The idea that the concept ‘knowledge’ has a distinctive function or social role is increasingly influential within contemporary epistemology. Perhaps the best-known account of the function of ‘knowledge’ is that developed in Edward Craig's Knowledge and the State of Nature (1990, Oxford University Press), on which (roughly) ‘knowledge’ has the function of identifying good informants. Craig's account of the function of ‘knowledge’ has been appealed to in support of a variety of views, and in this paper I'm concerned with the claim that it supports a sort of epistemic contextualism, which is (roughly) the view that the semantic contents and truth-conditions of ‘knowledge’ ascriptions – instances of ‘S knows that p’ – depend on and vary with the context of ascription (see, for instance, John Greco's ‘What's wrong with contextualism’, Philosophical Quarterly [2008]). Prima facie, this claim should strike us as surprising. A number of concepts and linguistic items (words, sentences) serve functions that have little or nothing to do with semantics. However, I argue that, on the best interpretation of talk of the function of a concept such as ‘knowledge’, the function of ‘knowledge’ is relevant to semantics. Along the way I also suggest how to improve on what I call the ‘usual argument’ that Craig's account of the function of ‘knowledge’ supports epistemic contextualism.

Classical logic is explosive in the face of contradiction, yet we find ourselves using inconsistent theories. Mark Colyvan, one of the prominent advocates of the indispensability argument for realism about mathematical objects, suggests that such use can be garnered to develop an argument for commitment to inconsistent objects and, because of that, a paraconsistent underlying logic. I argue to the contrary that it is open to a classical logician to make distinctions, also needed by the paraconsistent logician, which allow a more nuanced ranking of theories in which inconsistent theories can have different degrees of usefulness and productivity. Facing inconsistency does not force us to adopt an underlying paraconsistent logic. Moreover we will see that the argument to best explanation deployed by Colyvan in this context is unsuccessful. I suggest that Quinean approach which Colyvan champions will not lead to the revolutionary doctrines Colyvan endorses.

Certain combinations of attitudes are manifestly unreasonable. It is unreasonable to believe that dogs bark, for example, if one concedes that one has no justification to believe this. Why are the irrational combinations irrational? One suggestion is that these are attitudes that a subject cannot have justification to have. If this is right, we can test claims about the structure of propositional justification by relying on our observations about which combinations of attitudes constitute Moorean absurd pairs. In a recent defense of access internalism, Smithies argues that only access internalism can explain why various combinations of attitude are irrational. In this paper, I shall argue that access internalism cannot explain the relevant data. Reflection on Moore's Paradox will not tell us much of anything about propositional justification and cannot support access internalism.

Under the traditional system of peer-reviewed publication, the degree of prestige conferred to authors by successful publication is tied to the degree of the intellectual rigor of its peer review process: ambitious scientists do well professionally by doing well epistemically. As a result, we should expect journal editors, in their dual role as epistemic evaluators and prestige-allocators, to have the power to motivate improved author behavior through the tightening of publication requirements. Contrary to this expectation, I will argue that the publication bias literature in academic medicine demonstrates that editor interventions have had limited effectiveness in improving the health of the publication and trial registration record, suggesting that much stronger interventions are needed.

I examine how particular social arrangements and incentive structures encourage the honest reporting of experimental results and minimize fraudulent scientific work. In particular I investigate how epistemic communities can achieve this goal by promoting members to police the community. Using some basic tools from game theory, I explore a simple model in which scientists both conduct research and have the option of investigating the findings of their peers. I find that this system of peer policing can in many cases ensure high levels of honesty.

Drug discovery traditionally has occurred behind closed doors in for-profit corporations hoping to develop best-selling medicines that recoup initial research investment, sustain marketing infrastructures, and pass on healthy returns to shareholders. Only corporate Pharma has the man- and purchasing-power to synthesize the thousands of molecules needed to find a new drug and to conduct the clinical trials that will make the drug legal. Against this view, individual physician-scientists have suggested that the promise of applied genomics work calls for a new form of social organization – the open source sharing of molecular probes – in order to speed the generation and understanding of new therapeutics. Recent successes in open source drug discovery show it is possible to produce valuable, empirically adequate, and sustainable collective beliefs without secrecy, proprietary attitudes, initial cooperation from Pharma, or outsized monetary incentives. After reviewing and differentiating these successes, I diagnose the source of this healthy new epistemic strategy.

A scientific community can be modeled as a collection of epistemic agents attempting to answer questions, in part by communicating about their hypotheses and results. We can treat the pathways of scientific communication as a network. When we do, it becomes clear that the interaction between the structure of the network and the nature of the question under investigation affects epistemic desiderata, including accuracy and speed to community consensus. Here we build on previous work, both our own and others’, in order to get a firmer grasp on precisely which features of scientific communities interact with which features of scientific questions in order to influence epistemic outcomes.

Here we introduce a measure on the landscape meant to capture some aspects of the difficulty of answering an empirical question. We then investigate both how different communication networks affect whether the community finds the best answer and the time it takes for the community to reach consensus on an answer. We measure these two epistemic desiderata on a continuum of networks sampled from the Watts–Strogatz spectrum. It turns out that finding the best answer and reaching consensus exhibit radically different patterns. The time it takes for a community to reach a consensus in these models roughly tracks mean path length in the network. Whether a scientific community finds the best answer, on the other hand, tracks neither mean path length nor clustering coefficient.

I explore an agent-based model of the development and dissemination of scientific theory that makes very little use of any pre-defined “social structure” (such as partnerships or collaborations). In these models, under a broad range of values of the parameters, widespread (but not universal) “agreement” about scientific theory emerges. Moreover, the residual disagreement turns out to be important to developing new theories in the face of new evidence.