The thesis of instrumental convergence holds that a wide range of ends have common means: for instance, self preservation, desire preservation, self improvement, and resource acquisition. Bostrom contends that instrumental convergence gives us reason to think that "the default outcome of the creation of machine superintelligence is existential catastrophe". I use the tools of decision theory to investigate whether this thesis is true. I find that, even if intrinsic desires are randomly selected, instrumental rationality induces biases towards certain kinds of (...) choices. Firstly, a bias towards choices which leave less up to chance. Secondly, a bias towards desire preservation, in line with Bostrom's conjecture. And thirdly, a bias towards choices which afford more choices later on. I do not find biases towards any other of the convergent instrumental means on Bostrom's list. I conclude that the biases induced by instrumental rationality at best weakly support Bostrom's conclusion that machine superintelligence is likely to lead to existential catastrophe. (shrink)

Choices confront us with questions. How we act depends on our answers to those questions. So the way our beliefs guide our choices is not just a function of their informational content, but also depends systematically on the questions those beliefs address. This paper gives a precise account of the interplay between choices, questions and beliefs, and harnesses this account to obtain a principled approach to the problem of deduction. The result is a novel theory of belief-guided action that explains (...) and predicts the decisions of agents who, like ourselves, fail to be logically omniscient: that is, of agents whose beliefs may not be deductively closed, or even consistent. (Winner of the 2021 Isaac Levi Prize.). (shrink)

Infallibilism leads to skepticism, and fallibilism is plagued by the threshold problem. Within this narrative, the pragmatic turn in epistemology has been marketed as a way for fallibilists to address the threshold problem. In contrast, pragmatic versions of infallibilism have been left unexplored. However, I propose that going pragmatic offers the infallibilist a way to address its main problem, the skeptical problem. Pragmatic infallibilism, however, is committed to a shifty view of epistemic certainty, where the strength of a subject’s epistemic (...) state can vary depending upon the practical context. Pragmatic views of epistemic strength are quite radical so the argumentative goal of my discussion is to argue for its plausibility, thereby arguing for the plausibility of pragmatic infallibilism. To do so, I discuss the role that the framing of decision problems and the construction of our deliberative attitudes have in our theories of rational choice. And I explore a way in which these necessary components of a normative theory of rationality can be used to motivate pragmatic views of evidential support. Finally, since the plausibility of our pragmatic account will depend upon the plausibility of a constructive account of rational choice, I conclude by comparing the adequacy of the constructive account with orthodox Bayesian and knowledge-based accounts. (shrink)

Causation appears to present us with an interpretative difficulty. While arguably a redundant relation given fundamental physics, it is nevertheless apparently pragmatically indispensable. This chapter revisits certain arguments made previously by the author for these claims with the benefit of hindsight, starting with the role of causal models in the human sciences, and attempting to explain why it is not possible to straightforwardly ground such models in fundamental physics. This suggests that further constraints, going beyond physics, are needed to legitimate (...) such models. These supplementary constraints could be reified, but that would seem to conflict with the completeness of physics. A response is to emphasize the practical role of causal talk, and the author suggests that a fictionalist approach might be worth exploring. After clarifying fictionalism as a general approach, they carry out in some detail the project of clarifying what a fictionalist attitude to causation would involve. (shrink)

David Lewis rejected consequentialism in ethics. However, two aspects of his meta-ethical views make it a challenge to see how consequentialism could be resisted. Lewis endorses a maximising conception of rationality, where to be rational is to maximise value of a certain sort; he appears to think it is possible to be both rational and moral; and yet he rejects conceptions of moral action as acting to maximise moral value. The second tension in Lewis's views arises from his meta-ethics. Lewis's (...) naturalisation of morality is in terms of a naturalised account of value, including moral value. Insofar as he offers a motivational story for moral judgements, it is one that goes via a motivational connection to value. Yet his anti-consequentialism seems to commit him to rejecting a story of moral action entirely in terms of pursuit of moral value. In both cases there are things to be said about how to weave these strands of Lewis's thought together. After setting out these two apparent tensions in Lewis's thought about ethics, the paper discusses what a Lewisian could best say about these puzzles. Lewisians do have a number of plausible options for dealing with the first challenge. While there are things a Lewisian could attempt to extend Lewis's moral naturalism in a way that could deal with the second challenge, it is harder to see how to do this in a satisfying way. Attention then turns to what Lewis said in his letters on these topics: on that basis, perhaps Lewis was close to being an agent-relative consequentialist after all. (shrink)

