The Generality Problem for Reliabilism

This paper provides an empirical defense of credit theories of knowing against Mark Alfano’s challenges to them based on his theses of inferential cognitive situationism and of epistemic situationism. In order to support the claim that credit theories can treat many cases of cognitive success through heuristic cognitive strategies as credit-conferring, the paper develops the compatibility between virtue epistemologies qua credit theories, and dual-process theories in cognitive psychology. It also a response to Lauren Olin and John Doris’ “vicious minds” thesis, (...) and their “tradeoff problem” for virtue theories. A genuine convergence between virtue epistemology and dual-process theory is called for, while acknowledging that this effort may demand new and more empirically well-informed projects on both sides of the division between Conservative virtue epistemology (including the credit theory of knowing) and Autonomous virtue epistemology (including projects for providing guidance to epistemic agents). (shrink)

Epistemic luck has been the focus of much discussion recently. Perhaps the most general knowledge-precluding type is veritic luck, where a belief is true but might easily have been false. Veritic luck has two sources, and so eliminating it requires two distinct conditions for a theory of knowledge. I argue that, when one sets out those conditions properly, a solution to the generality problem for reliabilism emerges.

Here I present and defend an etiological theory of objective, doxastic justification, and related theories of defeat and evidence. The theory is intended to solve a problem for reliabilist epistemologies— the problem of identifying relevant environments for assessing a process's reliability. It is also intended to go some way to accommodating, neutralizing, or explaining away many internalist-friendly elements in our epistemic thinking.

In this paper I critically examine the Generality Problem and argue that it does not succeed as an objection to reliabilism. Although those who urge the Generality Problem are correct in claiming that any process token can be given indefinitely many descriptions that pick out indefinitely many process types, they are mistaken in thinking that reliabilists have no principled way to distinguish between relevant and irrelevant process types.

The generality problem is widely considered to be a devastating objection to reliabilist theories of justification. My goal in this paper is to argue that a version of the generality problem applies to all plausible theories of justification. Assume that any plausible theory must allow for the possibility of reflective justification—S's belief, B, is justified on the basis of S's knowledge that she arrived at B as a result of a highly (but not perfectly) reliable way of reasoning, R. The (...) generality problem applies to all cases of reflective justification: Given that is the product of a process-token that is an instance of indefinitely many belief-forming process-types (or BFPTs), why is the reliability of R, rather than the reliability of one of the indefinitely many other BFPTs, relevant to B's justificatory status? This form of the generality problem is restricted because it applies only to cases of reflective justification. But unless it is solved, the generality problem haunts all plausible theories of justification, not just reliabilist ones. (shrink)

This paper raises three critical questions about Jarrett Leplin's version of reliabilism.

In “Why the generality problem is everybody’s problem,” Michael Bishop argues that every theory of justification needs a solution to the generality problem. He contends that a solution is needed in order for any theory to be used in giving an acceptable account of the justificatory status of beliefs in certain examples. In response, first I will describe the generality problem that is specific to process reliabilism and two other sorts of problems that are essentially the same. Then I will (...) argue that the examples that Bishop presents pose no such problem for some theories. I will illustrate the exempt theories by describing how an evidentialist view can account for the justification in the examples without having any similar problem. It will be clear that other views about justification are likewise unaffected by anything like the generality problem. (shrink)

This article is concerned with a statistical proposal due to James R. Beebe for how to solve the generality problem for process reliabilism. The proposal is highlighted by Alvin I. Goldman as an interesting candidate solution. However, Goldman raises the worry that the proposal may not always yield a determinate result. We address this worry by proving a dilemma: either the statistical approach does not yield a determinate result or it leads to trivialization, i.e. reliability collapses into truth (and anti-reliability (...) into falsehood). Various strategies for avoiding this predicament are considered, including revising the statistical rule or restricting its application to natural kinds. All amendments are seen to have serious problems of their own. We conclude that reliabilists need to look elsewhere for a convincing solution to the generality problem. (shrink)

Guided by the work of William Alston, Jonathan Adler and Michael Levin propose a solution to the generality problem for reliabilism. In some respects their proposal improves on those we have discussed. We argue that the problem remains unsolved.

