Reflective equilibrium and underdetermination in epistemology

The basic aim of Alvin Goldman’s approach to epistemology, and the tradition it represents, is naturalistic; that is, epistemological theories in this tradition aim to identify the naturalistic, nonnormative criteria on which justified belief supervenes (Goldman, 1986; Markie, 1997). The basic method of Goldman’s epistemology, and the tradition it represents, is the reflective equilibrium test; that is, epistemological theories in this tradition are tested against our intuitions about cases of justified and unjustified belief (Goldman, 1986; Markie, 1997). I will argue that the prospect of having to reject their standard methodology is one epistemologists have to take very seriously; and I will do this by arguing that some current rival theories of epistemic justification are in fact in reflective equilibrium with our intuitions about cases of justified and unjustified belief. That is, I will argue that intuition underdetermines theory choice in epistemology, in much the way that observation underdetermines theory choices in empirical sciences. If reflective equilibrium leads to the underdetermination problem I say it leads to, then it cannot satisfy the aims of contemporary epistemology, and so cannot serve as its standard methodology.     

Reliabilism and Brains in Vats

According to epistemic internalism, the only facts that determine the justificational status of a belief are facts about the subject’s own mental states, like beliefs and experiences. Externalists instead hold that certain external facts, such as facts about the world or the reliability of a belief-producing mechanism, affect a belief’s justificational status. Some internalists argue that considerations about evil demon victims and brains in vats provide excellent reason to reject externalism: because these subjects are placed in epistemically unfavorable settings, externalism seems unable to account for the strong intuition that these subjects’ beliefs are nonetheless justified. I think these considerations do not at all help the internalist cause. I argue that by appealing to the anti-individualistic nature of perception, it can be shown that skeptical scenarios provide no reason to prefer internalism to externalism.     

Reliabilist Justification: Basic,Easy, and Brute

Process reliabilists hold that in order for a belief to be justified, it must result from a reliable cognitive process. They also hold that a belief can be basically justified: justified in this manner without having any justification to believe that belief is reliably produced. Fumerton (1995), Vogel (2000), and Cohen (2002) have objected that such basic justification leads to implausible easy justification by means of either epistemic closure principles or so-called track record arguments. I argue that once we carefully distinguish closure principles from transmission principles, and epistemic consequences from epistemic preconditions, neither version of this objection succeeds.

Scepticism,context and modal reasoning

I analyze some classical solutions of the skeptical argument and some of their week points (especially the contextualist solution). First I have proposed some possible improvement of the contextualist solution (the introduction of the explicit-implicit belief and knowledge distinction beside the differences in the relevance of some counter-factual alternatives). However, this solution does not block too fast jumps of the everyday context (where empirical knowledge is possible) into skeptical context (where empirical knowledge is impossible). Then I analyze some formal analogies between some modal arguments on the contingency of empirical facts (and the world as whole) and the skeptical arguments against empirical knowledge. I try to show that the skeptical conclusion “Empirical knowledge does not exist” is logically coherent with the thesis that they are empirical facts and that we have true belief on them. In order to do that without contradictions I have to accept a non-classical definition of knowledge: S knows that p:= S is not justified to allow that non-p. Knowledge and justified allowance function here as some pseudo-theoretical concepts which allow only some partial and conditional definitions by some “empirical” terms and logical conditions.     

Understanding, Integration, and Epistemic Value

Understanding enjoys a special kind of value, one not held by lesser epistemic states such as knowledge and true belief. I explain the value of understanding via a seemingly unrelated topic, the implausibility of veritism. Veritism holds that true belief is the sole ultimate epistemic good and all other epistemic goods derive their value from the epistemic value of true belief. Veritism entails that if you have a true belief that p, you have all the epistemic good qua p. Veritism is a plausible and widely held view; I argue that it is untenable. I argue that integration among beliefs possesses epistemic value independent from the good of true belief, and so has value veritism cannot account for. I argue further that this integration among beliefs comprises the distinctive epistemic value of understanding.     

