1. The role of significance in Strategic Reliabilism

It is easy to despair of coming up with a notion of significance that might

be useful for epistemology. Take a reasoner in a context who is trying to

figure out what is the most significant problem she faces. Will a general

theory of significance provide useful information about this case? If it is an

easy case, the answer will be so obvious that a theory of significance will be

superfluous. If she is concentrating very hard on the mind-body problem

while driving up to a railroad crossing with the sound of a train in the distance,

we don’t need any fancy theories to tell us which problem is more

significant. In many hard cases, the answer will be nonobvious because the

reasoner doesn’t know enough about the idiosyncratic details of his situation

to know which problem is the most significant. For example, a

reasoner’s boss might ask him to develop a business plan; it might not be

clear whether he should be developing what he thinks is the best plan or

whether he should be developing what he thinks his boss thinks is the best

plan. In this case, no theory of significance is going to be useful because

what the reasoner lacks isn’t a sense of what problems are generally significant.

The reasoner lacks knowledge about what his boss really wants, and

that knowledge won’t be provided by a theory of significance. We might

despair because no significance-based epistemology is going to be able to

provide advice that is specific to a reasoner’s particular situation that is

both nonobvious and definitive. So one might reasonably conclude that

while significance is an important notion, there is nothing much useful for

an epistemological theory to say about it.

We endorse the premises of the above argument: Issues of significance

arise in particular cases. And an epistemological theory that includes an

account of significance will not be able to be applied directly and fruitfully

to many particular cases. In other words, we should not expect a recipe

book that will allow a reasoner to identify beforehand all and only the

significant problems that confront her. It does not follow, however, that

epistemology need not concern itself with significance. Let’s step back and

investigate the role significance ought to play in a normative epistemology.

The primary aim of epistemology, from our perspective, is to provide

useful, general advice about reasoning. Such advice is inevitably going to

depend on fairly general judgments of epistemic significance. These judgments

will typically involve claims about what kinds of problems are likely

to be significant for reasoners in general. For example, in their day-to-day

lives, people often rely on causal reasoning. We can avoid pain and misery

if we can accurately predict the causal outcomes of various actions—our

own, those of others, and those of nature. That’s not to say that every

reasoning problem that calls for causal reasoning is significant. But many

of the significant problems that people generally face involve causal reasoning.

As a result, a significance-based epistemology can recommend that we be prepared to expend a fair amount of time and energy improving our

reasoning about causal matters. This general truth about significance also

has important implications for the practice of epistemology. In filling out

the practical, prescriptive content of a significance-based epistemology, we

can focus our attention on uncovering biases to which people are most

prone in reasoning about significant matters. We can then offer well-tested

correctives to those biases, or we can suggest new, replacement strategies

for reasoning about those matters.

It is easy to despair of coming up with a notion of significance that might

be useful for epistemology. Take a reasoner in a context who is trying to

figure out what is the most significant problem she faces. Will a general

theory of significance provide useful information about this case? If it is an

easy case, the answer will be so obvious that a theory of significance will be

superfluous. If she is concentrating very hard on the mind-body problem

while driving up to a railroad crossing with the sound of a train in the distance,

we don’t need any fancy theories to tell us which problem is more

significant. In many hard cases, the answer will be nonobvious because the

reasoner doesn’t know enough about the idiosyncratic details of his situation

to know which problem is the most significant. For example, a

reasoner’s boss might ask him to develop a business plan; it might not be

clear whether he should be developing what he thinks is the best plan or

whether he should be developing what he thinks his boss thinks is the best

plan. In this case, no theory of significance is going to be useful because

what the reasoner lacks isn’t a sense of what problems are generally significant.

The reasoner lacks knowledge about what his boss really wants, and

that knowledge won’t be provided by a theory of significance. We might

despair because no significance-based epistemology is going to be able to

provide advice that is specific to a reasoner’s particular situation that is

both nonobvious and definitive. So one might reasonably conclude that

while significance is an important notion, there is nothing much useful for

an epistemological theory to say about it.

We endorse the premises of the above argument: Issues of significance

arise in particular cases. And an epistemological theory that includes an

account of significance will not be able to be applied directly and fruitfully

to many particular cases. In other words, we should not expect a recipe

book that will allow a reasoner to identify beforehand all and only the

significant problems that confront her. It does not follow, however, that

epistemology need not concern itself with significance. Let’s step back and

investigate the role significance ought to play in a normative epistemology.

The primary aim of epistemology, from our perspective, is to provide

useful, general advice about reasoning. Such advice is inevitably going to

depend on fairly general judgments of epistemic significance. These judgments

will typically involve claims about what kinds of problems are likely

to be significant for reasoners in general. For example, in their day-to-day

lives, people often rely on causal reasoning. We can avoid pain and misery

if we can accurately predict the causal outcomes of various actions—our

own, those of others, and those of nature. That’s not to say that every

reasoning problem that calls for causal reasoning is significant. But many

of the significant problems that people generally face involve causal reasoning.

As a result, a significance-based epistemology can recommend that we be prepared to expend a fair amount of time and energy improving our

reasoning about causal matters. This general truth about significance also

has important implications for the practice of epistemology. In filling out

the practical, prescriptive content of a significance-based epistemology, we

can focus our attention on uncovering biases to which people are most

prone in reasoning about significant matters. We can then offer well-tested

correctives to those biases, or we can suggest new, replacement strategies

for reasoning about those matters.