The connection between whole and part is intimate: not only can we share the same space, but I’m incapable of leaving my parts behind; settle the nonmereological facts and you thereby settle what is a part of what; wholes don’t seem to be an additional ontological commitment over their parts. Composition as identity promises to explain this intimacy. But it threatens to make the connection too intimate, for surely the parts could have made a different whole and the whole have (...) had different parts. In this paper I attempt to offer an account of parthood that is intimate enough but not too intimate: the parts generate the whole, but they are not themselves the whole. (shrink)

Non-well-founded set theories allow set-theoretic exotica that standard ZFC will not allow, such as a set that has itself as its sole member. We can distinguish plenitudinous non-well-founded set theories, such as Boffa set theory, that allow infinitely many such sets, from restrictive theories, such as Finsler-Aczel or AFA, that allow exactly one. Plenitudinous non-well-founded set theories face a puzzle: nothing seems to explain the identity or distinctness of various of the sets they countenance. In this paper I aim to (...) sharpen this puzzle, make clear who it does and does not apply to and, ultimately, to argue in favor of a plenitudinous theory like Boffa. (shrink)

We examine moving spotlight theories of time: theories according to which there are past and future events and an objective present moment. In Section 1, we briefly discuss the origins of the view. In Section 2, we describe the traditional moving spotlight view, which we understand as an ‘enriched’ B-theory of time, and raise some problems for that view. In the next two sections, we describe versions of the moving spotlight view that we think are better and which solve those (...) problems. In Section 3, we describe a version of the view that combines permanentism – the thesis that all things always exist – with propositional temporalism, the thesis that some propositions are sometimes true and sometimes false. In Section 4, we discuss a version of the view that is like an ‘enriched’ presentism. We conclude with some brief thoughts on issues that remain outstanding. (shrink)

In a recent essay in this journal, Ross Cameron presents a novel solution to the problem of musical creation.1 The solution is of the ‘using a sledgehammer to crack a nut’ variety, since side by side with a dissolution of the problem of musical creation, his approach, if successful, would yield a swift answer to pretty much every central question in the ontology of art, and, for that matter, to a wide variety of perennial metaphysical difficulties. Nothing of (...) this magnitude should be nonchalantly swept aside. Unfortunately, Cameron's approach does not survive close scrutiny. (shrink)

Ross Cameron's The Moving Spotlight argues that of the three most common dynamical theories of time – presentism, the growing block theory and the moving spotlight theory – his version of the MST is the best. This paper focuses on Cameron's response the epistemic objection. It considers two of Cameron's arguments: that a standard version of the MST can successfully resist the epistemic objection, and that Cameron's preferred version of the MST has an additional avenue (...) open to it for resisting the objection, one that is consistent with an appealing account of truthmaking. I argue that neither argument succeeds....By SMST, I shall mean the view that there exists a static four-dimensional block of events such that if an event ever... (shrink)

Ross Cameron proposes to reconcile presentism and truth-maker theory by invoking temporal distributional properties, instantiated by present entities, as the truth-makers for truths about the past. This chapter argues that Cameron's proposal fails because objects can change which temporal distributional properties they instantiate and this entails that the truth-values of truths about the past can change in an objectionable way.

Ross Cameron (2022) argues that metaphysical infinitists should reject the generally accepted idea that metaphysical determination relations back metaphysical explanations. Otherwise it won’t be possible for them to come up with successful explanations for the existence of dependent entities in non-wellfounded chains of dependence. I argue that his argument suffers from what he calls the finitist dogma, although indirectly so. However, there is a better way of motivating Cameron’s conclusion. Assuming Cameron’s principle of Essence, explanations for (...) the existence of dependent entities turn out to be circular if determination relations back explanations. This latter argument provides a stronger case as it puts the foundationalist under significant pressure, besides putting the infinitist under some pressure, to deny the idea that determination relations back explanations. (shrink)

This paper defends Priorianism, a theory in the philosophy of time which combines three theses: first, that there is a metaphysical distinction between the present time and non-present times; second, that there are temporary propositions, that is, propositions that change in truth-value simpliciter over time; and third, that there is change over time only if there are temporary propositions. Priorianism is accepted by many Presentists, Growing Block Theorists, and Moving Spotlight Theorists. However, it is difficult to defend the view without (...) appealing to premises that those who reject the view find controversial. My aim in this paper is to defend Priorianism in a way that largely avoids appealing to such premises. I do three things: first (Section 1), I describe the component theses of Priorianism and the relations between them. Next (Section 2), I show how Priorians can respond to the argument that the B-theory implies that there are temporary propositions, and therefore satisfies the Priorian condition for there being change over time. Finally (Section 3), I defend the Priorian thesis that there is change over time only if there are temporary propositions against an alternative principle of change defended by Ross Cameron (The Moving Spotlight, 2015). (shrink)