A Priori knowledge contextualised and Benacerraf’s dilemma

In this article, I discuss Hawthorne’s contextualist solution to Benacerraf’s dilemma. He wants to find a satisfactory epistemology to go with realist ontology, namely with causally inaccessible mathematical and modal entities. I claim that he is unsuccessful. The contextualist theories of knowledge attributions were primarily developed as a response to the skeptical argument based on the deductive closure principle. Hawthorne uses the same strategy in his attempt to solve the epistemologist puzzle facing the proponents of mathematical and modal realism, but this problem is of a different nature than the skeptical one. The contextualist theory of knowledge attributions cannot help us with the question about the nature of mathematical and modal reality and how they can be known. I further argue that Hawthorne’s account does not say anything about a priori status of mathematical and modal knowledge. Later, Hawthorne adds to his account an implausible claim that in some contexts a gettierized belief counts as knowledge.     

Coherentism,Truth, and Witness Agreement

Coherentists on epistemic justification claim that all justification is inferential, and that beliefs, when justified, get their justification together (not in isolation) as members of a coherent belief system. Some recent work in formal epistemology shows that “individual credibility” is needed for “witness agreement” to increase the probability of truth and generate a high probability of truth. It can seem that, from this result in formal epistemology, it follows that coherentist justification is not truth-conducive, that it is not the case that, under the requisite conditions, coherentist justification increases the probability of truth and generates a high probability of truth. I argue that this does not follow.     

Reflective Luck and Belief Ownership

A belief is reflectively lucky if it is a matter of luck that the belief is true, given what a subject is aware of on reflection alone. Various epistemologists have argued that any adequate theory of knowledge should eliminate reflective luck, but doing so has proven difficult. This article distinguishes between two kinds of reflective luck arguments in the literature: local arguments and global arguments. It argues that local arguments are best interpreted as demanding, not that one be reflectively aware of the reliability of the sources of one’s beliefs, but that one’s beliefs be attributable to one as one’s own. The article then argues that global arguments make illegitimate demands, because they require that we be ultimately answerable for our beliefs. In the end, the article argues that epistemologists should shift their focus away from reflective luck and toward the conditions under which beliefs are attributable to cognitive agents.     

Expressing and Describing Experiences. A Case of Showing Versus Saying

Experiences are interpreted as conscious mental occurrences that are of phenomenal character. There is already a kind of (weak) intentionality involved with this phenomenal interpretation. A stricter conception of experiences distinguishes between purely phenomenal experiences and intentional experiences in a narrow sense. Wittgenstein’s account of psychological (experiential) verbs is taken over: Usually, expressing mental states verbally is not describing them. According to this, “I believe” can be seen as an expression of one’s own belief, but not as an expression of a belief about one’s belief. Hence, the utterance “I believe it is raining” shows that I believe that it is raining, although it is not said by these words that I believe that it is raining. Thinking thoughts such as “I believe it is raining, but it is not raining” (a variant of Moore’s paradox) is an absurdity between what is already said by silently uttering “It is not raining” and what is shown by silently uttering “I believe it is raining.” The paper agrees with a main result of Wittgenstein’s considerations of Moore’s paradox, namely the view that logical structure, deducibility, and consistency cannot be reduced solely to propositions—besides a logic of propositions, there is, for example, a logic of assertions and of imperatives, respectively.     

Entitlement, Justification, and the Bootstrapping Problem

According to the bootstrapping problem, any view that allows for basic knowledge (knowledge obtained from a reliable source prior to one??s knowing that that source is reliable) is forced to accept that one can utilize a track-record argument to acquire justification for believing that one??s belief source is reliable; yet, we tend to think that acquiring justification in this way is too easy. In this paper I argue, first, that those who respond to the bootstrapping problem by denying basic knowledge succumb to over-intellectualizing epistemology, and secondly, reliabilist views avoid over-intellectualization only at the expense of sanctioning bootstrapping as a benign procedure. Both of these outcomes are difficult to bear. To ward off each of these unsavory outcomes, I propose an alternative solution that draws on a distinction between two separate epistemic concepts: entitlement and justification.     

