Two Shapes of Pragmatism

1 Introduction

The question of what we should believe is central to the ethics of belief. This question can be broadened to other doxastic attitudes such as disbelief and suspension of judgment and it has been specified in the following ways: what is the rational doxastic attitude we should adopt? What are the reasons to adopt a specific doxastic attitude?

Evidence is a central concept in epistemology. It indicates the truth of some proposition. Evidence is typically considered as providing normative reasons for or against belief. A reason counts in favour of, or against, a doxastic attitude. Evidence for p speaks in favour of believing p, in the sense that it indicates the truth of the belief’s content. Furthermore, a reason helps determine whether our attitude is rational.^[1]

Evidentialism makes a further claim. According to it, evidence is the only source of reasons for or against belief.^[2] Evidentialists also claim that the rationality of one’s belief is determined by evidence only. On this view, non-evidential considerations do not contribute to the rationality of one’s doxastic attitude.^[3] In this paper, I will focus on the following two evidentialist theses:

Evidentialism faces two challenges of pragmatist inspiration. Both challenges state that practical considerations can be relevant to determine which doxastic attitude we should adopt. Those challengers can be called pragmatists. But pragmatism comes in two shapes.

According to an older version of pragmatism, lets dub it traditional pragmatism, practical considerations provide practical reasons for or against belief.^[4] As reasons are relevant to determine the rationality of the attitude, if there are practical reasons for or against belief, then the rationality of one’s belief is not fully determined by one’s evidence.

Pragmatic encroachment poses a different challenge to evidentialism.^[5] Encroachers claim that practical considerations affect some epistemic status such as epistemic rationality or knowledge.^[6] In Kvanvig’s slogan, the practical encroaches on the epistemic (2004). Whether a subject’s belief is epistemically rational or amounts to knowledge depends partly on practical considerations.^[7]

To have a first grasp on the distinction between both types of pragmatism, it is important to note that the central question of the ethics of belief is ambiguous between an epistemic, a practical, and an all-things-considered sense of “should”. As such, pragmatic encroachment is not directly opposed to the preceding evidentialist theses, but only to their epistemic counterparts:

According to pragmatic encroachment, practical considerations affect which doxastic attitude we should have by affecting which doxastic attitude is epistemically rational. As we shall see, traditional pragmatism does not think that epistemic rationality is the relevant type of rationality when it comes to decide which doxastic attitude we should adopt.

In the literature, the distinction between both pragmatisms is not always made. Philosophers rarely discuss them together. Occasionally, the distinction is mentioned but only to be set aside. And sometimes, it is simply overlooked.^[8] If the distinction is important for evidentialists which face two distinct pragmatist challenges, it matters for pragmatists too. There is an asymmetry in the pragmatist literature. If encroachers take care to distinguish their position from traditional pragmatism, traditional pragmatists do not discuss the encroachment claim.^[9] But as I see it, traditional pragmatism has good reasons to see pragmatic encroachment as an opponent and vice-versa.

This paper’s goals are to get clearer on the distinction between both types of pragmatism and to argue that they are independent. In fact, I will argue that traditional pragmatism has good reasons to reject pragmatic encroachment and vice versa. The paper is structured as follow. In Sections 2 and 3, i will present traditional pragmatism and pragmatic encroachment and develop their respective dialectic against evidentialism. Once the taxonomy is in place, I will argue, in Section 4, that if one accepts one version of pragmatism, one should reject the other.

2 Traditional Pragmatism

According to traditional pragmatism, the fact that having some doxastic attitude is beneficial or harmful is a genuine normative reason for or against that attitude. In this sense, the consequences of being in a certain doxastic attitude are relevant to determine whether we should have that attitude. Consider the following examples:

Athlete: You are a Roman athlete participating in the Olympics Games. Believing that you will win is likely to improve your performances.

Magic Potion: You are a Roman athlete participating in the Olympics Games. You have heard that one of your Gallic competitors has a magic potion that gives him invincible strength. Believing that he will win demoralizes you and negatively affects your athletic performances.