According to Intellectualism knowing how to V is a matter of knowing a suitable proposition about a way of V-ing. In this paper, I consider the question of which ways of acting might figure in the propositions which Intellectualists claim constitute the object of knowledge-how. I argue that Intellectualists face a version of the Generality Problem – familiar from discussions of Reliabilism – since not all ways of V-ing are such that knowledge about them suffices for knowledge-how. I consider various (...) responses to this problem, and argue that none are satisfactory. (shrink)

This paper has two goals. First, it aims to investigate the empirical assumptions of a recent proposal due to Olsson (forthcoming), according to which the generality problem for process-reliabilism can be approached by recruiting patterns and models from the basic-level research in cognitive psychology. Second, the paper attempts to generalize findings in the basic-level literature pertaining to concrete nouns to the abstract verbs that denote belief-forming processes. I will demonstrate that verbs for belief-forming processes exhibit the kind of linguistic convergence (...) that is characteristic of basic-level words, although these words are not neatly taxonomically organized or associated with large feature sets. Next, I will evaluate and partially validate Olsson's proposal in light of these findings. I will provide some discussion of possible explanations of the results are discussed, as well as the impact these results have for structural models of basic-level advantage, and for the feasibility of the explanatory strategy that these models presuppose. Finally, I will conclude that even though no particular model is compromised by these results, they call into question the underlying explanatory strategy by highlighting its parochial nature. (shrink)

Process-reliabilist analyses of justification and knowledge face the generality problem. Recent discussion of this problem turns on certain untested empirical assumptions that this paper investigates. Three experiments are reported: two are free-naming studies that support the existence of a basic level in the previously unexplored domain of names for belief-forming processes; the third demonstrates that reliability judgments for the basic-level belief-forming process types are very strongly correlated with the corresponding justification and knowledge judgments. I argue that these results lend support (...) to process-reliabilism; I then go on to explore several different ways to make use of them in solutions to the generality problem. (shrink)

The Generality Problem is widely recognized to be a serious problem for reliabilist theories of justification. James R. Beebe's Statistical Solution is one of only a handful of attempted solutions that has garnered serious attention in the literature. In their recent response to Beebe, Julien Dutant and Erik J. Olsson successfully refute Beebe's Statistical Solution. This paper presents a New Statistical Solution that countenances Dutant and Olsson's objections, dodges the serious problems that trouble rival solutions, and retains the theoretical virtues (...) that made Beebe's solution so attractive in the first place. There indeed exists a principled, rigorous, conceptually sparse, and plausible solution to the Generality Problem: it is the New Statistical Solution. (shrink)

The purpose of this paper is to offer a diagnosis and a resolution to generality problem. I state the generality problem and suggest a distinction between criteria of relevance and what I call a theory of determination. The generality problem may concern either of these. While plausible criteria of relevance would be convenient for the externalist, he does not need them. I discuss various theories of determination, and argue that no existing theory of determination is plausible. This provides a case (...) for the no determination view: there are no facts that determine relevant types. This is the diagnosis of the generality problem. The externalist, however, may embrace the no determination view. This is what provides a resolution to the generality problem. (shrink)

In this paper, I aim to develop a novel virtue reliabilist account of justified belief, which incorporates insights from both process reliabilism and extant versions of virtue reliabilism. Like extant virtue reliabilist accounts of justi- fied belief, the proposed view takes it that justified belief is a kind of competent performance and that competent performances require reliable agent abilities. However, unlike extant versions of virtue reliabilism, the view takes abilities to essentially involve reliable processes. In this way, the proposed takes (...) a leaf from process reliabilism. Finally, I will provide reason to believe that the view compares favourably with both extant versions of virtue reliabilism and process reliabilism. In particular, I will show that in taking abilities to essentially involve reliable processes, the view has an edge over extant versions of virtue reliabilism. Moreover, I will argue that the proposed view can either solve or defuse a number of classical problems of process reliabilism, including the new evil demon problem, the problem of clairvoyant cases and the generality problem. (shrink)

In his debates with Daniel Kahneman and Amos Tversky, Gerd Gigerenzer puts forward a stricter standard for the proper representation of judgment heuristics. I argue that Gigerenzer’s stricter standard contributes to naturalized epistemology in two ways. First, Gigerenzer’s standard can be used to winnow away cognitive processes that are inappropriately characterized and should not be used in the epistemic evaluation of belief. Second, Gigerenzer’s critique helps to recast the generality problem in naturalized epistemology and cognitive psychology as the methodological problem (...) of identifying criteria for the appropriate specification and characterization of cognitive processes in psychological explanations. I conclude that naturalized epistemologists seeking to address the generality problem should turn their focus to methodological questions about the proper characterization of cognitive processes for the purposes of psychological explanation. (shrink)

