a reason to shutter the windows, to believe that my house is in danger of being damaged, and to feel anxious about that impending damage. It may also seem that a hurricane’s pending arrival provides such reasons even if I am unaware that it is coming, having not seen the weather report. In The Range of Reasons, Daniel Whiting offers an original view of the nature of reasons that attempts to unify all the diverse roles that reasons play. Yet, his unifying project also attempts to preserve this diversity. As he puts it, a theme that runs through the book is “unity in diversity” (5). This book is ambitious, rigorous, and very carefully argued.

The central idea is that we should think of reasons modally. What makes a consideration a reason to x is its relationship to rightness across possible worlds. By de-coupling reasons from probabilities and from favoring, Whiting’s view can overcome a lot of objections launched at other theories of reasons while accounting for the intuitions that motivate them. Indeed, that is Whiting’s strategy for defending his view—to see how fruitful it can be to accept his theory insofar as it can accommodate most of the commitments of competitors while not exhibiting their weaknesses. I can imagine that someone who has an alternate view of how to understand the metaphysics of reasons may find a case that poses problems for Whiting’s view, or to argue that Whiting’s way of addressing an objection or worry is not convincing. I won’t be doing that here. My critical comments are going to focus primarily on Whiting’s discussion of rationality, and especially on his view of rational belief which is the topic of the book’s final chapter. My worry is that treating practical and epistemic rationality as completely distinct leads to an impoverished view of what it is to be a believer.

Before developing his own account of the nature of reasons, Whiting begins with a “job description” for reason, highlighting three features that seem central in the way we think about reasons. I have already alluded to the three important roles reasons play: “An important role reasons play is a justificatory one…A second key role reasons play is that of guiding action…The final role I will single out that reasons play is an explanatory one” (15). In explicating his view, the normative relation that reasons have to action and belief is what is most emphasized. That is, questions that relate, broadly, to reason’s justificatory role. In his discussion of how reasons can play this role, he distinguishes between justifying and demanding reasons. A reason stands in a justifying relation to x if it could not easily be wrong to x: “A justifying reason to act is a reliable indicator or safe evidence in which it is right to act” (64). A reason stands in a demanding relation to x if, by not performing x, a wrong is easily committed: “A demanding reason is a warning sign of a respect in which it is wrong not to act” (68). When Whiting first introduces his view, it is within the domain of objective reasons. There are facts that play certain roles, and this is the case whether or not someone has any access to them, or is in a position to know them.

When it comes to reasons’ explanatory role, subjective reasons need to be added to the picture. While justifying or demanding objective reasons can explain, they may not be sufficient. For example, in trying to explain Letitia’s crossing of the bridge, we can appeal to the fact that someone on the other side needed her help (37), but if she didn’t know or believe such a person was there, this would not be a good explanation. Subjective reasons do not only concern facts, they also take into account the subject’s perspective, and understanding their modal relationship to rightness requires the introduction of epistemically possible worlds in addition to the more familiar metaphysical ones. When introducing the idea of epistemic space, Whiting tells us “some proposition is epistemically possible for a person just in case they cannot know a priori that it is not true…So an epistemically possible world is a world which is not ruled out a priori, a world which is a priori coherent” (89). In explaining what is for an epistemically possible world to be near, Whiting employs the notion of similarity: “an epistemically possible world is near to the extent that it is similar in relevant respects to a person’s perspective” (91). With this idea in place, Whiting characterizes subjective reasons formally as follows:

Subjective: Necessarily, what appears to a person to be the case, P, is a subjective reason for them to act only if: (i) R is a respect in which it is right for them to act; (ii) in some nearby epistemically possible world in which P obtains, R obtains; (iii) they have the general ability to act for the reason that P (110).

Informally, and here distinguishing justifying and demanding, he says:

There is a subjective justifying reason for a person to act just when from their perspective, given what appears to them to be the case, doing so could not easily fail to be right in some way, say, that it keeps a promise. And there is a subjective demanding reason for a person to act just when from their perspective, given what appears to them to be the case, failing to do so could easily be wrong in some way, say, that it breaks a promise (111).

