TagLast edit: 11 Jun 2021 8:19 UTC by EA Wiki assistant

Epistemology is the study of how people should form beliefs and credences about the nature of the world. Beliefs and credences are purely evaluative attitudes: they are simply about the way that we think the world is. A person might believe that it will rain, for example, even though they hope that it will not.

Beliefs are all-or-nothing attitudes: we either believe that it will rain or we don’t believe that it will rain. Credences, on the other hand, reflect how likely we think it is that something is true, expressed as a real number between 0 and 1. For example, we might think that there is a 80% chance that it will rain, and therefore have a credence of 0.8 that it will rain.

It is widely held that beliefs are rational if they are supported by our evidence. And credences are rational if they follow the probability axioms (e.g. a credence should never be greater than 1 in any event) and are revised in accordance with Bayes’ rule.

Improving the accuracy of beliefs

One way to improve a person’s capacity to do good is to increase the accuracy of their beliefs. Since people’s actions are determined by their desires and their beliefs, a person aiming to do good will generally do more good the more accurate their beliefs are.

Examples of belief-improving work include reading books, crafting arguments in moral philosophy, writing articles about important problems, and making scientific discoveries.

Two distinctions are relevant in this context. First, a person can build capacity by improving either their factual or their normative beliefs. Second, a person can build capacity by improving either particular beliefs or general processes of belief-formation.


Chignell, Andrew (2018) The ethics of belief, sect. 4, March 5, in Edward Zalta (ed.), The Stanford Encyclopedia of Philosophy.

Related entries

assessing and assimilating evidence | Bayesian epistemology | decision theory | epistemic humility | rationality

Defer­ence for Bayesians

Halstead13 Feb 2021 12:33 UTC
92 points
30 comments7 min readEA link

How mod­est should you be?

Halstead28 Dec 2020 17:47 UTC
23 points
10 comments7 min readEA link

[Link] The Op­ti­mizer’s Curse & Wrong-Way Reductions

Chris Smith4 Apr 2019 13:28 UTC
75 points
61 comments1 min readEA link

[Question] Is the cur­rent defi­ni­tion of EA not rep­re­sen­ta­tive of hits-based giv­ing?

Venkatesh26 Apr 2021 4:37 UTC
42 points
14 comments1 min readEA link

Se­quence think­ing vs. cluster thinking

GiveWell25 Jul 2016 10:43 UTC
4 points
0 comments28 min readEA link

[Link] EAF Re­search agenda: “Co­op­er­a­tion, Con­flict, and Trans­for­ma­tive Ar­tifi­cial In­tel­li­gence”

stefan.torges17 Jan 2020 13:28 UTC
63 points
0 comments1 min readEA link

Epistemic Trade: A quick proof sketch with one example

Linch11 May 2021 9:05 UTC
19 points
2 comments8 min readEA link

[Question] Mat­sés—Are lan­guages pro­vid­ing epistemic cer­tainty of state­ments not of the in­ter­est of the EA com­mu­nity?

mikbp8 Jun 2021 19:25 UTC
15 points
12 comments1 min readEA link

Against Modest Epistemology

EliezerYudkowsky14 Nov 2017 21:26 UTC
9 points
11 commentsEA link
No comments.