3. The potential unavailability of objective reasons

The significance of a problem is determined by the strength of the reasons

one has for devoting resources to it. But often people through no fault of

Epistemic Significance 99

their own don’t have access to those reasons. We often reasonably believe

that a problem is significant when it’s not. Trying to predict what gift a

spouse would enjoy for an anniversary might seem a fairly significant

problem, deserving to be pondered in considerable detail. But it’s probably

not if the spouse runs off with the neighbor a week before the anniversary.

Further, often people through no fault of their own will be correct about

whether a problem is significant but wrong about why. Stich offers the

example of the person who reasons to a true belief about when her plane

leaves but who doesn’t know the plane is doomed to crash (1990). The

problem of finding out when the flight left was significant, but not for

the reason she thought. The fact that the significance of particular problems

will sometimes (or perhaps regularly) be unavailable to reasoners

might seem like a serious problem for our view. We claim to be offering

a normative epistemological theory that provides reasoning guidance. A

central aspect of our theory is that reasoners should focus on significant

problems. But we admit that reasoners will often not know which are the

significant problems. So how can our theory offer people useful reasoning

advice?

Even though a reasoner might not have a good sense of what problems

are most significant for him to tackle, this does not undermine our

theory. First, any theory that takes significance to be important will have

this problem; and any theory that does not take significance to be important

will be incapable of making positive normative recommendations.

So this worry is an unfortunate feature of the human condition, not a

weakness of our view. Second, recall the role that significance is supposed

to play in our epistemological theory. It directs reasoners to be prepared to

spend more resources improving or replacing reasoning strategies that as a

general rule tend to have in their range significant reasoning problems. (It

might also direct reasoners to avoid spending resources on problems that

are negatively significant.) So the notion of significance plays a regulative

role in guiding the research of a prescriptive epistemology. A priority for

epistemology is to develop excellent reasoning strategies (i.e., reliable and

tractable for ordinary reasoners) that can be used on significant problems.

The fact that individual reasoners will sometimes be quite mistaken about

which problems facing them are the significant ones does not undermine

the epistemological project of recommending excellent reasoning strategies

(i.e., strategies that are robustly reliable, tractable, and applicable to

significant problems).

The call to allocate our cognitive resources to significant problems

places specific demands on the excellent reasoner. The most important demand concerns setting priorities. The priorities of the excellent reasoner

(and more generally, of the wise person) are set so that they may serve as a

means to human flourishing. Sometimes the excellent reasoner must replace

hot, spontaneous judgment with the cool administration of our

epistemic priorities. We might prioritize our projects so that they keep us

happily occupied. We might place a priority on family and friends because

their well-being matters to us. But when our interests are stable and

healthy, we don’t have to explicitly arrange these priorities—our interests

spontaneously direct us to significant problems. Our decision to have

children is more often the spontaneous result of a loving relationship than

it is the issue of a cold calculation that it will pay off in the long run. A

toned body can just as easily be the result of pleasant sport as it can be the

joyless consequence of scheduled maintenance. And like the beautiful dust

on a butterfly’s wings, these spontaneous interests result from natural ends

that subtly sculpt our lives. When determining the significance of the problems

we face, we should attend to these contours.

We should emphasize that a successful epistemological tradition will

not demand that the responsibility for reasoning excellence be shouldered

entirely by individuals. The well-ordered social presence of a reason-guiding

epistemology should promote the proper distribution of epistemic responsibility.

Institutions can make it more likely that individuals will act responsibly,

through for example, proper training, institutional procedures,

a well-designed system of incentives, or formal or informal sanctions.

The objection we are considering is an instance of a more general

worry about our theory. Strategic Reliabilism is a theory that sets forth the

conditions of reasoning excellence. This theory also holds out the promise

of an applied component, which will include reasoning advice we have

strong empirical reason to think is good reasoning advice. At the moment,

the practical content of Strategic Reliabilism is limited by the current state

of our well-tested, empirical knowledge about what sorts of reasoning

strategies are robustly reliable, tractable, and focused on significant matters.