Of the three roles Whiting originally identified, guidance is what gets the least attention. Not everyone thinks reasons need to guide. Indeed, Whiting thinks that objective reasons often do not guide, but he wants to make room for the possibility that they can. It is this area, particularly concerning reasons’ role in deliberation, that is the focus of my concerns. One can only be guided by reasons one possesses. For Whiting, to possess a reason (to act) is “to be in a position to act in light of it or to be guided by it in action” (75). These are often what, in the literature on reasons, are called motivating reasons, and Whiting sometimes uses this terminology. When adapting this view to the epistemic realm, Whiting considers how one can believe in light of a reason. An asymmetry that Whiting often points out in his treatment of the practical and epistemic domains is that he takes “no stand… on the substantive question in normative ethics of what the right-making features of actions are, that is, in what respects it is right to act. But I do take a stand on the corresponding question in (what one might call) normative epistemology of what the right-making features of belief are, that is, in what respects it is right to believe a proposition” (123). I am not sure if he is entirely agnostic in the practical domain; the examples he provides in ways an action can be right always refer to benefits or violation of duties. It is unclear that his view would tolerate a first-order ethics of pure egoism. But he explicitly is committed to Truth, namely the view that “Necessarily, it is right for a person to believe a proposition if and only if that proposition is true.” He argues that this can account for all the other epistemic norms, such as knowledge, and offers a way of thinking of knowledge as justified true belief in a way that is not undermined by Gettier cases. Briefly, this is because a belief cannot be justified in cases where the belief could easily be false. The safety condition that is sometimes appealed to in theories of knowledge is already built into what it is to be a justifying reason.

A few times, Whiting brushes aside the idea that so-called non-epistemic reasons can be reasons for belief. He says, for example, that his focus is restricted to “epistemic norms for believing, as opposed to, say, moral, prudential or aesthetic norms for doing so. I will not here consider whether it is non-epistemically right or wrong to believe certain propositions, whether there are non-epistemic reasons for or against doing so, or whether there are things a person non-epistemically may, must, or ought to believe” (125). But when it comes to assessments of rationality, I don’t think this purity can hold. In his introduction of possessed reasons, Whiting only considers objective reasons, that is, facts that bear a certain modal relationship to rightness. But, as noted, he also wants to allow for the possibility of subjective reasons, which need not be facts but depend on a subjects’ perspective. Often when trying to explain why someone acts or believes as they do, these will be the kinds of reasons to which we appeal. Further, what makes an action or a belief rational, according to Whiting, is whether someone has a subjective reason for the action or belief. He is thus entirely internalist when it comes to rationality. Some theorists argue that only facts can be motivating reasons, not the appearance of facts. But again, Whiting is attempting to account for all the ways we employ reasons in our practice, and one of them does appeal to this subjective sense: “The propositions that provide subjective reasons can be false. So, if subjective reasons are possessed reasons, possessed reasons can be false. If a person possesses a reason, they have the opportunity to act in light of it. So, if possessed reasons can be false, motivating reasons—the reasons in light of which a person acts—can be false” (116).

When introducing the idea of “possessing” or “having” a reason, Whiting separates two ways such an idea is characterized. First as already mentioned, they are sometimes thought of in terms of guidance: “possessed reasons are capable of guiding a person’s actions; unpossessed reasons are not” and they are sometimes thought of in terms of rationality “Possessed reasons determine what it is rational for a person to do; unpossessed reasons do not” (99). Whiting separates his discussion of guidance and rationality. When concentrating on how reasons guide, he is primarily concerned with facts, namely possessed objective reasons (which is the nature of reasons referred to in the above quotations), and in trying to characterize what it means for its status as a reason to be apparent to a subject so that they can act in light of it. But in discussing rationality, it is possessed subjective reasons that matter, and these can be non factive. It seems such reasons can also guide; deliberation, which is tied closely to guidance, will often appeal to such reasons; I am not sure if Whiting meant to preclude subjective reasons from playing a guiding role.