The Growing Block Theory of time says that the metaphysical openness of the future should be understood in terms of there not being any future objects or events. But in a series of works, Ross Cameron, Elizabeth Barnes, and Robbie Williams have developed a competing view that understands metaphysical openness in terms of it being indeterminate whether there exist future objects or events. I argue that the three reasons they give for preferring their account are not compelling. And (...) since the notion of “indeterminate existence” suffers conceptual problems, the Growing Block is the preferable view. (shrink)

According to truthmaker theory, particular truths are true in virtue of the existence of particular entities. Truthmaker maximalism holds that this is so for all truths. Negative existential and other ‘negative’ truths threaten the position. Despite this, maximalism is an appealing thesis for truthmaker theorists. This motivates interest in parsimonious maximalist theories, which do not posit extra entities for truthmaker duty. Such theories have been offered by David Lewis and Gideon Rosen, Ross Cameron, and Jonathan Schaffer. But these (...) theories cannot be sustained, I’ll argue, and hence maximalism comes with a serious ontological cost. Neither Armstrong’s invocation of totality facts nor the Martin-Kukso line on absences can meet this cost satisfactorily. I’ll claim that negative facts are the best (and perhaps only) way out of the problem for the truthmaker maximalist. (shrink)

I apply the notion of truthmaking to the topic of fundamentality by articulating a truthmaker theory of fundamentality according to which some truths are truth-grounded in certain entities while the ones that don't stand in a metaphysical-semantic relation to the truths that do. I motivate this view by critically discussing two problems with Ross Cameron's truthmaker theory of fundamentality. I then defend this view against Theodore Sider's objection that the truthmaking approach to fundamentality violates the purity constraint. Truthmaker (...) theorists can have a trouble-free theory of fundamentality. (shrink)

Presentists face a familiar problem. If only present objects exist, then what 'makes true' our true claims about the past? According to Ross Cameron, the 'truth-makers' for past and future tensed propositions are presently instantiated Temporal Distributional Properties. We present an argument against Cameron's view. There are two ways that we might understand the term 'distribute' as it appears. On one reading, the resulting properties are not up to the task of playing the truth-maker role; on the (...) other, the properties are incompatible with presentism. (shrink)

The Epistemic Objection says that certain theories of time imply that it is impossible to know which time is absolutely present. Standard presentations of the Epistemic Objection are elliptical—and some of the most natural premises one might fill in to complete the argument end up leading to radical skepticism. But there is a way of filling in the details which avoids this problem, using epistemic safety. The new version has two interesting upshots. First, while Ross Cameron alleges that (...) the Epistemic Objection applies to presentism as much as to theories like the growing block, the safety version does not overgeneralize this way. Second, the Epistemic Objection does generalize in a different, overlooked way. The safety objection is a serious problem for a widely held combination of views: “propositional temporalism” together with “metaphysical eternalism”. (shrink)

Truthmaker says that things, broadly construed, are the ontological grounds of truth and, therefore, that things make truths true. Recently, there have been a number of arguments purporting to show that if one embraces Truthmaker, then one ought to embrace Truthmaker Maximalism—the view that all non-analytic propositions have truthmakers. But then if one embraces Truthmaker, one ought to think that negative existentials have truthmakers. I argue that this is false. I begin by arguing that recent attempts by Ross (...) class='Hi'>Cameron and Jonathan Schaffer to provide negative existentials with truthmakers fail. I then argue that the conditional—if one embraces Truthmaker, the one ought to embrace Truthmaker Maximalism—is false by considering worlds where very little, if anything at all, exists. The conclusion is that thinking that negative existentials do not have truthmakers, and therefore rejecting Truthmaker Maximalism, need not worry Truthmaker embracers. (shrink)