Ownership, Authorship and External Justification

Some of the most well-known arguments against epistemic externalism come in the form of thought experiments involving subjects who acquire beliefs through anomolous means such as clairvoyance. These thought experiments purport to provide counterexamples to the reliabilist conception of justification: their subjects are intuitively epistemically unjustified, yet meet reliabilist externalist criteria for justification. In this article, I address a recent defence of externalism due to Daniel Breyer, who argues that externalists need not consider such subjects justified, since they fail to own those beliefs in a way required for epistemic evaluability. I argue that the concept of belief ownership Breyer adopts leaves his account open to related counterexamples, and suggest a modification, drawing on analogies between these cases and cases of delusions, such as thought insertion. I will argue that a concept of authorship developed in the literature on delusions better grounds the sense of attribution required for epistemic evaluability.     

Wisdom, Understanding and Knowledge: A Virtue-Theoretic Proposal

Reaching understanding is one of our central epistemic goals, dictated by our important motivational epistemic virtue, namely inquisitiveness about the way things hang together. Understanding of humanly important causal dependencies is also the basic factual-theoretic ingredient of wisdom on the anthropocentric view proposed in the article. It appears at two levels. At the first level of immediate, spontaneous wisdom, it is paired with practical knowledge and motivation (phronesis), and encompasses understanding of oneself (a distinct level of self-knowledge having to do with one??s dispositions capacities, vulnerabilities, and ways of reacting), of other people, of possibilities of meaningful life, and of relevant courses of events. It is partly resistant to skeptical scenarios, but not completely. On the second and reflective level, understanding helps fuller holistic integration of one??s first order practical and factual-theoretic knowledge and motivation, thus minimizing potential and actual conflicts between all these components. It also participates in the reflective control whose exercise is needed for full reflective wisdom, the crucial epistemic-practical virtue or human excellence.     

Inferior Disagreement

Literature in the epistemology of disagreement has focused on peer disagreement: disagreement between those with shared evidence and equal cognitive abilities. Additional literature focuses on the perspective of amateurs who disagree with experts. However, the appropriate epistemic reaction from superiors who disagree with inferiors remains underexplored. Prima facie, this may seem an uninteresting set of affairs. If A is B’s superior, and A has good reason to believe she is B’s superior, A appears free to dismiss B’s disagreement. However, a closer look will show otherwise. I first distinguish competent from incompetent inferiors and then argue that disagreement from the former often gives superiors reason to adjust credence and reevaluate belief. In other words, epistemic inferiority alone is insufficient grounds for dismissing opinion. More nuanced difficulties arise with incompetent inferiors. When superiors disagree with incompetents, this might provide evidence to bolster belief credence; however, agreement from incompetents can defeat justification. In either instance, inferior opinion carries epistemic weight. Yet, this fails to cover all ground; at times, superiors learn nothing from inferior disagreement. I finish by exploring these uninformative disagreements, how to distinguish them from the informative cases, and the proper epistemic reactions thereof.     

Probabilism,entropies and strictly proper scoring rules

Accuracy arguments are the en vogue route in epistemic justifications of probabilism and further norms governing rational belief. These arguments often depend on the fact that the employed inaccuracy measure is strictly proper. I argue controversially that it is ill-advised to assume that the employed inaccuracy measures are strictly proper and that strictly proper statistical scoring rules are a more natural class of measures of inaccuracy. Building on work in belief elicitation I show how strictly proper statistical scoring rules can be used to give an epistemic justification of probabilism.An agent's evidence does not play any role in these justifications of probabilism. Principles demanding the maximisation of a generalised entropy depend on the agent's evidence. In the second part of the paper I show how to simultaneously justify probabilism and such a principle. I also investigate scoring rules which have traditionally been linked with entropies.     

