Geertjan Holtrop

Strong Phenomenal Conservatism, as defended by Michael Huemer, says that having an appearance that p is a necessary and sufficient condition for having a justified belief that p in the absence of defeaters. I argue that the no-defeater condition is problematic in Strong Phenomenal Conservatism because it leads to the problem of mutual defeat due to conflicting appearances, a problem that becomes especially clear with optical illusions. I show that the problem of mutual defeat cannot be avoided by looking at differences in strengths of appearances, differences in quantity and quality of appearances, higher-order appearances, or unjustified beliefs. The only way the mutual defeat problem can be avoided is by relying on justified beliefs that are not ultimately based on appearances, but this means that Strong Phenomenal Conservatism is false. Therefore, phenomenal conservatives should opt for a weaker version of it.

We have many beliefs, yet we cannot immediately show sufficient supporting evidence for many of them, nor are we able to point to reliable sources from which many of our beliefs originate. Take for example beliefs such as that I am awake, that I like coffee, or that studying philosophy is not forbidden by law. Nonetheless, we hold on to such beliefs and we do not seem to be unreasonable for doing so. Epistemic conservatism can explain why this is so. This principle says that if a person, S, in fact believes a proposition, P, then S is epistemically justified in holding on to the belief that P unless there is something that defeats this justification. Although there a number of arguments in favour of conservatism –such as that it is cognitive efficient and explains our intuitions about lost evidence and memory beliefs, many philosophers think epistemic conservatism is implausible. Among them are Richard Foley, Hamid Vahid, David Christensen, Richard Feldman, and Matthew Frise. In this dissertation I have evaluated their objections against epistemic conservatism, subsequently formulating the following eight criteria for a plausible version thereof: Criterion 1: avoiding infinite regression A plausible principle of justification does not require a person S to believe an infinite number of propositions in order for a belief that P to be justified for S. Criterion 2: avoiding circularity A plausible principle of justification avoids circularity, such that the belief that P is not justified by a chain of reasons that is grounded in the belief that P itself. Criterion 3: avoiding arbitrariness A plausible principle of justification is able to explain the difference between a justified and an unjustified belief. Criterion 4: avoiding the fallacy ad ignorantiam A plausible principle of justification avoids arbitrariness as a consequence of the fallacy ad ignorantiam. This means that the principle explains the difference between passing judgment and suspending judgment. Criterion 5: avoiding uninterestingness A plausible principle of justification produces epistemically interesting results. This means that it produces justification of such nature and degree that it contributes sufficiently to attaining an epistemic goal. Criterion 6: avoiding backward reasoning A plausible principle of justification does not use the fact that a person already holds a belief as a reason to form that belief. Criterion 7: avoiding overly liberal defeaters A plausible principle of justification avoids overly liberal defeaters. This means that a version of conservatism that relies on a defeat condition that is satisfied for every belief a person has, is implausible. Criterion 8: avoiding evidential ignorance In a plausible principle of justification, evidence related to a belief is not ignored—evidence has a function in the principle. I have put various versions of conservatism to the test, concluding that the versions of Roderick Chisholm, Jonathan Kvanvig and Kevin McCain are either implausible or fundamentally failing to be a proper version of conservatism in the first place. Instead, I propose my own version, the No-Doubt Account of Conservatism, which I argue is plausible. It is important to note that epistemic conservatism is that is a principle of justified belief maintenance instead of a principle of justified belief formation. I argue that this does not necessarily mean that epistemic conservatism can only be of use in concert with another principle that provides for the justification of belief formation. I argue that given conservatism, it is plausible that a belief might be justifiably maintained even though it is not justifiably formed. Finally, I compare epistemic conservatism to phenomenal conservatism, in particular Michael Huemer’s version of it, dogmatism, and credulism –the latter two views having been espoused by James Pryor.

According to Epistemic Conservatism (EC), the fact that a person has a belief without defeaters is sufficient for assigning a positive epistemic status to that belief. An objection to EC is that it cannot be evaluated because it is impossible to imagine ‘bare beliefs’ in real life, i.e. there are no beliefs that have no connection to other beliefs, personal memory, or epistemic context. I argue that EC is not committed to the existence of ‘bare beliefs’ and that there is a different and better way to evaluate EC. Namely, by considering how well EC accommodates deep-seated epistemic intuitions about holding a belief. I argue that EC accommodates these intuitions well, also in comparison to other theories such as Common Sense epistemology, epistemic self-trust, epistemic entitlement, and phenomenal conservatism. Moreover, I give an explanation for the source of the positive epistemic status for beliefs on EC that involves considering the whole of our cognitive faculties. This also brings with it a new response to the ‘extra-boost objection’ against EC. I conclude that EC can be positively evaluated and can therefore be a useful epistemic theory.