Well-Founded Belief

New Essays on the Epistemic Basing Relation

J. Adam Carter, Patrick Bondy, J. Adam Carter, Patrick Bondy

• Publisher: Routledge
• Year: 2019
• eBook ISBN: 9781351382434
• Edition: 1
• Topic: Philosophy
• Subtopic: Epistemology in Philosophy

Epistemological theories of knowledge and justification draw a crucial distinction between one's simply having good reasons for some belief and one's actually basing one's belief on good reasons. While the most natural kind of account of basing is causal in nature—a belief is based on a reason if and only if the belief is properly caused by the reason—there is hardly any widely accepted, counterexample-free account of the basing relation among contemporary epistemologists. Further inquiry into the nature of the basing relation is therefore of paramount importance for epistemology. Without an acceptable account of the basing relation, epistemological theories remain both crucially incomplete and vulnerable to errors that can arise when authors assume an implausible view of what it takes for beliefs to be held on the basis of reasons.

Well-Founded Belief brings together 16 essays written by leading epistemologists to explore this important topic in greater detail. The chapters in this collection are divided into two broad categories: (i) the nature of the basing relation; and (ii) basing and its applications. The chapters in the first section are concerned, principally, with positively characterizing the epistemic basing relation and criticizing extant accounts of it, including extant accounts of the relationship between epistemic basing and propositional and doxastic justification. The latter chapters connect epistemic basing with other topics of interest in epistemology as well as ethics, including: epistemic disjunctivism, epistemic injustice, agency, epistemic conservativism, epistemic grounding, epistemic genealogy, practical reasoning, and practical knowledge.

Part I
The Nature of the Basing Relation

1
A Doxastic-Causal Theory of Epistemic Basing

Ru Ye

1 Introduction

Suppose that two students, Anna and Barry, both have strong evidence for believing that they will get an A for the epistemology course they are taking, because both have done exceptionally well in their papers and exams. However, while Anna believes that she will get an A on the basis of that evidence, Barry comes to this belief because he hopes it is true. Intuitively, Anna’s belief is epistemically better than Barry’s, even though her evidence for the belief is no better than Barry’s.

What makes Anna’s belief better? Nowadays, a common answer is to distinguish doxastic justification from propositional justification. A popular way to characterize the distinction says that the former demands more than the latter: one’s belief is propositionally justified when one has good reason for the belief, and the belief is doxastically justified when it’s based on good reasons. So, Anna’s belief is better than Barry’s because it is not only propositionally justified but also doxastically justified.

Given its important role in the characterization of how doxastic justification differs from propositional justification, we need to get clear on what exactly the relation of epistemic basing involves. And this is not a task faced only by those who treat propositional justification as more fundamental. Those who think doxastic justification is more fundamental also have vested interest in getting clear on basing. For they will say that what makes a set of considerations R something that provides propositional justification to believe p is that, if one were to believe p on the basis of R, the belief would thereby be doxastically justified.¹

The importance of getting clear on basing goes beyond understanding doxastic justification. It’s also important for understanding reasoning or inference. For intuitively, one’s belief is a result of reasoning or inference from some premises only if one comes to hold the belief on the basis of the premises.²

I offer a new doxastic-causal theory of basing in this chapter, a theory that I call ‘Causation Caused by Believing’ (CCB). It says that one’s belief that p is based on reason R just in case R causes the belief and the causation happens because one believes that R supports p. I argue that this theory has two main virtues. First, it avoids the problem of deviant-causation, a problem plaguing simple causal theories of basing. Second, it implies a plausible theory of proper basing that can nicely account for the heatedly debated phenomenon of higher-order defeat.

Here is the plan. In Section 2, I defend a theory that I call ‘Causation Caused by Taking’ (CCT) as a stepping stone of CCB. CCT says that one’s belief is based on R when it is caused by R and when the causation happens because one takes R to support p. In Section 3, I argue that the taking condition should be understood as a belief about evidential connection. These two steps together constitute my argument for CCB. In Section 4, I answer some objections.

2 Avoiding Deviant Causation: Causation Caused by Taking

Although I am offering a doxastic-causal account of basing, I won’t say much to defend the ‘causal’ element in the account. That causation must be involved in basing is both intuitive and widely accepted.³ For basing is an explanatory relation. For me to hold a belief on the basis of some reason, that reason must explain why I hold the belief, and the most natural candidate of that explanation is a causal one. (See Turri (2011) for a more thorough defense of the causal element.)

Of course, mere causation is not enough for basing. There are many ways in which one’s belief can be caused by one’s reasons, and not all of them qualify as ways of basing. So, it becomes a vexed problem for causal theorists of basing to say which kind of causation qualifies as basing and which kind doesn’t, a problem known as ‘deviant causation.’ In Section 2.1, I criticize some current solutions to the problem in a way that would motivate my causal theory CCT. In Section 2.2, I explain CCT in greater detail.

2.1 Deviant Causation

Let’s consider the following two paradigm cases of deviant causation discussed in the literature of epistemic basing relation.

Late and Birds

I believe that I am going to be late to my class, and that causes me to run on a slippery sidewalk, lose my footing, and fall down, whereupon I find myself flat on my back looking up at the birds in the tree above me. I thereby believe that there are birds in the tree.

(Pollock and Cruz 1999, p. 36.)

Seeing and Hurting

Suddenly seeing Sylvia, I form the belief that I see her; as a result, I become rattled and drop my cup of tea, scalding my leg. I then form the belief that my leg hurts.

(Plantinga 1993, p. 69, fn. 8.)

In the first case, my belief that I am late causes my belief that there are birds in the tree, but intuitively I don’t hold the latter belief on the basis of the former. In the second case, my belief that I see Sylvia causes my belief that my leg hurts, but intuitively the latter belief is not based on the former.