Objections to reliabilist theories of knowledge and justification have looked insuperable. Reliability is a property of the process of belief formation. But the generality problem apparently makes the specification of any such process ambiguous. The externalism of reliability theories clashes with strongly internalist intuitions. The reliability property does not appear closed under truth-preserving inference, whereas closure principles have strong intuitive appeal. And epistemic paradoxes, like the preface and the lottery, seem unavoidable if knowledge or justification depends on the frequency with (...) which a process generates true beliefs. The present theory has the conceptual resources to meet these challenges. It requires that a justificatory belief-formation process be intentionally applied. It distinguishes the justification of beliefs from that of the believer. And it avoids a frequency interpretation of reliability by introducing a notion of the normalcy of conditions under which processes are intentionally used. (shrink)

Reliabilist theories of knowledge face the “generality problem”; any token of a belief-forming processes instantiates types of different levels of generality, which can vary in reliability. I argue that we exploit this situation in epistemic evaluation; we appraise beliefs in different ways by adverting to reliability at different levels of generality. We can detect at least two distinct uses of reliability, which underlie different sorts of appraisals of beliefs and believers.

Reliabilism holds that knowledge is true belief reliably caused. Reliabilists should say something about individuating processes; critics deny that the right degree of generality can be specified without arbitrariness. It is argued that this criticism applies as well to processes mentioned in scientific explanations. The gratuitous puzzles created thereby show that the “generality problem” is illusory.

The generality problem is perhaps the most notorious problem for process reliabilism. Several recent responses to the generality problem have claimed that the problem has been unfairly leveled against reliabilists. In particular, these responses have claimed that the generality problem is either (i) just as much of a problem for evidentialists, or (ii) if it is not, then a parallel solution is available to reliabilists. Along these lines, Juan Comesaña has recently proposed solution to the generality problem—well-founded reliabilism. According to (...) Comesaña, the solution to the generality problem lies in solving the basing problem, such that any solution to the basing problem will give a solution to the generality problem. Comesaña utilizes Conee and Feldman’s evidentialist account of basing (Conee and Feldman’s well-foundedness principle) in forming his version of reliabilism. In this paper I show that Comesaña’s proposed solution to the generality problem is inadequate. Well-founded reliabilism both fails to solve the generality problem and subjects reliabilism to new damning verdicts. In addition, I show that evidentialism does not face any parallel problems, so the generality problem remains a reason to prefer evidentialism to reliabilism. (shrink)

The Generality Problem for process reliabilism is to outline a procedure for determining when two beliefs are produced by the same process, in such a way as to avoid, on the one hand, individuating process types so narrowly that each type is instantiated only once, or, on the other hand, individuating them so broadly that beliefs that have different epistemic statuses are subsumed under the same process type. In this paper, I offer a solution to the problem which takes belief‐independent (...) processes to be functions that take as inputs information about distal states of affairs, and produce beliefs as outputs. Processes are individuated narrowly, so as to avoid the latter aspect of the Generality problem, but, by holding process tokens to be of the same type when they take perceptually equivalent scenes as inputs, and produce beliefs of the same kind as outputs, the former aspect of the problem is avoided too. Having argued that this method of typing process tokens solves the Generality Problem, I then argue that my solution does not fall prey to objections that have been, or might be, raised for similar proposals. (shrink)

The orthodox position in epistemology, for both externalists and internalists, is that a subject in a ‘bad case’—a sceptical scenario—is so epistemically badly off that they cannot know how badly off they are. Ofra Magidor contends that externalists should break ranks on this question, and that doing so is liberating when it comes time to confront a number of central issues in epistemology, including scepticism and the new evil demon problem for process reliabilism. In this reply, I will question whether (...) Magidor’s argument should persuade externalists, whether it really engages with the orthodox view on what subjects in bad cases can know, and whether the dispute is, as Magidor insists, a significant one for contemporary epistemology. (shrink)

Process reliabilism says that a belief is justified iff the belief-forming process that produced it is sufficiently reliable. But any token belief-forming process is an instance of a number of different belief-forming process types. The problem of specifying the relevant type is known as the ‘generality problem’ for process reliabilism. This paper proposes a broadly relativist solution to the generality problem. The thought is that the relevant belief-forming process type is relative to the context. While the basic idea behind the (...) solution is from Mark Heller (1995), the solution defended here departs from Heller on a crucial point. Because of this departure, my solution avoids a serious problem with Heller’s solution. (shrink)

