Epistemic attitudes and levels of belief

Revisiting core definitions after three years

The Death of Socrates, Jacques-Louis David, 1787.

It is three years today since my first post at Humble Knowledge, where I introduced epistemic attitudes like epistemic humility and confidence. While these concepts, along with epistemic certainty and skepticism, have been foundational to my writing, I'm not sure that they have been easy to understand. So, for this anniversary post, I thought it might be helpful to provide an alternate set of definitions. These are based on the concept of credence, or levels of belief, from my most recent post.

Firstly, as used in this context, an epistemic attitude is a philosophical disposition or assumption about the nature of knowledge. It describes our baseline expectations about what is possible and how knowledge should work. It isn't meant to capture our attitude towards any particular fact or situation, but rather our expectations about knowledge writ large.

In one of my early posts, I offered the following set of definitions:

The first is epistemic confidence, namely that humans can acquire knowledge, and wide ranging knowledge, with complete or at least considerable certainty. At its most confident, this can drift into epistemic certainty, the conviction that humans have (or at least the author has) achieved genuine and reliable knowledge (about certain topics) and no-one can reasonably doubt it.

In contrast is the attitude of epistemic humility, that genuine knowledge is difficult to acquire for everyone, even with the best of care and attention. A bolder, but related, conviction is epistemic skepticism: the belief that knowledge is not achievable and we genuinely know nothing (or very little).

I want to now offer some alternates based on the concept of a credence, introduced in my previous post. A credence is a measure of how strongly we believe a statement, typically rated on a scale of 0% (complete disbelief) to 100% (complete certainty). Using this rating scale as an indicative measure, we can explain these four epistemic attitudes:

1. Epistemic certainty assumes that the norm for human knowledge is a credence of 100%, that is, if we do the right thing we will achieve certainty. There may be practical reasons why we don't achieve it in some cases, but they occur when something goes wrong.

2. Epistemic confidence assumes that an achievable goal for human knowledge is a credence of 100%, but in practice we often fall short. The default expectation is that our credence in any knowledge should be 90+%.

3. Epistemic humility assumes that the achievable norm for human knowledge is a credence of around 60 - 80% and that, importantly, a credence of 100% is only possible in a few unique situations. In almost all situations, the maximum possible is ~99% because there always is a chance that we'll be wrong.

4. Epistemic skepticism assumes that a credence of over 40 or 50% is never justified and so we don't really know anything.

I’ll be interested in your comments on whether these definitions make the concepts clearer.

As a summary of the past three years, my broad argument here has been that we live in a culture that accepts epistemic certainty as a basic philosophical assumption, but this is mistaken as epistemic humility is not just desirable, but is an accurate description of human knowledge. This isn’t just an idle philosophical curiosity but has practical effects on a range of societal systems and actions, including on how we make decisions. More on this final point will come in the next post.