This paper argues that expected utility theory for actions in chancy environments should be formulated in terms of centered chances. The subjective expected utility of an option A may be seen as a weighted sum of the utilities of A in different possible worlds, with weights being the credences that the agent assigns to these worlds. The utility of A in a given world is then definable as a weighted sum of the values of A’s different possible outcomes, with weights (...) being the conditional chances of these outcomes if A were performed. On the centered-chance view, the chances to be used as weights in the definition of utility are centered. Unlike ordinary chances, centered chances depend not only on what happens prior to the agent’s choice but also on the events that occur after the choice. Thus, to give an example, suppose that the action under consideration results in a bad outcome due to some event whose ordinary chance of occurring was very low at the time of choice. Then the utility of that action in the actual world could be high on the non-centered view, but on the centered view that utility is negative, since the centered chance of the event in question given the action was one, given that it did actually take place. A precise definition of centered chances is not easy to frame, but the concept can be made intuitively clear. The resulting decision theory is, in my opinion, philosophically more satisfactory than the extant proposals, even though it doesn’t differ much in its practical recommendations, with the exception of some rather peculiar cases. (shrink)

This dissertation looks at a set of interconnected questions concerning the foundations of probability, and gives a series of interconnected answers. At its core is a piece of old-fashioned philosophical analysis, working out what probability is. Or equivalently, investigating the semantic question of what is the meaning of ‘probability’? Like Keynes and Carnap, I say that probability is degree of reasonable belief. This immediately raises an epistemological question, which degrees count as reasonable? To solve that in its full generality would (...) mean the end of human inquiry, so that won’t be attempted here. Rather I will follow tradition and merely investigate which sets of partial beliefs are coherent. -/- The standard answer to this question, what is commonly called the Bayesian answer, says that degrees of belief are coherent iff they form a probability function. I disagree with the way this is usually justified, but subject to an important qualification I accept the answer. The important qualification is that degrees of belief may be imprecise, or vague. Part one of the dissertation, chapters 1 to 6, looks largely at the consequences of this qualification for the semantic and epistemological questions already mentioned. It turns out that when we allow degrees of belief to be imprecise, we can discharge potentially fatal objections to some philosophically attractive theses. Two of these, that probability is degree of reasonable belief and that the probability calculus provides coherence constraints on partial beliefs, have been mentioned. Others include the claim, defended in chapter 4, that chance is probability given total history. -/- As well as these semantic and epistemological questions, studies of the foundations of probability usually include a detailed discussion of decision theory. For reasons set out in chapter 2, I deny we can gain epistemological insights from decision theory. Nevertheless, it is an interesting field to study on its own, and it might be expected that there would be decision theoretic consequences of allowing imprecise degrees of belief. As I show in part two, this expectation seems to be mistaken. Chapter 9 shows that there aren’t interesting consequences of this theory for decision theory proper, and chapters 10 and 11 show that Keynes’s attempt to use imprecision in degrees of belief to derive a distinctive theory of interest rates is unsound. -/- Chapters 7 and 8 provide a link between these two parts. In chapter 7 I look at some previous philosophical investigations into the effects of imprecision. In chapter 8 I develop what I take to be the best competitor to the theory defended here – a constructivist theory of probability. On this view degrees of belief are precise, but the relevant coherence constraint is a constructivist probability calculus. This view is, I think, mistaken, but the calculus has some intrinsic interest, and there are at least enough arguments for it to warrant a chapter-length examination. (shrink)

The notion of existential catastrophe is increasingly appealed to in discussion of risk management around emerging technologies, but it is not completely clear what this notion amounts to. Here, I provide an opinionated survey of the space of plausibly useful definitions of existential catastrophe. Inter alia, I discuss: whether to define existential catastrophe in ex post or ex ante terms, whether an ex ante definition should be in terms of loss of expected value or loss of potential, and what kind (...) of probabilities should be involved in any appeal to expected value. (shrink)