Some facts ground other facts. Some fact is fundamental iff there are no other facts which partially or fully ground that fact. According to metaphysical foundationalism, every non-fundamental fact is fully grounded by some fundamental fact(s). In this paper I examine and defend some neglected considerations which might be made in favor of metaphysical foundationalism. Building off of work by Ross Cameron, I suggest that foundationalist theories are more unified than, and so in one important respect simpler than, (...) non-foundationalist theories, insofar as foundationalist theories allow us to derive all non-fundamental facts from some fundamental fact(s). Non-foundationalist theories can enjoy a similar sort of theoretical unification only by taking on objectionable metaphysical laws. (shrink)

We regularly make graded normative judgements in the epistemic domain. Recent work in the literature examines degrees of justification, degrees of rationality, and degrees of assertability. This paper addresses a different dimension of the gradeability of epistemic normativity, one that has been given little attention. How should we understand degrees of epistemic criticizability? In virtue of what sorts of factors can one epistemic failing be worse than another? The paper develops a dual-factor view of degrees of epistemic criticizability. According to (...) the view, degrees of epistemic criticizability are (i) an inverse function of degrees of doxastic justification and (ii) a function of degrees of agent culpability. The paper develops an account of each factor, and explains how they should be weighted. The paper also addresses the importance of modelling degrees of epistemic criticizability in a broader context. I focus on the role that such a model can play in the ethics of epistemic criticism. (shrink)

I respond to Tim Smartt’s (2023) skepticism about epistemic blame. Smartt’s skepticism is based on the claims that (i) mere negative epistemic evaluation can better explain everything proponents of epistemic blame say we need epistemic blame to explain; and (ii) no existing account of epistemic blame provides a plausible account of the putative force that any response deserving the label “blame” ought to have. He focuses primarily on the prominent “relationship-based” account of epistemic blame to defend these claims, arguing that (...) the account is explanatorily idle, and cannot distinguish between epistemically excused and epistemically blameworthy agents. I argue that Smartt mischaracterizes the account’s role for judgments of epistemic relationship impairment, leading to mistaken claims about the account’s predictions. I also argue that the very feature of the account that Smartt mischaracterizes is key to understanding what epistemic blame does for our epistemic responsibility practices that mere negative epistemic evaluation cannot. (shrink)

For A. J. Ayer, the occurrence of delusions confutes the notion that we perceive the world directly. He argues instead that perceptions are caused by immaterial “sense data” which somehow represent the properties of material things to us in our experiences. J. L. Austin systematically rejects Ayer’s claims, arguing that the occurrence of delusions does not preclude the possibility of direct perception, and that, indeed, our normal perception is direct. I challenge both philosophers’ ideas by examining how they deal with (...) the phenomenon of colour. (shrink)

Why comply with epistemic norms? In this paper, I argue that complying with epistemic norms, engaging in epistemically responsible conduct, and being epistemically trustworthy are constitutive elements of maintaining good epistemic relations with oneself and others. Good epistemic relations are in turn both instrumentally and finally valuable: they enable the kind of coordination and knowledge acquisition underpinning much of what we tend to associate with a flourishing human life; and just as good interpersonal relations with others can be good for (...) their own sake, standing in good epistemic relations is good for its own sake. On my account, we have reason to comply with epistemic norms because it is a way of respecting the final value of something that also tends to be an instrumentally valuable thing: good epistemic relations. Situating the account within the recent social turn in debates about epistemic instrumentalism, I argue that the dual‐value aspect of good epistemic relations can explain important anti‐instrumentalist intuitions, in a well‐motivated way, within a broadly instrumentalist framework. (shrink)

Does Aristotle offer a definition of the soul? In fact, he rejects the possibility of defining the soul univocally. Because “life” is a homonymous concept, so too is “soul”. Given the specific causal role that Aristotle envisages for form and essence, the soul requires multiple different definitions to capture how it functions as a cause in each form of life. Aristotle suggests demonstrations can be given which express these causal definitions; I reconstruct these demonstrations.

There is a distinction between merely having the right belief, and further basing that belief on the right reasons. Any adequate epistemology needs to be able to accommodate the basing relation that marks this distinction. However, trouble arises for Bayesianism. I argue that when we combine Bayesianism with the standard approaches to the basing relation, we get the result that no agent forms their credences in the right way; indeed, no agent even gets close. This is a serious problem, for (...) it prevents us from making epistemic distinctions between agents that are doing a reasonably good job at forming their credences and those that are forming them in clearly bad ways. I argue that if this result holds, then we have a problem for Bayesianism. However, I show how the Bayesian can avoid this problem by rejecting the standard approaches to the basing relation. By drawing on recent work on the basing relation, we can develop an account of the relation that allows us to avoid the result that no agent comes close to forming their credences in the right way. The Bayesian can successfully accommodate the basing relation. (shrink)