Vogel argues that sensitivity accounts of knowledge are implausible because they entail that we cannot have any higher-level knowledge that our beliefs are true, not false. Becker and Salerno object that Vogel is mistaken because he does not formalize higher-level beliefs adequately. They claim that if formalized correctly, higher-level beliefs are sensitive, and can therefore constitute knowledge. However, these accounts do not consider the belief-forming method as sensitivity accounts require. If we take bootstrapping as the belief-forming method, as the discussed (...) cases suggest, then we face a generality problem. Our higher-level beliefs as formalized by Becker and Salerno turn out to be sensitive according to a wide reading of bootstrapping, but insensitive according to a narrow reading. This particular generality problem does not arise for the alternative accounts of process reliabilism and basis-relative safety. Hence, sensitivity accounts not only deliver opposite results given different formalizations of higher-level beliefs, but also for the same formalization, depending on how we interpret bootstrapping. Therefore, sensitivity accounts do not fail because they make higher-level knowledge impossible, as Vogel argues, and they do not succeed in allowing higher-level knowledge, as Becker and Salerno suggest. Rather, their problem is that they deliver far too heterogeneous results. (shrink)

The No-Miracles Argument (NMA) is often used to support scientific realism. We can formulate this argument as an inference to the best explanation this accusation of circularity by appealing to reliabilism, an externalist epistemology. In this paper I argue that this retreat fails. Reliabilism suffers from a potentially devastating difficulty known as the Generality Problem and attempts to solve this problem require adopting both epistemic and metaphysical assumptions regarding local scientific theories. Although the externalist can happily adopt the former, if (...) he adopts the latter then the Generality Problem arises again, but now at the level of scientific methodology. Answering this new version of the Generality Problem is impossible for the scientific realist without making the important further assumption that there exists the possibility of a unique rule of Doing this however would make the NMA viciously premise circular. (shrink)

In this paper, we seek a reliabilist account of justified credence. Reliabilism about justified beliefs comes in two varieties: process reliabilism (Goldman, 1979, 2008) and indicator reliabilism (Alston, 1988, 2005). Existing accounts of reliabilism about justified credence comes in the same two varieties: Jeff Dunn’s is a version of process reliabilism (Dunn, 2015) while Weng Hong Tang offers a version of indicator reliabilism (Tang, 2016). As we will see, both face the same objection. If they are right about what justification (...) is, it is mysterious why we care about justification, for neither of the accounts explains how justification is connected to anything of epistemic value. We will call this the Connection Problem. I begin by describing Dunn’s process reliabilism and Tang’s indicator reliabilism. I argue that, understood correctly, they are, in fact, extensionally equivalent. That is, Dunn and Tang reach the top of the same mountain, albeit by different routes. However, I argue that both face the Connection Problem. In response, I offer my own version of reliabilism, which is both process and indicator, and I argue that it solves that problem. Furthermore, I show that it is also extensionally equivalent to Dunn’s reliabilism and Tang’s. Thus, I reach the top of the same mountain as well. Having done that, I consider some objections to the account I propose and some consequences of it. (shrink)

Matthew Frise claims that reliabilist theories of justification have a temporality problem—the problem of providing a principled account of the temporal parameters of a process’s performance that determine whether that process is reliable at a given time. Frise considers a representative sample of principled temporal parameters and argues that there are serious problems with all of them. He concludes that the prospects for solving the temporality problem are bleak. Importantly, Frise argues that the temporality problem constitutes a new reason to (...) reject reliabilism. On this point, I argue that Frise is mistaken. There are serious interpretive difficulties with Frise’s argument. In this essay, I show that there are principled and reasonable temporal parameters for the reliabilist to adopt that successfully undermine the interpretations of Frise’s argument that only invoke plausible premises. There are interpretations of Frise’s argument that leave reliabilism without a clear parameter solution. However, I argue that these interpretations invoke controversial premises that are at best unmotivated, and at worst they merely re-raise older disputes about reliabilism. In any event, the temporality problem fails to constitute a new reason to reject reliabilism. (shrink)