Although limited, our view still recommends a number of specific strategies

that most people should adopt (e.g., frequency formats for diagnosis problems,

the consider-the-opposite strategy to counteract overconfidence, and

others to be discussed in chapter 9). But we cannot guarantee that people

will follow this advice—some will not follow our advice because they have

never been introduced to it, others because they decide to ignore it. But

these possibilities are no objection to our theory. Our theory provides

useful advice—but that doesn’t mean it provides advice that everyone

can always use no matter what. An analogy might be helpful. The owner’s manual for a car provides useful advice. It doesn’t follow that everyone

regardless of their skill or knowledge can use that advice profitably.

There is another aspect to the owner’s manual analogy. If there exists

only one copy of a Chevy Vega owner’s manual and it is locked in a vault

in Detroit, it is not available enough to be genuinely useful to Vega owners

(who are likely to need genuine help). Similarly, if the advice of Strategic

Reliabilism is to be restricted to highly specialized journals, then it will not

be available enough to be genuinely useful. That is why our view takes

seriously the idea that epistemology, like any science, ought to be a wellordered

social system (Kitcher 2001). A well-ordered social system for

epistemology would have at least two features. First, in order to achieve its

ameliorative potential, epistemology should be organized so that it provides

a way to effectively communicate its established findings, particularly

its practical advice, to appropriate audiences. Second, in order to minimize

the risk of promoting harmful or mistaken findings, epistemology

should be organized so that whatever findings are communicated widely

have passed rigorous empirical scrutiny.

Recognizing the importance of significance in epistemology opens up

a pair of empirical issues that are perhaps deserving of more study: First,

what sorts of problems are significant that people tend to think are not

significant and so perhaps reason poorly or not enough about? For example,

people tend to unduly discount the future, as when they overvalue

small current increments of money compared to their compounded value

in the future. And second, what sorts of problems are not significant (or

perhaps ‘‘negatively’’ significant) that people tend to believe are significant

and so perhaps spend too much time and energy on? For example, people

tend to unduly focus on vivid low probability risks at the expense of pallid

but much higher probability risks. Given the empirical nature of significance,

no theory can guarantee that significant problems are psychologically

available to us. The best our theory can do is to ensure that it

recommend strategies that will improve our reasoning about matters of

significance.

The significance of a problem is determined by the strength of the reasons

one has for devoting resources to it. But often people through no fault of

Epistemic Significance 99

their own don’t have access to those reasons. We often reasonably believe

that a problem is significant when it’s not. Trying to predict what gift a

spouse would enjoy for an anniversary might seem a fairly significant

problem, deserving to be pondered in considerable detail. But it’s probably

not if the spouse runs off with the neighbor a week before the anniversary.

Further, often people through no fault of their own will be correct about

whether a problem is significant but wrong about why. Stich offers the

example of the person who reasons to a true belief about when her plane

leaves but who doesn’t know the plane is doomed to crash (1990). The

problem of finding out when the flight left was significant, but not for

the reason she thought. The fact that the significance of particular problems

will sometimes (or perhaps regularly) be unavailable to reasoners

might seem like a serious problem for our view. We claim to be offering

a normative epistemological theory that provides reasoning guidance. A

central aspect of our theory is that reasoners should focus on significant

problems. But we admit that reasoners will often not know which are the

significant problems. So how can our theory offer people useful reasoning

advice?

Even though a reasoner might not have a good sense of what problems

are most significant for him to tackle, this does not undermine our

theory. First, any theory that takes significance to be important will have

this problem; and any theory that does not take significance to be important

will be incapable of making positive normative recommendations.

So this worry is an unfortunate feature of the human condition, not a

weakness of our view. Second, recall the role that significance is supposed

to play in our epistemological theory. It directs reasoners to be prepared to

spend more resources improving or replacing reasoning strategies that as a

general rule tend to have in their range significant reasoning problems. (It

might also direct reasoners to avoid spending resources on problems that

are negatively significant.) So the notion of significance plays a regulative

role in guiding the research of a prescriptive epistemology. A priority for

epistemology is to develop excellent reasoning strategies (i.e., reliable and

tractable for ordinary reasoners) that can be used on significant problems.

The fact that individual reasoners will sometimes be quite mistaken about

which problems facing them are the significant ones does not undermine

the epistemological project of recommending excellent reasoning strategies

(i.e., strategies that are robustly reliable, tractable, and applicable to

significant problems).