In both cases, you have a practical, more specifically prudential, consideration that counts in favour of believing (that you will perform well) or suspending judgment (as to whether your competitor will win).^[10] Even though those considerations are not truth-related, traditional pragmatists claim that they are genuine normative reasons for or against belief. They are relevant to determine whether one should believe, or suspend judgement about, the target propositions. In other words, practical reasons help rationalize one’s doxastic attitude.

There is disagreement among the proponents of this view. Pluralists, such as Marušić (2011) or Reisner (2008, 2009, claim that practical and epistemic reasons compete with one another to determine a general verdict about which doxastic attitude we should have.^[11] Both kind of reasons are normative reasons for or against belief. Consequently, one should believe what is rational to believe all-things-considered.

Other pragmatists make stronger claims. Robust pragmatists, such as Rinard (2015, 2019, claim that the only genuine normative reasons for or against belief are practical. On this picture, if there is a practical value in believing what the evidence supports, then evidence provides us with practical reasons for belief (Rinard 2015: 218). For such a robust pragmatism, one should believe what is practically rational to believe.

Pluralist and robust pragmatists quarrel about whether the reasons for or against belief are always practical. Pluralists deny it and recognize practical and epistemic reasons. Or to put the question more stringently, they quarrel about whether evidence is always a reason for belief. Robust pragmatists claim that if there is no practical value to believe p, the evidence for p’s truth does not provide us with any normative reason to believe that p.

Nevertheless, pluralist and robust pragmatists alike argue against evidentialism that practical considerations constitute genuine normative reasons for or against belief. We should believe what is practically or all-things-considered rational to believe. By contrast, evidentialism claims that the ethics of belief is only concerned with the question of what we should epistemically believe.

So, traditional pragmatism opposes Evidentialism about Rationality be rejecting Evidentialism about Reasons. But interestingly, traditional pragmatism, at least in its pluralist form, makes no claim about what determine epistemic status.^[12] As pluralists maintain the distinction between practical and epistemic reasons, the distinction between practical and epistemic rationality is preserved. Whether one’s belief is epistemically rational might well be determined by evidence alone.^[13]

3 Pragmatic Encroachment

According to pragmatic encroachment, practical factors affect epistemic rationality. Whether one should believe that p depends partly on some practical factors that determine the epistemic rationality of one’s belief. In agreement with the evidentialist, the encroachers may well accept that beliefs’ rationality is just epistemic rationality. Nevertheless, they claim that epistemic rationality is not fully determined by evidence. So, encroachers deny evidentialism by rejecting Evidentialism About Epistemic Rationality. As we shall see, some encroachers also deny Evidentialism about Epistemic Reasons.

To have a first grasp on the encroachment claim, consider the following pair of cases from Schroeder (2012a: 266–267):^[14]

Low Stakes: Hannah and her wife Sarah are driving home on a Friday afternoon. They plan to stop at the bank on the way home to deposit their paychecks. It is not important that they do so, as they have no impending bills. But as they drive past the bank, they notice that the lines inside are very long, as they often are on Friday afternoons. Hannah remembers the bank being open on Saturday morning a few weeks ago, so she says, “Fortunately, it will be open tomorrow, so we can just come back”. In fact, Hannah is right – the bank will be open on Saturday.

High Stakes: Hannah and her wife Sarah are driving home on a Friday afternoon. They plan to stop at the bank on the way home to deposit their paychecks. Since their mortgage payment is due one Sunday, they have very little in their account, and they are on the brink of foreclosure, it is very important that they deposit their paychecks by Saturday. But as they drive past the bank, they notice that the lines inside are very long, as they often are on Friday afternoons. Hannah remembers the bank being open on Saturday morning a few weeks ago, so she says, “Fortunately, it will be open tomorrow, so we can just come back”. In fact, Hannah is right – the bank will be open on Saturday.^[15]

Interestingly, High Stakes and Low Stakes do not differ in their epistemic features. In both cases, Hannah has the same evidence.^[16] The cases differ only in their practical features. In High Stakes, by contrast to Low Stakes, much depends on forming a false belief. Forming a false belief and acting on it will have terrible consequences.