Causal essentialists hold that a property essentially bears its causal and nomic relations. Further, as many causal essentialists have noted, the main motivations for causal essentialism also motivate holding that properties are individuated in terms of their causal and nomic relations. This amounts to a kind of identity of indiscernibles thesis; properties that are indiscernible with respect to their causal and nomic relations are identical. This can be compared with the more well-known identity of indiscernibles thesis, according to which particulars (...) that are qualitatively indiscernible are identical. Robert Adams has developed a well-known objection to this thesis by considering a series of possibilities involving nearly qualitatively indiscernible particulars that naturally leads to a possibility involving qualitatively indiscernible particulars. I argue that we can construct parallel cases involving a series of possibilities involving properties that are nearly indiscernible with respect to their causal and nomic relations that naturally lead to possibilities involving properties that are indiscernible with respect to their causal and nomic relations. The same features that make Adams’ argument forceful also carry over to my cases, giving us a powerful objection to the causal essentialist identity of indiscernibles thesis. (shrink)

Action must be taken to combat climate change. Yet, how the costs of climate action should be allocated among states remains a question. One popular answer—the polluter-pays principle (PPP)—stipulates that those responsible for causing the problem should pay to address it. While intuitively plausible, the PPP has been subjected to withering criticism in recent years. It is timely, following the Paris Agreement, to develop a new version: one that does not focus on historical production-based emissions but rather allocates climate burdens (...) in proportion to each state’s annual consumption-based emissions. This change in carbon accounting results in a fairer and more environmentally effective principle for distributing climate duties. (shrink)

Evaluating counterfactuals in worlds with deterministic laws poses a puzzle. In a wide array of cases, it does not seem plausible that if a non-actual event were to occur that either the past would be different or that the laws would be different. But it’s also difficult to see how we can avoid this result. Some philosophers have argued that we can avoid this dilemma by allowing that a proposition can be a law even though it has violations. On this (...) view, for the relevant cases, the past and the laws would still hold, but the laws would have a violation. In this paper, I raise a problem for the claim that the laws and the past are preserved for all of the relevant counterfactual antecedents. I further argue that this problem undermines motivating the possibility of violations on the grounds that they allow us to hold that the past and the laws are typically counterfactually preserved, even if they are not always preserved. (shrink)

Plenitude, roughly, the thesis that for any non-empty region of spacetime there is a material object that is exactly located at that region, is often thought to be part and parcel of the standard Lewisian package in the metaphysics of persistence. While the wedding of plentitude and Lewisian four-dimensionalism is a natural one indeed, there are a hand-full of dissenters who argue against the notion that Lewisian four-dimensionalism has exclusive rights to plentitude. These ‘promiscuous’ three-dimensionalists argue that a temporalized version (...) of plenitude is entirely compatible with a three-dimensional ontology of enduring entities. While few would deny the coherence of such a position, and much work has been done by its proponents to appease critics, there has been surprisingly little by way of exploring the various forms such an ontology might take as well as the potential advantages of one plenitudinous three-dimensional ontology over another. Here I develop a novel form of plenitudinous three-dimensionalism, what John Hawthorne (Metaphysical essays, 2006a, b) has called “Neo-Aristotelian Plenitude,” and argue that if one is inclined to endorse an abundant three-dimensional ontology, one is wise to opt for a plenitude of accidental unities. (shrink)

In this short paper, I critically examine Veli Mitova’s proposal that social-identity groups can have collective epistemic reasons. My primary focus is the role of privileged access in her account of how collective reasons become epistemic reasons for social-identity groups. I argue that there is a potentially worrying structural asymmetry in her account of two different types of cases. More specifically, the mechanisms at play in cases of “doxastic reasons” seem fundamentally different from those at play in cases of “epistemic-conduct (...) reasons”. The upshot is a need for further explanation of what unifies these dimensions of the account. (shrink)

Is authoritarian power ever legitimate? The contemporary political theory literature—which largely conceptualizes legitimacy in terms of democracy or basic rights—would seem to suggest not. I argue, however, that there exists another, overlooked aspect of legitimacy concerning a government’s ability to ensure safety and security. While, under normal conditions, maintaining democracy and rights is typically compatible with guaranteeing safety, in emergency situations, conflicts between these two aspects of legitimacy can and often do arise. A salient example of this is the COVID-19 (...) pandemic, during which severe limitations on free movement and association have become legitimate techniques of government. Climate change poses an even graver threat to public safety. Consequently, I argue, legitimacy may require a similarly authoritarian approach. While unsettling, this suggests the political importance of climate action. For if we wish to avoid legitimating authoritarian power, we must act to prevent crises from arising that can only be resolved by such means. (shrink)