Advocates of Causal Decision Theory (CDT) argue that Evidential Decision Theory (EDT) is inadequate because it gives the wrong result in Newcomb problems. Egan (2007) provides a recipe for converting Newcomb problems to counterexamples to CDT, arguing that CDT is inadequate too. Proposed by Edgington (2011), the Conditional Causal Decision Theory (CCDT) is widely taken uncritically in the recent literature as a version of CDT that conforms to the supposedly correct pre-theoretic judgments about the rationality of acts in Newcomb problems (...) and Egan cases. While the renewed attention to the debate between EDT and CDT is mostly focused on how various theories treat Egan cases, this paper scrutinizes Edgington’s treatment of Newcomb problems and argues that, despite Edgington’s promise, CCDT actually fails in Newcomb problems. This, I assume, shows that CCDT is superior neither to the classic visions of CDT nor to EDT. The paper concludes by arguing, in general, that CCDT is systematically bound to deliver the same results as does EDT in any decision problem. Despite being formulated in terms of causal notions, therefore, it collapses back into the classic EDT. (shrink)

The classic formulation of causal decision theory (CDT) appeals to counterfactuals. It says that you should aim to choose an option that would have a good outcome, were you to choose it. However, this version of CDT faces trouble if the laws of nature are deterministic. After all, the standard theory of counterfactuals says that, if the laws are deterministic, then if anything—including the choice you make—were different in the present, either the laws would be violated or the distant past (...) would be changed. And as several authors have shown, it's easy to transform this upshot of the standard theory of counterfactuals into full-blown counterexamples to CDT. In response to these counterexamples, I argue here that the problem lies, not so much with CDT's guiding idea—that it's the expected causal consequences of your actions that matter for rational decision-making—but with the fact that the classic formulation of CDT doesn't pay sufficient attention to the context-sensitivity of counterfactuals. I develop a contextualist version of CDT which better accounts for this context-sensitivity. And I show that my theory avoids the problems faced by the classic formulation of CDT in determinstic worlds. (shrink)

This volume concerns Rational Agents - humans, players in a game, software or institutions - which must decide the proper next action in an atmosphere of partial information and uncertainty. The book collects formal accounts of Uncertainty, Rationality and Agency, and also of their interaction. It will benefit researchers in artificial systems which must gather information, reason about it and then make a rational decision on which action to take.

Contro una visione prettamente strumentale della razionalità, una tesi di questo libro è che il pensiero critico non può consistere solo di abilità di pensiero deduttivo o inferenziale, ma è più in generale espressione di abilità epistemiche e competenze etiche inerenti al processo della conoscenza intesa come questione complessa, poiché relativa alla formazione della persona che conosce e agisce nel mondo secondo determinati fini, valori, credenze. Una valutazione delle competenze che non tenga conto di questa differenza genera forme di ingiustizia (...) epistemica. La proposta di questo lavoro –– alla luce di alcune riflessioni su varie tendenze dell’epistemologia contemporanea –– è pertanto che il pensiero critico nasce essenzialmente da un processo di trasmissione dell’eredità culturale e va di pari passo con una didattica guidata. Il linguaggio rappresenta un bagaglio storico, olistico e sociale della tradizione delle ragioni e delle forme di vita che costituiscono le norme o il mezzo in base ai quali gli educatori mediano la loro interpretazione di colui che apprende e lo ammettono all’interno dello spazio delle ragioni. Il passaggio è così da una concezione improntata alla formazione di abilità volte alla ricerca di una verità oggettiva, all’idea che l’approccio critico, pur senza negare una verità oggettiva, e senza ricadere in una soluzione relativista, presuppone piuttosto un punto di vista prospettico sulla realtà. (shrink)

The role of causation and counterfactuals in causal decision theory is vexed and disputed. Recently, Brian Hedden (2023) argues that we should abandon causal decision theory in favour of an alternative: counterfactual decision theory. I argue that, pace Hedden, counterfactual decision theory is not a competitor to, but rather a version of, causal decision theory – the most popular version by far. I provide textual evidence that the founding fathers of causal decision theory (Stalnaker, Gibbard, Harper, Lewis, Skyrms, Sobel, and (...) Joyce) all endorse counterfactual decision theories. I additionally discuss why these theories came to be called ‘causal’, rather than ‘counterfactual’. And I argue that, properly understood, causal decision theory escapes Hedden's objections. (shrink)

The ontology of decision theory has been subject to considerable debate in the past, and discussion of just how we ought to view decision problems has revealed more than one interesting problem, as well as suggested some novel modifications of classical decision theory. In this paper it will be argued that Bayesian, or evidential, decision-theoretic characterizations of decision situations fail to adequately account for knowledge concerning the causal connections between acts, states, and outcomes in decision situations, and so they are (...) incomplete. Second, it will be argues that when we attempt to incorporate the knowledge of such causal connections into Bayesian decision theory, a substantial technical problem arises for which there is no currently available solution that does not suffer from some damning objection or other. From a broader perspective, this then throws into question the use of decision theory as a model of human or machine planning. (shrink)