According to pragmatic encroachment, the intuition behind such pair of cases is that in Low Stakes, it is epistemically rational for Hannah to believe that the bank will be open on Saturday, but it is not epistemically rational for her to believe it in High Stakes. In High Stakes, Hannah’s evidence is not enough for her belief to be rational. She needs more evidence and should inquire further before forming a belief and acting on it (Fantl and McGrath 2002: 68; Schroeder 2012a: 267).

As the two cases only differ in their practical features, encroachers claim that practical considerations affect epistemic rationality. They affect the amount of evidence needed for a belief to be epistemically rational. When the stakes are higher, more evidence is necessary. This gives us a way to characterize epistemic rationality. Epistemic rationality is a matter of being sufficiently supported by the evidence. The key word here is “sufficiently”. It makes this characterization of epistemic rationality different from the evidentialist one we considered so far:

Let me be clearer on this point. It is true that no evidentialist will deny what I dub the broad conception of epistemic rationality. That the wise person proportions her belief to her evidence is the evidentialist motto par excellence. But what does it mean to be sufficiently supported by the evidence? It is at this point that the encroachment claim breaks through.

Encroachers accuse evidentialists of not having explained the notion of “sufficient support” (Fantl and McGrath 2002: 71; Owens 2000: 25). According to this line of criticism, evidentialism does not provide us with a way to determine how much evidence is required for epistemic rationality. Indeed, evidentialists face the well-known “threshold problem” that is the problem to fix in a non-arbitrary way the amount of evidence required for a belief to be rational. It is argued that the evidence could not fix by itself the level of evidence that is sufficient for rational belief.

Pragmatic encroachment claims that practical considerations determine the relevant threshold (Fantl and McGrath 2002: 88; Owens 2002: 23). Encroachers posit a close link between rational belief and rational action. For example, according to Fantl and McGrath, the epistemic rationality of a subject’s belief depends on what is for her rational to do. In this sense, to believe that p is epistemically rational, only if it is rational to act as if p. Consider Hannah’s example. For her belief to be rational, it must be rational to act as if the bank will be open. In High Stakes, as it is not rational for her to act on this assumption, her belief is not epistemically rational (Fantl and McGrath 2002: 88). Hannah epistemically should not believe that the bank is open.

So according to pragmatic encroachment, practical considerations encroach on epistemic rationality by affecting the amounts of evidence required for one’s belief to be epistemically rational. This makes clear how practical stakes can bear on epistemic rationality: they raise the epistemic standard. Practical stakes have a different function than evidence. Evidence provides us with the epistemic reasons for or against belief, facts about stakes determine how much epistemic reasons is needed for a belief to be epistemically rational.

Pragmatic encroachment, or as is it called threshold pragmatism, challenges evidentialism because it states that epistemic rationality is not purely a matter of evidence. It is opposed to Evidentialism about Epistemic Rationality. A belief is epistemically rational just in case one’s evidence is good enough, but what is good enough is not itself a matter of evidence, it is determined by some practical considerations. Nevertheless, threshold pragmatism may accept Evidentialism about Epistemic Reasons: the only epistemic reasons are provided by the evidence.

There is another version of pragmatic encroachment that, for the sake of our distinctions, is worth having on our table. According to it, facts about stakes can affect the amount of evidence needed for a belief to be epistemically rational by providing epistemic reasons that compete against the epistemic reasons provided by the evidence.^[17] This version, lets dub it reasons-first encroachment, rejects Evidentialism about Epistemic Rationality by rejecting Evidentialism about Epistemic Reasons.^{[18] , [19]}

According to reasons-first encroachment, epistemic reasons are reasons that bear on epistemic rationality. Recall that a belief is epistemically rational iff it is sufficiently supported by the evidence. On this picture, evidence is clearly a source of epistemic reasons, but it may not be the only one. If a consideration bear on whether the evidence is good enough, then it also provides us with an epistemic reason (Schroeder 2012a: 272).