Is there a distinctively epistemic kind of blame? It has become commonplace for epistemologists to talk about epistemic blame, and to rely on this notion for theoretical purposes. But not everyone is convinced. Some of the most compelling reasons for skepticism about epistemic blame focus on disanologies, or asymmetries, between the moral and epistemic domains. In this paper, I defend the idea that there is a distinctively epistemic kind of blame. I do so primarily by developing an account of the (...) nature of epistemic blame. My account draws on a prominent line of theorizing in moral philosophy that ties blame to our relationships with one another. I argue that with my account of epistemic blame on hand, the most compelling worries about epistemic blame can be deflated. There is a distinctively epistemic kind of blame. (shrink)

I first offer a broad taxonomy of models of divine omnipresence in the Christian tradition, both past and present. I then examine the recent model proposed by Hud Hudson (2009, 2014) and Alexander Pruss (2013)—ubiquitous entension—and flag a worry with their account that stems from predominant analyses of the concept of ‘material object’. I then attempt to show that ubiquitous entension has a rich Latin medieval precedent in the work of Augusine and Anselm. I argue that the model of omnipresence (...) explicated by Augustine and Anselm has the resources to avoid the noted worry by offering an alternative account of the divide between the immaterial and the material. I conclude by considering a few alternative analyses of ‘material object’ that make conceptual room for a contemporary Christian theist to follow suite in thinking that at least some immaterial entities are literally spatially located when relating to the denizens of spacetime. (shrink)

There is a widely accepted distinction between being directly responsible for a wrongdoing versus being somehow indirectly or vicariously responsible for the wrongdoing of another person or collective. Often this is couched in analyses of complicity, and complicity’s role in the relationship between individual and collective wrongdoing. Complicity is important because, inter alia, it allows us to make sense of individuals who may be blameless or blameworthy to a relatively low degree for their immediate conduct, but are nevertheless blameworthy to (...) a higher degree for their implication in some larger (or another person’s) wrongdoing. In this paper, I argue that there is a distinctively epistemic kind of complicity. First, I motivate the distinction between direct and vicarious responsibility with three interlocking arguments, respectively appealing to: i) the structure of rational agency; ii) linguistic considerations; iii) the role of “principal vs. accomplice” in legal doctrine. I show how these arguments naturally extend to the epistemic domain, motivating an epistemic form of vicarious responsibility. I then examine complicity as a mechanism of vicarious epistemic responsibility. To fill this out, I engage with an epistemic analogue of the debate about the role of intention versus causal contribution in complicity. I propose a Casual Account of Epistemic Complicity, arguing that it accommodates a wide range of cases, and enables fine-grained explanations of degrees of culpability for epistemic complicity. With an adequate account of epistemic complicity on hand, we can explain what is objectionable about an important class of epistemic agent who, on an individual level, may be epistemically blameless or blameworthy to a relatively low degree, but whose relation to other individuals or collectives nevertheless makes them epistemically blameworthy to a higher degree. I explore some broader implications of this result. (shrink)

In artificial intelligence, recent research has demonstrated the remarkable potential of Deep Convolutional Neural Networks (DCNNs), which seem to exceed state-of-the-art performance in new domains weekly, especially on the sorts of very difficult perceptual discrimination tasks that skeptics thought would remain beyond the reach of artificial intelligence. However, it has proven difficult to explain why DCNNs perform so well. In philosophy of mind, empiricists have long suggested that complex cognition is based on information derived from sensory experience, often appealing to (...) a faculty of abstraction. Rationalists have frequently complained, however, that empiricists never adequately explained how this faculty of abstraction actually works. In this paper, I tie these two questions together, to the mutual benefit of both disciplines. I argue that the architectural features that distinguish DCNNs from earlier neural networks allow them to implement a form of hierarchical processing that I call “transformational abstraction”. Transformational abstraction iteratively converts sensory-based representations of category exemplars into new formats that are increasingly tolerant to “nuisance variation” in input. Reflecting upon the way that DCNNs leverage a combination of linear and non-linear processing to efficiently accomplish this feat allows us to understand how the brain is capable of bi-directional travel between exemplars and abstractions, addressing longstanding problems in empiricist philosophy of mind. I end by considering the prospects for future research on DCNNs, arguing that rather than simply implementing 80s connectionism with more brute-force computation, transformational abstraction counts as a qualitatively distinct form of processing ripe with philosophical and psychological significance, because it is significantly better suited to depict the generic mechanism responsible for this important kind of psychological processing in the brain. (shrink)