Proper functionalism claims that a belief has epistemic warrant only if it’s formed according to the subject’s truth-aimed cognitive design plan. The most popular putative counter-examples to proper functionalism all involve agents who form beliefs in seemingly warrant-enabling ways that don’t appear to proceed according to any sort of design. The Swampman case is arguably the most famous scenario of this sort. However, some proper functionalists accept that subjects like Swampman have warrant, opting instead to adopt a non-standard account of (...) design. But critics of proper functionalism hold that this strategy comes at a high cost: the design-plan condition now seems explanatorily superfluous. James Taylor construes cases like Swampman as posing a dilemma for the proper functionalist: either deny warrant in these cases and concede that proper functionalism doesn’t capture our intuitions, or affirm warrant and undermine the explanatory power of the design-plan condition. Proper functionalists have replied to both horns of this dilemma. Recently, Kenny Boyce and Andrew Moon have argued that warrant-affirming intuitions on cases like Swampman are motivated by a principle that has a clear counter-example. Also, Alvin Plantinga presents a set of cases that supposedly cause problems for any analysis of warrant that lacks a design-plan condition. In this essay, I present a counter-argument to Boyce and Moon’s argument, and show that a more robust reliability condition can accommodate Plantinga’s problem cases. I conclude that we’re left with no good reason to doubt that cases like Swampman raise a troubling dilemma for the proper functionalist. (shrink)

The generality problem is commonly seen as one of the most pressing issues for process reliabilism. The generality problem starts with the following question: of all the process types exemplified by a given process token, which type is the relevant one for measuring reliability? Defenders of the generality problem claim that process reliabilists have a burden to produce an informative account of process type relevance. As they argue, without such a successful account, the reasonability of process reliabilism is significantly undermined. (...) One way for the reliabilist to respond is to attempt to construct such a theory of type relevance. But another way of responding is to argue that, if finding an account of type relevance is a burden for the reliabilist, then it is also a burden for everyone else. Thus, the generality problem doesn’t present some unique reason to reject process reliabilism. I call this latter strategy a parrying response. In this essay, I examine the contemporary parrying responses of Michael Bishop and Juan Comesaña, which have both faced recent criticism. I respond to these critics, and argue that parrying responses are far stronger than defenders of the generality problem have appreciated. (shrink)

Many current popular views in epistemology require a belief to be the result of a reliable process (aka ‘method of belief formation’ or ‘cognitive capacity’) in order to count as knowledge. This means that the generality problem rears its head, i.e. the kind of process in question has to be spelt out, and this looks difficult to do without being either over or under-general. In response to this problem, I propose that we should adopt a more fine-grained account of the (...) epistemic basing relation, at which point the generality problem becomes easy to solve. (shrink)

The traditional generality problem for process reliabilism concerns the difficulty in identifying each belief forming process with a particular kind of process. Thatidentification is necessary since individual belief forming processes are typically of many kinds, and those kinds may vary in reliability. I raise a new kind of generality problem, one which turns on the difficulty of identifying beliefs with processes by which they were formed. This problem arises because individual beliefs may be the culmination of overlapping processes of distinct (...) lengths, and these processes may differ in reliability. I illustrate the force of this problem with a discussion of recent work on the bootstrapping problem. (shrink)

Critics of reliability theories of epistemic justificationoften claim that the `generality problem' is an insurmountabledifficulty for such theories. The generality problem is theproblem of specifying the level of generality at which abelief-forming process is to be described for the purposeof assessing its reliability. This problem is not asintractable as it seems. There are illuminating solutionsto analogous problems in the ethics literature. Reliabilistsought to attend to utilitarian approaches to choices betweeninfinite utility streams; they also ought to attend towelfarist approaches to social (...) choice situations that donot demand full aggregation of individual welfares.These analogies suggest that the traditional `single number'approach to reliability is misguided. I argue that a newapproach – the `vector reliability' approach – is preferable.Vector reliability theories associate target beliefs withreliability vectors – that is, structured collections ofreliability numbers – and construct criteria of epistemicjustification that appeal to these vectors. The bulk of thetheoretical labor involved in a reliability account of epistemicjustification is thus transferred from picking a uniquereliability number to constructing a plausible criterionof epistemic justification. (shrink)

I take issue with Robert Brandom's claim that on an analysis of knowledge based on objective probabilities it is not possible to provide a stable answer to the question whether a belief has the status of knowledge. I argue that the version of the problem of generality developed by Brandom doesn't undermine a truth-tracking account of noninferential knowledge that construes truth-tacking in terms of conditional probabilities. I then consider Sherrilyn Roush's claim that an account of knowledge based on probabilistic tracking (...) faces a version of the problem of generality. I argue that the problems she raises are specific to her account, and do not affect the version of the view that I have advanced. I then consider Brandom's argument that the cases that motivate reliabilist epistemologies are in principle exceptional. I argue that he has failed to make a cogent case for this claim. I close with the suggestion that the representationalist approach to knowledge that I endorse and Brandom rejects is in principle compatible with the kind of pragmatist approach to belief and truth that both Brandom and I endorse. (shrink)