On a reason-first view, for a belief to be epistemically rational, the epistemic reasons for belief provided by the evidence must outweigh the epistemic reasons against belief. But belief has two epistemic competitors: disbelief and suspension of judgment. It is argued that the epistemic reasons to suspend judgment are not evidence. The evidence for p and not-p provides us with epistemic reasons to believe and to disbelieve that p. But these reasons are not epistemic reasons to suspend judgment (Lord fc: 16; Schroeder 2012a: 276).

What are then the non-evidential epistemic reasons to suspend judgment? Paradigmatically, the fact that one’s evidence is counterbalanced provides us with an epistemic reason to suspend judgment. This fact is not strictly speaking an evidence, it does not indicate the truth of some propositions.^[20] Schroeder also claims that facts about stakes provide non-evidential epistemic reasons to suspend judgment. Consider Hannah’s case: Hannah has more evidence for the proposition that the bank will be open, than for the contrary proposition. But she also has a reason to suspend judgment as to whether the bank will be open: the costs of believing falsely. According to Schroeder, the practical costs compete against the epistemic reasons to believe that the bank will be open. If the costs of error are high enough, the epistemic reasons to believe it are then not sufficient. As the evidence is not sufficient for her belief to be epistemically rational, Hannah should suspend judgment and look out for more evidence. In other words, her reason to suspend judgment outweighs her epistemic reasons to believe that the bank will be open. Consequently, facts about stakes provide us with epistemic reasons because they affect the amount of evidence needed for one’s belief to be epistemically rational.^[21]

To conclude on pragmatic encroachment, note that despite the challenge they pose to evidentialism, both versions of pragmatic encroachment are sympathetic to evidentialist intuitions. When it comes to rationality, they are rather conservative. They accept the broad conception of epistemic rationality: a belief is epistemically rational iff it is sufficiently supported by evidence. Still, they remain pragmatic in the sense that the sufficiency of evidence is determined by practical considerations whether these are taken to be reasons or not.

4 Traditional Pragmatism and Pragmatic Encroachment Discussion

Both pragmatisms argue that practical considerations are relevant to settle the central question of the ethics of belief: which doxastic attitude we should have. In this sense, they are opposed to evidentialist view about the rationality of belief. However, traditional pragmatism and pragmatic encroachment are independent. They do not oppose the same evidentialist theses. Traditional pragmatism is opposed to the claim that the only normative reasons for or against belief are evidence. But as pluralists maintain the distinction between practical and epistemic reasons, they could grant that epistemic rationality is determined by evidence alone. Still, they will deny that the fact that a belief is epistemically rational settles the question of whether we should believe it.

By contrast, encroachers deny that only evidence determines epistemic rationality. Practical considerations are relevant to determine whether the evidence is sufficient for one’s belief to be epistemically rational. Pragmatic encroachment is neutral about a practical sense, or an all-things-considered sense of “should”. They may claim that believing p because it is beneficial for the believer is in “some sense” rational (Schroeder 2012a: 268). But they are not concerned with this sense of rationality.^[22] Figure 1 summarizes the distinct pragmatist views.

Figure 1:

Two shapes of pragmatism.

Noting the independency of both pragmatisms is important. It means that rejecting one version of pragmatism does not settle the other pragmatist debate. Moreover, accepting one version does not commit one to accept the other. But, while each pragmatism can stay neutral toward one another, I will give some reasons to think that they should not. Each brand has good reasons to reject the other.

Let us consider traditional pragmatism against pragmatic encroachment first. As noted by Leary, traditional pragmatists have the conceptual resources to vindicate the encroachers intuition (Leary ms: 5). Subscribing to pragmatic encroachment to explain it will simply be redundant. In High Stakes, the practical risks generate a practical reason against belief. This practical reason affects what we practically or all-things-considered should believe. For example, pluralist pragmatists might claim that Hannah’s belief is epistemically rational, but that she still ought, all-things-considered, not to believe it.