Judging an act’s causal efficacy plays a crucial role in causal decision theory. A recent development appeals to the causal modeling framework with an emphasis on the analysis of intervention based on the causal Bayes net for clarifying what causally depends on our acts. However, few writers have focused on exploring the usefulness of extending structural causal models to decision problems that are not ideal for intervention analysis. The thesis concludes that structural models provide a more general framework for rational (...) decision makers. (shrink)

The following four assumptions plausibly describe the ideal rational agent. (1) She knows what her beliefs are. (2) She desires to believe only truths. (3) Whenever she desires that P → Q and knows that P, she desires that Q. (4) She does not both desire that P and desire that ~P, for any P. Although the assumptions are plausible, they have an implausible consequence. They imply that the ideal rational agent does not believe and desire contradictory propositions. She neither (...) desires the world to be any different than she thinks it is, nor thinks it is any different than she desires it to be. The problem of preserving our intuitions about desire, without embracing the implausible conclusion, is what I call “the Wishful Thinking Puzzle.” In this paper, I examine how this puzzle arises, and I argue that it is surprisingly difficult to solve. Even the decision theoretic conception of desire is not immune to the puzzle. One approach, the contrastive conception of desire, does avoid the puzzle without being ad hoc, but it remains too inchoate to win our full confidence. (shrink)

In this paper we discuss how Causal Decision Theory should be modified to handle a class of problematic cases involving deterministic laws. Causal Decision Theory, as it stands, is problematically biased against your endorsing deterministic propositions (for example it tells you to deny Newtonian physics, regardless of how confident you are of its truth). Our response is that this is not a problem for Causal Decision Theory per se, but arises because of the standard method for assessing the truth of (...) certain counterfactuals. The truth of deterministic laws is `modally fragile' on the standard semantics for counterfactuals: if determinism is true and you were to do otherwise, the laws would be different. We provide two ways of avoiding this problem: 1) supplement the standard semantics for counterfactuals with impossible worlds, or 2) introduce rigid designators into the description of problematic decision situations. We argue that both of these approaches are well-motivated and can be readily incorporated into Lewisian Causal Decision Theory. (shrink)

Until recently, many philosophers took Causal Decision Theory to be more successful than its rival, Evidential Decision Theory. Things have changed, however, with a renewed concern that cases involving an extreme form of decision instability are counterexamples to CDT :392–403, 1984; Egan in Philos Rev 116:93–114, 2007). Most prominent among those cases of extreme decision instability is the Psychopath Button, due to Andy Egan; in that case, CDT recommends a seemingly absurd act that almost certainly results in your death. This (...) renewed attention to decision instability has spurned an array of modifications to and rejections of CDT. I argue, however, that the Psychopath Button and its ilk are no counterexamples to CDT. That is, given the causalist’s commitments in Newcomb Problems, they already have the tools to justify CDT’s verdict in Egan-style cases of extreme decision instability. I first argue that there is no reason to think the Psychopath Button is a counterexample to CDT; in particular, many philosophers have placed too much weight on pre-theoretic intuition in Egan-cases, and apart from pre-theoretic intuition, arguments against CDT in cases of extreme decision instability are flawed. My second claim is that the causalist can provide good reasons for following CDT in cases of extreme decision instability. I present a new case, the Two Button Defense, that highlights precisely why the causalist can reasonably follow CDT in even the Psychopath Button. CDT therefore stands as a viable decision theory, without need for modification, restriction, or rejection. (shrink)

_ doi:10.1093/analys/anx072 _, published: 27 June 2017.

Anti-exceptionalism about logic takes logical theories to be continuous with scientific theories. Scientific theories are subject to criteria of theoretical equivalence. This article compares two types of theoretical equivalence – one syntactic and one semantic – in the context of logical anti-exceptionalism, and argues that the syntactic approach leads to undesirable consequences. The anti-exceptionalist should therefore take a semantic approach when evaluating whether logical theories, understood as scientific theories, are equivalent. This article argues for a particular semantic approach, in terms (...) of categorical equivalence, to determine whether logical theories are equivalent. (shrink)

The 'Why ain'cha rich?' argument for one-boxing in Newcomb's problem allegedly vindicates evidential decision theory and undermines causal decision theory. But there is a good response to the argument on behalf of causal decision theory. I develop this response. Then I pose a new problem and use it to give a new 'Why ain'cha rich?' argument. Unlike the old argument, the new argument targets evidential decision theory. And unlike the old argument, the new argument is sound.