This paper provides a critical overview of recent work on epistemic blame. The paper identifies key features of the concept of epistemic blame and discusses two ways of motivating the importance of this concept. Four different approaches to the nature of epistemic blame are examined. Central issues surrounding the ethics and value of epistemic blame are identified and briefly explored. In addition to providing an overview of the state of the art of this growing but controversial field, the paper highlights (...) areas where future work is needed. (shrink)

One typical aim of responsibilist virtue epistemology is to employ the notion of intellectual virtue in pursuit of an ameliorative epistemology. This paper focuses on “political inquiry” as a case study for examining the ameliorative value of intellectual virtue. The main claim is that the case of political inquiry threatens to expose responsibilist virtue epistemology in a general way as focusing too narrowly on the role of individual intellectual character traits in attempting to improve our epistemic practices.

Under what conditions would an artificially intelligent system have wellbeing? Despite its clear bearing on the ethics of human interactions with artificial systems, this question has received little direct attention. Because all major theories of wellbeing hold that an individual’s welfare level is partially determined by their mental life, we begin by considering whether artificial systems have mental states. We show that a wide range of theories of mental states, when combined with leading theories of wellbeing, predict that certain existing (...) artificial systems have wellbeing. Along the way, we argue that there are good reasons to believe that artificial systems can have wellbeing even if they are not phenomenally conscious. While we do not claim to demonstrate conclusively that AI systems have wellbeing, we argue that there is a significant probability that some AI systems have or will soon have wellbeing, and that this should lead us to reassess our relationship with the intelligent systems we create. (shrink)

One challenge in developing an account of the nature of epistemic blame is to explain what differentiates epistemic blame from mere negative epistemic evaluation. The challenge is to explain the difference, without invoking practices or behaviors that seem out of place in the epistemic domain. In this paper, I examine whether the most sophisticated recent account of the nature of epistemic blame—due to Jessica Brown—is up for the challenge. I argue that the account ultimately falls short, but does so in (...) an instructive way. Drawing on the lessons learned, I put forward an alternative approach to the nature of epistemic blame. My account understands epistemic blame in terms of modifications to the intentions and expectations that comprise our “epistemic relationships” with one another. This approach has a number of attractions shared by Brown’s account, but it can also explain the significance of epistemic blame. (shrink)

Pessimistic empirical evidence about the reformatory and deterrent effects of punitive treatment poses a challenge for all justificatory theories of punishment. Yet, the dominant progressive view remains that punishment is required for the most serious crimes. This paper outlines an empirically sensitive prospectus for justifying punitive treatment through understanding the importance of reintegration. On this view, punishment can be viewed as a preferred alternative to the rigours of social ostracism, a common way of dealing with offenders in lieu of formal (...) criminal justice. Adopting reintegration as a primary aim encourages taking a longer view which focuses on desistance from criminality rather than only on reform at the point of release from formal punishment. The view outlined in this paper enables a vindication of reintegrative punishment even when it is not the most immediately efficient means of reforming offenders. In making this argument, I develop the modern retributive platform in criminal justice theory, identify various overlooked yet key nuances in the relationship between reintegration and reform, and argue for greater theoretical and practical attention to how the state can make the communities they serve more receptive to reintegrating offenders. (shrink)

Recent advances in natural language processing have given rise to a new kind of AI architecture: the language agent. By repeatedly calling an LLM to perform a variety of cognitive tasks, language agents are able to function autonomously to pursue goals specified in natural language and stored in a human-readable format. Because of their architecture, language agents exhibit behavior that is predictable according to the laws of folk psychology: they function as though they have desires and beliefs, and then make (...) and update plans to pursue their desires given their beliefs. We argue that the rise of language agents significantly reduces the probability of an existential catastrophe due to loss of control over an AGI. This is because the probability of such an existential catastrophe is proportional to the difficulty of aligning AGI systems, and language agents significantly reduce that difficulty. In particular, language agents help to resolve three important issues related to aligning AIs: reward misspecification, goal misgeneralization, and uninterpretability. (shrink)