The call to allocate our cognitive resources to significant problems

places specific demands on the excellent reasoner. The most important demand concerns setting priorities. The priorities of the excellent reasoner

(and more generally, of the wise person) are set so that they may serve as a

means to human flourishing. Sometimes the excellent reasoner must replace

hot, spontaneous judgment with the cool administration of our

epistemic priorities. We might prioritize our projects so that they keep us

happily occupied. We might place a priority on family and friends because

their well-being matters to us. But when our interests are stable and

healthy, we don’t have to explicitly arrange these priorities—our interests

spontaneously direct us to significant problems. Our decision to have

children is more often the spontaneous result of a loving relationship than

it is the issue of a cold calculation that it will pay off in the long run. A

toned body can just as easily be the result of pleasant sport as it can be the

joyless consequence of scheduled maintenance. And like the beautiful dust

on a butterfly’s wings, these spontaneous interests result from natural ends

that subtly sculpt our lives. When determining the significance of the problems

we face, we should attend to these contours.

We should emphasize that a successful epistemological tradition will

not demand that the responsibility for reasoning excellence be shouldered

entirely by individuals. The well-ordered social presence of a reason-guiding

epistemology should promote the proper distribution of epistemic responsibility.

Institutions can make it more likely that individuals will act responsibly,

through for example, proper training, institutional procedures,

a well-designed system of incentives, or formal or informal sanctions.

The objection we are considering is an instance of a more general

worry about our theory. Strategic Reliabilism is a theory that sets forth the

conditions of reasoning excellence. This theory also holds out the promise

of an applied component, which will include reasoning advice we have

strong empirical reason to think is good reasoning advice. At the moment,

the practical content of Strategic Reliabilism is limited by the current state

of our well-tested, empirical knowledge about what sorts of reasoning

strategies are robustly reliable, tractable, and focused on significant matters.

Although limited, our view still recommends a number of specific strategies

that most people should adopt (e.g., frequency formats for diagnosis problems,

the consider-the-opposite strategy to counteract overconfidence, and

others to be discussed in chapter 9). But we cannot guarantee that people

will follow this advice—some will not follow our advice because they have

never been introduced to it, others because they decide to ignore it. But

these possibilities are no objection to our theory. Our theory provides

useful advice—but that doesn’t mean it provides advice that everyone

can always use no matter what. An analogy might be helpful. The owner’s manual for a car provides useful advice. It doesn’t follow that everyone

regardless of their skill or knowledge can use that advice profitably.

There is another aspect to the owner’s manual analogy. If there exists

only one copy of a Chevy Vega owner’s manual and it is locked in a vault

in Detroit, it is not available enough to be genuinely useful to Vega owners

(who are likely to need genuine help). Similarly, if the advice of Strategic

Reliabilism is to be restricted to highly specialized journals, then it will not

be available enough to be genuinely useful. That is why our view takes

seriously the idea that epistemology, like any science, ought to be a wellordered

social system (Kitcher 2001). A well-ordered social system for

epistemology would have at least two features. First, in order to achieve its

ameliorative potential, epistemology should be organized so that it provides

a way to effectively communicate its established findings, particularly

its practical advice, to appropriate audiences. Second, in order to minimize

the risk of promoting harmful or mistaken findings, epistemology

should be organized so that whatever findings are communicated widely

have passed rigorous empirical scrutiny.

Recognizing the importance of significance in epistemology opens up

a pair of empirical issues that are perhaps deserving of more study: First,

what sorts of problems are significant that people tend to think are not

significant and so perhaps reason poorly or not enough about? For example,

people tend to unduly discount the future, as when they overvalue

small current increments of money compared to their compounded value

in the future. And second, what sorts of problems are not significant (or

perhaps ‘‘negatively’’ significant) that people tend to believe are significant

and so perhaps spend too much time and energy on? For example, people

tend to unduly focus on vivid low probability risks at the expense of pallid

but much higher probability risks. Given the empirical nature of significance,

no theory can guarantee that significant problems are psychologically

available to us. The best our theory can do is to ensure that it

recommend strategies that will improve our reasoning about matters of

significance.