It’s natural to think that the deviancy in causation has to do with how the causal path proceeds. In both cases, the causation is not direct but mediated through several stages of causation. And in both cases, the causal path involves elements that are entirely external to the subject’s cognitive system (running, losing footing, rattling, etc.). This may suggest that deviant causation happens when the causation is not direct, or when the causal path doesn’t happen entirely within one’s cognitive system (see Korcz 2000, p. 540 for the latter suggestion).

Both suggestions are problematic. They are simply too strong, given that the relevant conditions also obtain in intuitively non-deviant causation. Suppose I infer ‘p’ from ‘p and q,’ which is turn inferred from ‘p and q and r.’ Then my belief p is intuitively based on my belief ‘p and q and r,’ even though the causation is not direct, and even though the causation might involve purely physical activities happening in my brain, elements that are outside of my cognitive system: my belief ‘p and q’ might first cause some neurons firing, which in turn causes my belief p.

The thought that deviant causation must have to do with how the causal path proceeds also underlies the ‘causal-manifestation theory’ of basing recently proposed by Turri (2011, p. 393). According to the theory, deviant causation leading to your belief is causation that doesn’t manifest your cognitive disposition. And Turri takes a cognitive disposition to be a habit to form doxastic attitudes in certain circumstances, such as habitually taking experience at face value, habitually trusting testimony of others, or habitually reasoning in certain patterns. (In what follows, I will use ‘cognitive disposition’ and ‘cognitive habit’ interchangeably; it’s in line with Turri’s [2011, p. 391] own usage.)

This theory seems able to explain why the causation is deviant in Late and Birds and in Seeing and Hurting. In both cases, the causation is entirely accidental. It doesn’t manifest my cognitive habits: I simply don’t have the habit of believing that there are birds when I believe that I am late or the habit of believing that my leg hurts when I see Sylvia.

However, the causal-manifestation theory still doesn’t get to the heart of the problem. Consider the following variant of Late and Birds.

Late and Running

Joe believes that he is late for class. This causes him to run, which causes him to believe that he is running. And the causation chain is by no means accidental: He has a cognitive disposition to believe that he is running when he believes that he is late for class. This is because he has a habit to run when he believes that he is late for class, and he has a habit to believe that he is running when he is in fact running.

In this case, Joe’s belief that he is running is intuitively not based on his belief that he is late for class, even though the causation between the two beliefs is not accidental but is a manifestation of his cognitive disposition. (You might doubt whether Joe’s disposition to believe that he is running when he believes that he is late for class counts as a ‘cognitive’ disposition. But according to Turri, a cognitive disposition is just a habit to form beliefs in certain patterns, and by stipulation Joe does have a habit to form beliefs in the relevant pattern—he would believe that he is running whenever he believes that he is late for class. Or perhaps Turri will object that Joe’s disposition is not a cognitive one because it’s mediated through Joe’s disposition of running, a disposition outside of Joe’s cognitive system. But this objection will rule out too much. As I have explained earlier, mediation through purely physical activities in one’s brain can be involved in normal, non-deviant causation.) So, whether an instance of causation leading to a belief is deviant doesn’t seem to depend on whether the causation is accidental or is a manifestation of cognitive dispositions.⁴

2.2 How to Avoid Deviant Causation

What the previous diagnoses of deviant causation have in common is that they all focus on how the causation happens. They all focus on the causal paths and try to identify some features shared by those deviant causal paths. These attempts would fail because, as I see it, deviancy in causation doesn’t lie in how the causation happens, but in why it happens.⁵

Let’s consider some paradigm cases of non-deviant causation. I come to believe that I am mortal by inferring it from ‘All humans are mortal.’ Observing that the sun has risen every day in the past, I conclude that the sun will rise tomorrow. Receiving tons of witness testimony placing Jack at the murder scene, I come to believe that Jack is the murderer. Given clear visual experience of Tom in the library, I believe that Tom is in the library.

In these cases, the causal chains leading to my belief happen for a reason: they happen because I take the putative basis to support the proposition in question. When I believe that p on the basis of R, the causation between R and my belief is ‘sanctioned’ by my taking R to support p, and the causation wouldn’t happen without the taking. For instance, the reason why my belief that all humans are mortal causes my belief that I am mortal involves, at least partly, my taking ‘all humans are mortal’ to support ‘I am mortal.’

Note that the causation in the previous examples might be indirect because it might involve mediate steps of causation, and it might involve elements external to my cognitive systems such as pure physical brain activities, and it might be accidental in a sense because it doesn’t manifest my habits. But these factors are harmless, if the causation happens against the backdrop of my cognitive taking about the evidential support in question. In contrast, imagine such a case: I don’t take ‘all humans are mortal’ to support ‘I am mortal’ because I just don’t see the connection, and yet the causation still happens because of some glitches in my brain. In this case, the causation would be deviant given that it doesn’t happen because of my take on the evidential connection. Also, in Late and Birds, Seeing and Hurting, and Late and Running, the causation doesn’t happen because I take the putative basis to support the believed proposition.

To put the point another way, deviancy of a causal path isn’t located in how the causal path proceeds. Nothing about the causal path itself—whether it is direct, involves external elements, or is habitual—can suggest whether it’s deviant or not. When a reason R causes a belief p, the causation might be mediated by some purely physical brain activities both in deviant causation and in non-deviant causation. Nothing about those brain activities themselves marks the difference between deviant causation and non-deviant causation. What makes a stream of brain activities ‘a glitch’ instead of normal activities involved in basing is that the former doesn’t ha...

Table of contents

• Cover
• Half Title
• Series Page
• Title
• Copyright
• Contents
• Introduction
• PART I The Nature of the Basing Relation
• PART II Basing and Its Applications
• List of Contributors
• Index