It is generally assumed that existing artificial systems are not phenomenally conscious, and that the construction of phenomenally conscious artificial systems would require significant technological progress if it is possible at all. We challenge this assumption by arguing that if Global Workspace Theory (GWT) — a leading scientific theory of phenomenal consciousness — is correct, then instances of one widely implemented AI architecture, the artificial language agent, might easily be made phenomenally conscious if they are not already. Along the way, (...) we articulate an explicit methodology for thinking about how to apply scientific theories of consciousness to artificial systems and employ this methodology to arrive at a set of necessary and sufficient conditions for phenomenal consciousness according to GWT. (shrink)

The field of AI safety seeks to prevent or reduce the harms caused by AI systems. A simple and appealing account of what is distinctive of AI safety as a field holds that this feature is constitutive: a research project falls within the purview of AI safety just in case it aims to prevent or reduce the harms caused by AI systems. Call this appealingly simple account The Safety Conception of AI safety. Despite its simplicity and appeal, we argue that (...) The Safety Conception is in tension with at least two trends in the ways AI safety researchers and organizations think and talk about AI safety: first, a tendency to characterize the goal of AI safety research in terms of catastrophic risks from future systems; second, the increasingly popular idea that AI safety can be thought of as a branch of safety engineering. Adopting the methodology of conceptual engineering, we argue that these trends are unfortunate: when we consider what concept of AI safety it would be best to have, there are compelling reasons to think that The Safety Conception is the answer. Descriptively, The Safety Conception allows us to see how work on topics that have historically been treated as central to the field of AI safety is continuous with work on topics that have historically been treated as more marginal, like bias, misinformation, and privacy. Normatively, taking The Safety Conception seriously means approaching all efforts to prevent or mitigate harms from AI systems based on their merits rather than drawing arbitrary distinctions between them. (shrink)

Epistemological Disjunctivism is a view about paradigm cases of perceptual knowledge. Duncan Pritchard claims that it is particularly well suited to accounting for internalist and externalist intuitions. A number of authors have disputed this claim, arguing that there are problems for Pritchard’s way with internalist intuitions. I share the worry. However, I don’t think it has been expressed as effectively as it can be. My aim in this paper is to present a new way of formulating the worry, in terms (...) of an “explanatory challenge”. The explanatory challenge is a simple, yet powerful and illuminating challenge for Epistemological Disjunctivism. It is illuminating in the sense that it shows us why Epistemological Disjunctivism must take on certain internalistically problematic commitments. A secondary aim of this paper is to examine whether the recently much-discussed distinction between justifications and excuses in epistemology can support an adequate response. I will argue that it cannot. (shrink)

Existing work on gaslighting ties it constitutively to facts about the intentions or prejudices of the gaslighter and/or his victim’s prior experience of epistemic injustice. I argue that the concept of gaslighting is more broadly applicable than has been appreciated: what is distinctive about gaslighting, on my account, is simply that a gaslighter confronts his victim with a certain kind of choice between rejecting his testimony and doubting her own basic epistemic competence in some domain. I thus hold that gaslighting (...) is a purely epistemic phenomenon—not requiring any particular set of intentions or prejudices on the part of the gaslighter—and also that it can occur even in the absence of any prior experience of epistemic injustice. Appreciating the dilemmatic character of gaslighting allows us to understand its connection with a characteristic sort of epistemic harm, makes it easier to apply the concept of gaslighting in practice, and raises the possibility that we might discover its structure and the associated harm in surprising places. (shrink)

Contextology is the science of the dynamics of the conversational context. Contextology formulates laws governing how the shared information states of interlocutors evolve in response to assertion. More precisely, the contextologist attempts to construct a function which, when provided with just a conversation’s pre-update context and the content of an assertion, delivers that conversation’s post-update context. Most contextologists have assumed that the function governing the evolution of the context is simple: the post-update context is just the pre-update context intersected with (...) the content of the assertion. We argue that this assumption is wrong: not only is it false, it is also incoherent given standard contextological assumptions. Moreover, it is impossible in principle to revise it to correctly describe the dynamics of context. We conclude that there can be no science of Contextology. The laws governing the evolution of the context in response to assertion must make essential reference to the private information states of interlocutors. (shrink)