Moreover, traditional pragmatism has a unified position on practical reasons: practical reasons bear on practical rationality or all-things-considered rationality. I see no point in complicating this story by distinguishing between those practical reasons and the practical reasons that bear only on epistemic rationality. As encroachers want to be moderate, in the sense that not every practical consideration bear on epistemic rationality, they bear the burden of distinguishing between the “right” practical considerations (i.e., bearing on epistemic rationality) and the “wrong” ones (i.e., bearing on another sense of rationality).^[23] Traditional pragmatists do not share this burden.

Should encroachers accept traditional pragmatism? Again, I think the answer is negative. As I see it, encroachers are sympathetic toward evidentialist intuitions shared by many philosophers involved in the ethics of belief. Indeed, pragmatic encroachment accepts the evidentialist motto: a belief is epistemically rational iff it is sufficiently supported by evidence. Pragmatic encroachment is also compatible with the claim that the only epistemic reasons for belief are evidence.^[24] The fact that pragmatic encroachment adhere to an evidentialist view of rationality makes it look attractive for those who share the intuitions.^[25] Encroachers might be well advised to accept the further claim that the ethics of belief is concerned with the question of what we epistemically should believe.

Finally, encroachment pragmatism provides us with an elegant explanation of the interaction between practical factors and evidence. On a traditional pragmatist view, it is hard to explain how to resolve conflicts between practical and epistemic reasons. What should we believe when both kinds of reasons give different verdicts? On the encroachment picture, the practical factors raise the epistemic standard. This gives us a way to arbitrate apparent conflicts: as long as the amount of evidence is not above the necessary threshold one should not believe that p. I’d like to expand on bit more on this point before concluding.

It is important to note that the two brands of pragmatism consider differently the interaction between practical and epistemic factors. On an encroachment view, practical and epistemic considerations interact to determine a unique verdict: an epistemic one. Practical stakes raise the epistemic standard. On the other hand, pluralist pragmatists claim that practical and epistemic reasons determine two distinct verdicts: a practical one and an epistemic one. Those verdicts can be in conflict (e.g., cases in which a belief that p is epistemically rational but practically irrational). Some pluralists give us a model to resolve such conflicts (Reisner 2008). On Reisner’s picture, when we have strong practical reasons not to believe that p, the epistemic reasons are “silenced”, i.e., they do not count anymore to determine an all-things-considered verdict (Reisner 2008: 22). But if, on the contrary, the practical reasons are weak, one must follow the epistemic verdict. On such a view, both kinds of reasons do not really collaborate. All-things-considered, one should always follow one kind of reasons.

It is not to say that encroachers have a way to avoid tragic doxastic conflicts.^[26] The thought is rather the following: how can the framework of traditional pragmatism explain the encroacher’s intuition? Recall that in the bank case, the intuition is to say that Hannah needs more evidence before she forms her belief in order to limit the risk of error. As I see it, the claim that Hannah should not believe that the bank is open because it is not practically rational falls a little short, especially if her belief is considered as being epistemically rational (as pluralists might claim).^[27] I wonder how pluralist pragmatists would explain that at some point, when Hannah get more evidence for p, the balance of her reasons will tip on the other side: it will be rational, all-things-considered, to believe that the bank is open. Reisner’s model will not be able to capture this intuition. In high stakes cases, the epistemic reasons are not silenced, they are simply not sufficient. Therefore, the traditional pragmatist debunking story needs some refinements if it is to capture the dynamics in the encroachment cases.

5 Conclusion

In this paper, I presented and distinguished two pragmatist views involved in the ethics of belief debate. Both argues against evidentialism that practical considerations are relevant to determine which doxastic attitude we should adopt. Nevertheless, I have shown that traditional pragmatism and pragmatic encroachment make distinct claims and are independent. Finally, I argued that even if both versions of pragmatism are neutral toward each other, they should not be. I gave some reasons to think that they should consider each other as opponents.