When one acts in light of a subjective reason which is false, this can lead to having a false belief, but it can still be rational. Given Whiting’s commitment to Truth, it will not be right. It might irk some readers that rightness and rationality are divorced. I am on board with this division as I think some sense should be made of the idea that it can be rational to believe a falsehood. While, according to Whiting’s view, there cannot be justified false beliefs, (since justification is reserved for the relationship that holds between objective reasons and the belief), he makes room for this idea in what it can be rational to believe. So, for example, he argues that a “Moorean” belief such as “I believe it is raining, but it is not raining” can be rational. How so? It requires that someone take a kind of third-person alienated stance on their thoughts, and though this might violate some kind of normative standard, it does not violate the standard of what it is reasonable, or as Whiting says, “rational” to believe. I can look at the way that I am pulled in different directions when it comes to my belief. And if what I discover is that I am conflicted and divided, then having such a belief is rational. Whiting considers the “hackneyed example of a patient on a therapist’s couch who comes to think and thereby to know: I believe that my parents hate me, even though my parents do not hate me” (130).

The Moorean belief then can be “rational, even if the subject is thereby guaranteed to believe wrongly” (130). When we deem such beliefs rational, or reasonable, I think part of what is doing the rationalizing are non-epistemic considerations. How do I come to have beliefs about my mental states? What considerations lead me to believe that I am confused, conflicted, or anxious? A subjective epistemic reason for a person is one that, given how things appear to that person, that proposition could not easily be false (175). But one cannot rule out that some of the considerations that lead one to believe (and so take it to be true) are not themselves directly related to the likelihood of the proposition being true. To return to the therapist’s couch, discoveries (and perhaps knowledge) obtained there depend on non-epistemic factors such as trust in the therapist, attention to bodily sensations, or powerful emotions which arise in session. These sensations and emotions do not provide direct evidential support for the likelihood of any particular proposition. While the nature of trust is a complex and disputed topic, many argue that reasons to trust include the goodness of doing so. The trust I come to have in my therapist is not based on an assessment of their track record.¹

Whiting addresses this kind of concern when thinking about how his model can accommodate a typical way of thinking about deliberation. His view of “rational belief” is as follows “Necessarily, it is rational for a person to believe a proposition if and only if there is a subjective reason for them to do so” (179). It follows that “it is rational in some respect for a person to believe a proposition if and only it is rational overall for them to do so” (180). Whiting thinks a reader might worry that this view seems “to conflict with a very natural conception of deliberation: the weighing model. Reasoning as to whether some proposition is true, hence, whether to believe it, often seems to take this shape: On the one hand P, but on the other hand Q, and yet R, so S. This suggests that, in such reasoning, a person is weighing the reasons for and against the conclusion” (180).

His answer is that not every consideration in such deliberation counts as a reason. It can only be a reason, on his view, if it is one that relates to the likelihood of the proposition being true. But once in the realm of subjective reasons, how can he dictate what will make it seem, from the subject’s perspective what is more likely to be true? And how can bracket all the “non-epistemic” considerations that might show up in deliberation?

I will end with one more question on a different issue. An asymmetry in Whiting’s accounts of practical and epistemic reasons is that there are no demanding epistemic reasons. So one is never required or obligated to believe, only permitted, while there are obligations or duties in the practical realm. I have argued against purely epistemic duties as well from a different angle but I wonder if Whiting might concur with my reasoning. If the epistemic domain is completely separate from the practical, then it becomes unclear whether or to what degree its verdicts are substantially normative. I have argued if a flaw is purely epistemic, then one isn’t subject to blame in any sense. In the context of believing, blame is only appropriate, when the flaw extends into the broadly practical domain. If violations of duty make one subject to blame, then there are no purely epistemic duties.² For Whiting, the standard of rightness for belief is truth, and he equates this with correctness or fittingness. But correctness or fittingness conditions can be specified in many normative domains without carrying any kind of substantial normative force. A standard example is the domain of etiquette. Does believing rightly matter on Whiting’s account? Are the reasons we have to care about epistemic reasons themselves not epistemic reasons? Whiting has tried to unify the domains at the structural level through the standard of rightness while keeping them entirely distinct in terms of substance. My worry is that if one allows no unity at the substantial level, the epistemic realm is drained of substance.