How should we determine the distribution of psychological traits—such as Theory of Mind, episodic memory, and metacognition—throughout the Animal kingdom? Researchers have long worried about the distorting effects of anthropomorphic bias on this comparative project. A purported corrective against this bias was offered as a cornerstone of comparative psychology by C. Lloyd Morgan in his famous “Canon”. Also dangerous, however, is a distinct bias that loads the deck against animal mentality: our tendency to tie the competence criteria for cognitive capacities (...) to an exaggerated sense of typical human performance. I dub this error “anthropofabulation”, since it combines anthropocentrism with confabulation about our own prowess. Anthropofabulation has long distorted the debate about animal minds, but it is a bias that has been little discussed and against which the Canon provides no protection. Luckily, there is a venerable corrective against anthropofabulation: a principle offered long ago by David Hume, which I call “Hume’s Dictum”. In this paper, I argue that Hume’s Dictum deserves a privileged place next to Morgan’s Canon in the methodology of comparative psychology, illustrating my point through a discussion of the debate over Theory of Mind in nonhuman animals. (shrink)

The paper critically examines recent work on justifications and excuses in epistemology. I start with a discussion of Gerken’s claim that the “excuse maneuver” is ad hoc. Recent work from Timothy Williamson and Clayton Littlejohn provides resources to advance the debate. Focusing in particular on a key insight in Williamson’s view, I then consider an additional worry for the so-called excuse maneuver. I call it the “excuses are not enough” objection. Dealing with this objection generates pressure in two directions: one (...) is to show that excuses are a positive enough normative standing to help certain externalists with important cases; the other is to do so in a way that does not lead back to Gerken’s objection. I show how a Williamson-inspired framework is flexible enough to deal with both sources of pressure. Perhaps surprisingly, I draw on recent virtue epistemology. (shrink)

Our prominent definitions of cognition are too vague and lack empirical grounding. They have not kept up with recent developments, and cannot bear the weight placed on them across many different debates. I here articulate and defend a more adequate theory. On this theory, behaviors under the control of cognition tend to display a cluster of characteristic properties, a cluster which tends to be absent from behaviors produced by non-cognitive processes. This cluster is reverse-engineered from the empirical tests that comparative (...) psychologists use to determine whether a behavior was generated by a cognitive or a non-cognitive process. Cognition should be understood as the natural kind of psychological process that non-accidentally exhibits the properties assessed by these tests (as well as others we have not yet discovered). Finally, I review two plausible neural accounts of cognition's underlying mechanisms?one based in localization of function to particular brain regions and another based in the more recent distributed networks approach to neuroscience?which would explain why these properties non-accidentally cluster. While this notion of cognition may be useful for a number of debates, I here focus on its application to a recent crisis over the distinction between cognition and association in comparative psychology. (shrink)

Recent philosophical work on causation has focused on distinctions across types of causal relationships. This paper argues for another distinction that has yet to receive attention in this work. This distinction has to do with whether causal relationships have “material continuity,” which refers to the reliable movement of material from cause to effect. This paper provides an analysis of material continuity and argues that causal relationships with this feature are associated with a unique explanatory perspective, are studied with distinct causal (...) investigative methods, and provide different types of causal control over their effects. (shrink)

A plausible condition on having the standing to blame someone is that the target of blame's wrongdoing must in some sense be your “business”—the wrong must in some sense harm or affect you, or others close to you. This is known as the business condition on standing to blame. Many cases of epistemic blame discussed in the literature do not obviously involve examples of someone harming or affecting another. As such, not enough has been said about how an individual's epistemic (...) failing can really count as another person's business. In this paper, I deploy a relationship-based account of epistemic blame to clarify the conditions under which the business condition can be met in the epistemic domain. The basic idea is that one person's epistemic failing can be another's business in virtue of the way it impairs their epistemic relationship. (shrink)

This paper compares two alternative explanations of pragmatic encroachment on knowledge (i.e., the claim that whether an agent knows that p can depend on pragmatic factors). After reviewing the evidence for such pragmatic encroachment, we ask how it is best explained, assuming it obtains. Several authors have recently argued that the best explanation is provided by a particular account of belief, which we call pragmatic credal reductivism. On this view, what it is for an agent to believe a proposition is (...) for her credence in this proposition to be above a certain threshold, a threshold that varies depending on pragmatic factors. We show that while this account of belief can provide an elegant explanation of pragmatic encroachment on knowledge, it is not alone in doing so, for an alternative account of belief, which we call the reasoning disposition account, can do so as well. And the latter account, we argue, is far more plausible than pragmatic credal reductivism, since it accords far better with a number of claims about belief that are very hard to deny. (shrink)