The dissertation had two parts that fulfilled the goals described in my overview.
Part I: Objective Reasons, Rationality, and Justification
In Chapter 1 (“Reasons and the Metaphysics of Epistemology”), I defended a reasons-based account of justification that profited from ideas and distinctions in theorizing about reasons outside of epistemology. I argued that externalists in epistemology can and should hold a reasons-based account, and that the best general views about reasons (which I take from meta-ethics) strongly support an externalist---indeed reliabilist---epistemology. This chapter is crucial for the analogies with meta-ethics that underpin my justification/rationality distinction. But the chapter has independent interest: among other things, it shows that it is a mistake to think that the distinction between internalist and reliabilist theories of justification corresponds to the distinction between reasons-based and non-reasons-based theories.
In Chapter 2 (“What Apparent Reasons Appear to Be”), I laid the foundations for the other half of my distinction between justification and rationality. Many meta-ethicists understand rationality as a matter of correctly responding to apparent normative reasons, where apparent normative reasons are not necessarily real normative reasons. I am sympathetic to this general idea, but think that the accounts of apparent reasons that meta-ethicists have so far offered are flawed. In this chapter, I argued against them, and defended my own account. On my account, apparent normative reasons are apparent facts that we are competently disposed to treat like objective normative reasons, where the notion of competence is analyzed indirectly in terms of objective reasons and a competence/performance distinction is drawn. While my account allows that apparent reasons can easily fail to be objective reasons, it implies that the two are objectively likely to coincide when conditions for the exercise of competence are normal.
In Chapter 3 (“Rationality and Justification: Reasons to Divorce?”), I defended my distinction between rationality and justification. I also explained its implications for issues like the possibility of higher-order defeat, the normative impact of seemings with “checkered” etiologies, and several disagreements between internalists and externalists. On my view, rationality has a structural side and a substantive side, and both sides are separable from justification. Structural rationality consists in complying with the pressures of coherence exerted by our attitudes, which are hypothetical in the sense that one can rationally escape these pressures by dropping the relevant attitudes. Substantive rationality consists in heeding the apparent reasons, where these are understood along the lines suggested in Chapter 2. Unlike rational PHI-ing, justified PHI-ing requires one to possess an objective reason to PHI that is not outweighed by other objective reasons to PHI that one possesses.
Part II: How Does Rationality Matter from the Epistemic Point of View?
In Chapter 4 (“Veritism without Instrumentalism”), I defended a picture of epistemic value that proved crucial for my second goal. I argued that friends of the view that true belief is the fundamental epistemic value (“Veritism”) should reject the idea that all non-fundamental epistemic value is merely a species of instrumental epistemic value ("Instrumentalism"). Instrumentalism, I argued, is responsible for a generalized version of the swamping problem. Veritists can solve this problem if and only if they reject Instrumentalism. To make this concrete, I sketched an alternative version of Veritism that profits from an alternative model of value derivation. I took part of this model from Thomas Hurka's work on virtue. For a toy illustration outside of epistemology, consider beauty and its proper appreciation. Appreciating beauty is good because beauty is good. But that ‘because’ signals no instrumental explanation: appreciating beauty does not reliably cause more beauty to exist, nor is it the product of anything beauty-conducive. How is this type of value derivation relevant to epistemology? On my picture, there are certain ways of placing value on accuracy in thought that are manifested by rational belief, justified belief, and knowledge. These forms of belief are epistemically good because they manifest these ways of placing value on accuracy, and the ways of placing value on accuracy are, in turn, epistemically good because accuracy is epistemically good. But like the ‘because’ in the earlier example, these are not purely instrumental ‘because’s. This makes a crucial difference, I argued, because this form of value derivation is immune from swamping by the presence of the more fundamental value.
In Chapter 5 (“Rationality and Fundamental Epistemic Value”), I deepened the objection to the instrumentalist model of epistemic value by arguing that there is no satisfactory instrumentalist explanation of the epistemic value of rationality or rational beliefs. Instrumentalists must reject the idea that rationality has any necessary value from the epistemic point of view. Extending the arguments from Chapter 4, I argued that this is really a symptom of a more general problem, and that instrumentalists cannot even explain the epistemic value of forms of belief (e.g., justified belief) that are instrumentally connected to accuracy. Using the account developed in Chapter 4, I then showed that we can explain why rationality has necessary rather than merely contingent epistemic value if we reject the instrumentalist model. Indeed, I show that the account that we should give of why rationality matters from the epistemic point of view will have the same form as the account we should give of why justification and knowledge matter from the epistemic point of view.
The dissertation was successfully defended in December 2013 and I was awarded the PhD officially in the spring. I'd originally planned a sixth chapter, which became my new paper "Respect and the Reality of Apparent Reasons". Because I defended earlier than expected due to getting the post at Southampton, this was cut from the published version of the dissertation. But I regard this piece as perhaps the most important part of the project. So if you're interested in reading the whole dissertation, I would recommend reading this piece as if it were part of it.