Skip to main content

The Nature of Degrees of Belief

Periodic Reporting for period 1 - Degrees of Belief (The Nature of Degrees of Belief)

Reporting period: 2017-02-01 to 2019-01-31

Ordinary humans have mental states capable of representing different ways the world might be; that is, they have representational content. Such states include our beliefs, which are a kind of all-or-nothing state: one either believes that P, or one does not. In recent decades, however, philosophers have increasingly emphasized that believing is not always an all-or-nothing matter—there are degrees of belief. For instance, one might be 99.99% confident that the sun will rise tomorrow, but only 80% confident that it will rain tomorrow. These are quantitative degrees of belief; they refer to specific levels of confidence, and are usually represented numerically. There are also qualitative degrees of belief. For example, one might be more confident that the sun will rise tomorrow than that it will rain tomorrow.

Talk of degrees of belief allows us to describe in a more nuanced way how humans represent the world, allowing us to distinguish between the relative likelihoods that individuals attach to the different claims that they consider. Doing so is central for our best accounts of decision making, which form the basis of a very significant amount of contemporary philosophical, psychological, and economic theory.

However, very little has been said about what degrees of belief are, exactly, nor how quantitative and qualitative degrees of belief relate to one another. Furthermore, to the extent that work has been done on this matter, the focus has been on the degrees of belief of unrealistically rational beings, quite unlike any ordinary human. This is unfortunate: without an adequate account of what degrees of belief are for ordinary agents, it is hard to settle many of the philosophical issues that decision theorists today engage upon.

The project’s overall aim is to use newly developed formal results to construct a novel approach to understanding the degrees of belief of ordinary, non-ideal human beings. More specifically, the project aims to bring together two historically distinct strands of thinking on the nature of degrees of belief: those who seek to make central use of so-called ‘representation theorems’ in their account, and those who hold that degrees of belief are psychologically real.
The primary research objectives for the project were to produce five research articles relating to the topic of what degrees of belief are, particularly as they appear in non-ideal agents, and ultimately aiming towards the development of a novel approach to answering this question. The researcher was successful in these aims. The researcher has submitted for review five research articles relating to the goals of this project. At the conclusion of the project, one of these articles had already been published, and a second had received a conditional acceptance. The final article includes a detailed discussion of the researcher’s novel approach to explaining what degrees of belief are, with aspects of the approach supported by the results of the earlier articles. In addition, the researcher organised and funded an international workshop on the nature of degrees of belief.
The researcher’s main scientific achievements during the course of this project includes (i) the development of a novel approach to degrees of belief, supported by formal results published by the researcher prior to the beginning of the project; (ii) novel philosophical arguments and positions relating to the connection between quantitative and qualitative degrees of belief; and (iii) advances in the formal modelling of unawareness and suspension of judgement, both important kinds of mental state that are closely related to understanding what degrees of belief are.
In all three of the above, progress has been made towards the much broader research goal—shared by philosophers, psychologists and economists alike—of developing a more adequate formal model of ordinary human agents’ degrees of belief and their relationship to decision making.
Researcher