Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)





First Advisor

Prof. ROWBOTTOM Darrell

Second Advisor

Prof. NADO Jennifer


In this thesis, I explore answers to three central questions: (i) what are beliefs, (ii) why do we have them, and (iii) how should we interpret doxastic correctness, the principle that it is correct to believe that p if and only if p? The first question has a long history in philosophy of mind, and in various forms can be dated at least as far back as David Hume. For that reason, I refer to the problem as Hume’s Problem. As I interpret the question, the main difficulty with accounting for what beliefs are, is in distinguishing them from forms of acceptance, where acceptances are understood as regarding-as-true attitudes—these include assuming, supposing, guessing, and (propositional) imagining. I take this technical use of ‘acceptance’ from Velleman (2000). The second question I interpret as a biological question about why we, as organisms with beliefs, have beliefs. Answering this question depends on how we answer the first question, in the sense that, in explaining why we have beliefs, we are explaining why we have an attitude that meets the conditions that we set for beliefs to meet. And the third question is largely about the important relation between belief and truth; why is it that, as believers, we emphasise the apparent platitude that a belief is correct if and only if it is true? To address these questions, I consider three different theoretical approaches to understanding beliefs. Specifically, I discuss teleological theories, normative theories, and a functional theory. For various reasons, I argue that teleological accounts fail to provide satisfactory answers to the three central questions (Part I); that normative accounts also fail to answer our central questions (Part II); but that a functional account, appropriately understood, can provide answers to these questions (Part III). In particular, I argue for what I call the doxastic effects thesis, which defines belief according to the effects beliefs have (or their outputs); and I propose that we interpret the components of this thesis as functional statements. The doxastic effects thesis allows us to answer Hume’s Problem, by proposing necessary and sufficient conditions for beliefs to meet; and interpreting these conditions as functions allows us to explain, at least in part, why we have beliefs. Concerning doxastic correctness, I argue that our commitment to the principle has arisen as a social construct, and therefore should be given (what I call) a thin reading; as opposed to the substantive reading that we get on the teleological and normative accounts, such that doxastic correctness states an essential fact about belief. Finally, in Part IV, I extend the first question, what beliefs are, to a third doxastic attitude. Namely, suspended belief (suspension). I argue that one further advantage of my functional theory of belief is that it can account for suspension as a doxastic attitude, unlike the teleological and normative alternatives.



Recommended Citation

Atkinson, C. J. (2018). On belief: Aims, norms, and functions (Doctoral thesis, Lingnan University, Hong Kong). Retrieved from

Included in

Philosophy Commons