Market selection: hungry misers and bloated bankrupts

The Market Selection Hypothesis is a principle which (informally) proposes that ‘less knowledgeable’ agents are eventually eliminated from the market. This elimination may take the form of starvation (the proportion of output consumed drops to zero), or may take the form of going broke (the proportion of asset held drops to zero), and these are not the same thing. Starvation may result from several causes, diverse beliefs being only one. We firstly identify and exclude these other possible causes, and then prove that starvation is equivalent to inferior belief, under suitable technical conditions. On the other hand, going broke cannot be characterized solely in terms of beliefs, as we show. We next present a remarkable example with two agents with different beliefs, in which one agent starves yet amasses all the capital, and the other goes broke yet consumes all the output—the hungry miser and the bloated bankrupt. This example also serves to show that although an agent may starve, he may have long-term impact on the prices. This relates to the notion of price impact introduced by Kogan et al. (Market selection, working paper, 2009), which we correct in the final section, and then use to characterize situations where asymptotically equivalent pricing holds.     

Possible Worlds of Doubt

A prominent contemporary anti-skeptical strategy, most famously articulated by Keith DeRose, aims to cage the skeptic′s doubts by contextualizing subjunctive conditional accounts of knowledge through a conversational rule of sensitivity. This strategy, I argue, courts charges of circularity by selectively invoking heavy counterfactual machinery. The reason: such invocation threatens to utilize a metric for modal comparison that is implicitly informed by judgments of epistemic sameness. This gives us reason to fear that said modal metric is selectively cherry-picked in advance to support the very anti-skeptical conclusion for which the contextualist longs.     

Armchair luck: Apriority,intellection and epistemic luck

The paper argues that there is such a thing as luck in acquisition of candidate apriori beliefs and knowledge, and that the possibility of luck in this “armchair” domain shows that definitions of believing by luck thatp offered in literature are inadequate, since they mostly rely on the possibility of it being the case that not-p. Whenp is necessary, such a definition should be supplemented by one pointing to variation in belief, not in the fact believed. Thus the paper suggests a focus upon the agent and her epistemic virtue in the account of epistemic luck in general.     

Proper belief revision and rationalizability in dynamic games

In this paper we develop an epistemic model for dynamic games in which players may revise their beliefs about the opponents’ utility functions as the game proceeds. Within this framework, we propose a rationalizability concept that is based upon the following three principles: (1) at every instance of the game, a player should believe that his opponents are carrying out optimal strategies, (2) a player, at information set h, should not change his belief about an opponent’s relative ranking of two strategies s and s′ if both s and s′ could have led to h, and (3) the players’ initial beliefs about the opponents’ utility functions should agree on a given profile u of utility functions. Common belief in these events leads to the concept of persistent rationalizability for the profile u of utility functions. It is shown that for a given game tree with observable deviators and a given profile u of utility functions, every properly point-rationalizable strategy is a persistently rationalizable strategy for u. This result implies that persistently rationalizable strategies always exist for all game trees with observable deviators and all profiles of utility functions. We provide an algorithm that can be used to compute the set of persistently rationalizable strategies for a given profile u of utility functions. For generic games with perfect information, persistent rationalizability uniquely selects the backward induction strategy for every player.     

An analysis of the a priori and a posteriori

I present and defend a unified, non-reductive analysis of the a priori and a posteriori. It is a mistake to remove all epistemic conditions from the analysis of the a priori (as, for example, Alvin Goldman has recently suggested doing). We can keep epistemic conditions (like unrevisability) in the analysis as long as we insist that a priori and a posteriori justification admit of degrees. I recommend making the degree to which a belief’s justification is a priori or a posteriori solely dependent on the revisability relations that obtain among the faculties that deliver the belief and all other faculties.