Editing Belief (section)

=== Full and partial ===
An important dispute in formal epistemology concerns the question of whether beliefs should be conceptualized as ''full'' beliefs or as ''partial'' beliefs.<ref name="Genin">{{cite web |last1=Genin |first1=Konstantin |last2=Huber |first2=Franz |title=Formal Representations of Belief |url=https://plato.stanford.edu/entries/formal-belief/ |website=The Stanford Encyclopedia of Philosophy |publisher=Metaphysics Research Lab, Stanford University |access-date=1 April 2021 |date=2021 |archive-date=13 April 2021 |archive-url=https://web.archive.org/web/20210413164951/https://plato.stanford.edu/entries/formal-belief/ |url-status=live }}</ref> Full beliefs are all-or-nothing attitudes: either one has a belief in a proposition or one does not. This conception is sufficient to understand many belief ascriptions found in everyday language: for example, Pedro's belief that the Earth is bigger than the Moon. But some cases involving comparisons between beliefs are not easily captured through full beliefs alone: for example, that Pedro's belief that the Earth is bigger than the Moon is more certain than his belief that the Earth is bigger than Venus. Such cases are most naturally analyzed in terms of partial beliefs involving degrees of belief, so-called ''[[Credence (statistics)|credences]]''.<ref name="Genin"/><ref name="Olsson"/> The higher the degree of a belief, the more certain the believer is that the believed proposition is true.<ref name="Hartmann"/> This is usually formalized by numbers between 0 and 1: a degree of 1 represents an absolutely certain belief, a belief of 0 corresponds to an absolutely certain disbelief and all the numbers in between correspond to intermediate degrees of certainty. In the [[Bayesian epistemology|Bayesian approach]], these degrees are interpreted as [[subjective probabilities]]:<ref>{{cite web |last1=Hájek |first1=Alan |title=Interpretations of Probability: 3.3 The Subjective Interpretation |url=https://plato.stanford.edu/entries/probability-interpret/#SubPro |website=The Stanford Encyclopedia of Philosophy |publisher=Metaphysics Research Lab, Stanford University |access-date=6 March 2021 |date=2019 |archive-date=17 February 2021 |archive-url=https://web.archive.org/web/20210217013520/https://plato.stanford.edu/entries/probability-interpret/#SubPro |url-status=live }}</ref><ref name="Pettigrew">{{cite journal |last1=Pettigrew |first1=Richard |title=Précis of Accuracy and the Laws of Credence |journal=Philosophy and Phenomenological Research |date=2018 |volume=96 |issue=3 |pages=749–754 |doi=10.1111/phpr.12501 |url=https://philpapers.org/rec/PETPOA-3 |access-date=3 April 2021 |archive-date=1 June 2021 |archive-url=https://web.archive.org/web/20210601042948/https://philpapers.org/rec/PETPOA-3 |url-status=live |hdl=1983/d9f3e1c4-1bc9-4e04-b74c-dba4eb795393 |hdl-access=free }}</ref> e.g. a belief of degree 0.9 that it will rain tomorrow means that the agent thinks that the probability of rain tomorrow is 90%. Bayesianism uses this relation between beliefs and probability to define the norms of rationality in terms of the laws of probability.<ref name="Hartmann">{{cite book |last1=Hartmann |first1=Stephan |last2=Sprenger |first2=Jan |title=The Routledge Companion to Epistemology |date=2010 |publisher=London: Routledge |pages=609–620 |url=https://philpapers.org/rec/BOVSIO |chapter=Bayesian Epistemology |access-date=3 April 2021 |archive-date=16 May 2021 |archive-url=https://web.archive.org/web/20210516095047/https://philpapers.org/rec/BOVSIO |url-status=live }}</ref> This includes both synchronic laws about what one should believe at any moment and diachronic laws about how one should revise one's beliefs upon receiving new evidence.<ref name="Olsson">{{cite book |last1=Olsson |first1=Erik J. |title=Introduction to Formal Philosophy |date=2018 |publisher=Springer |pages=431–442 |url=https://philpapers.org/rec/OLSBE |chapter=Bayesian Epistemology |access-date=3 April 2021 |archive-date=16 May 2021 |archive-url=https://web.archive.org/web/20210516095057/https://philpapers.org/rec/OLSBE |url-status=live }}</ref><ref name="Hartmann"/>

The central question in the dispute between full and partial beliefs is whether these two types are really distinct types or whether one type can be explained in terms of the other.<ref name="Genin"/> One answer to this question is called the ''Lockean thesis''. It states that partial beliefs are basic and that full beliefs are to be conceived as partial beliefs above a certain threshold: for example, every belief above 0.9 is a full belief.<ref name="Genin"/><ref>{{cite journal |last1=Dorst |first1=Kevin |title=Lockeans Maximize Expected Accuracy |journal=Mind |date=2019 |volume=128 |issue=509 |pages=175–211 |doi=10.1093/mind/fzx028 |url=https://philpapers.org/rec/DORLME-2 |access-date=3 April 2021 |archive-date=20 October 2020 |archive-url=https://web.archive.org/web/20201020201413/https://philpapers.org/rec/DORLME-2 |url-status=live }}</ref><ref>{{cite journal |last1=Locke |first1=Dustin Troy |title=The Decision-Theoretic Lockean Thesis |journal=Inquiry: An Interdisciplinary Journal of Philosophy |date=2014 |volume=57 |issue=1 |pages=28–54 |doi=10.1080/0020174x.2013.858421 |s2cid=85521556 |url=https://philpapers.org/rec/LOCTDL |access-date=3 April 2021 |archive-date=5 July 2016 |archive-url=https://web.archive.org/web/20160705133728/http://philpapers.org/rec/LOCTDL |url-status=live }}</ref> Defenders of a primitive notion of full belief, on the other hand, have tried to explain partial beliefs as full beliefs about probabilities.<ref name="Genin"/> On this view, having a partial belief of degree 0.9 that it will rain tomorrow is the same as having a full belief that the probability of rain tomorrow is 90%. Another approach circumvents the notion of probability altogether and replaces degrees of belief with degrees of disposition to revise one's full belief.<ref name="Genin"/> From this perspective, both a belief of degree 0.6 and a belief of degree 0.9 may be seen as full beliefs. The difference between them is that the former belief can readily be changed upon receiving new evidence while the latter is more stable.<ref name="Genin"/>