Suppose that you have to take a test tomorrow but you do not want to study. Unfortunately you should study, since you care about passing and you expect to pass only if you study. Is there anything you can do to make it the case that you should not study? Is there any way for you to ‘rationalize’ slacking off? I suggest that such rationalization is impossible. Then I show that if evidential decision theory is true, rationalization is not only (...) possible but sometimes advisable. (shrink)

There are arguments which purport to rebut psychological determinism by appealing to its alleged incompatibility with rationality. I argue that they all fail. Against Davidson, I argue that rationality does not preclude the existence of psychological laws. Against Popper, I argue that rationality is compatible with the possibility of predicting human actions. Against Schlesinger, I claim that Newcomb's problem cannot be invoked to show that human actions are unpredictable. Having vindicated the possibility of a rationally-based theory of action, I consider (...) the form it might take. (shrink)

Fallibilism is an attractive epistemological position, avoiding the Scylla of rationalism, and the Charybdis of scepticism. Acknowledging, on the one hand, human imperfection, yet claiming that science and rational inquiry are possible. Fallibilism is a thesis, but equally importantly – an epistemological recommendation. that we should never be absolutely sure of anything. My aim in this paper is to drive a wedge between the thesis and the recommendation. The (eminently plausible) doctrine, I shall argue, cannot be used to ground the (...) epistemological prescription for caution. (shrink)

In some decision problems adoption of an option furnishes evidence about the option's consequences. Rational decisions take account of that evidence, although it makes an option's adoption changes the option's expected utility.

Suppose that it is rational to choose or intend a course of action if and only if the course of action maximizes some sort of expectation of some sort of value. What sort of value should this definition appeal to? According to an influential neo-Humean view, the answer is “Utility”, where utility is defined as a measure of subjective preference. According to a rival neo-Aristotelian view, the answer is “Choiceworthiness”, where choiceworthiness is an irreducibly normative notion of a course of (...) action that is good in a certain way. The neo-Humean view requires preferences to be measurable by means of a utility function. Various interpretations of what exactly a “preference” is are explored, to see if there is any interpretation that supports the claim that a rational agent’s “preferences” must satisfy the “axioms” that are necessary for them to be measurable in this way. It is argued that the only interpretation that supports the idea that the rational agent’s preferences must meet these axioms interprets “preferences” as a kind of value-judgment. But this turns out to be version of the neo-Aristotelian view, rather than the neo-Humean view. Rational intentions maximize expected choiceworthiness, not expected utility. (shrink)

Sometimes, we think of belief as a phenomenon that comes in degrees – that is, in the many different levels of confidence that a thinker might have in various different propositions. Sometimes, we think of belief as a simple two-place relation that holds between a thinker and a proposition – that is, as what I shall here call "outright belief".

This article proposes a new theory of rational decision, distinct from both causal decision theory (CDT) and evidential decision theory (EDT). First, some intuitive counterexamples to CDT and EDT are presented. Then the motivation for the new theory is given: the correct theory of rational decision will resemble CDT in that it will not be sensitive to any comparisons of absolute levels of value across different states of nature, but only to comparisons of the differences in value between the available (...) options within states of nature; however, the correct theory will also resemble EDT in that it will rely on conditional probabilities (not unconditional probabilities). The new theory gives a prominent role to the notion of a “benchmark” for each state of nature, by comparison with which the value of the available options in that state of nature are measured, and so it has been called the Benchmark Theory (BT). It is argued that BT gives the right verdict on the cases that seem to be counterexamples to CDT and EDT. Finally, some objections to BT are considered and answered. (shrink)

It is commonly thought, in line with the position defended in an influential paper by David Lewis, that the decision problems faced in the prisoner’s dilemma and the Newcomb situation are essentially the same problem. José Luis Bermúdez has recently attacked the case Lewis makes for this claim. While I think the claim is false, I contend that Bermúdez’s reason for rejecting Lewis’s argument is inadequate, and then outline what I take to be a better reason for doing so.

I explore the debate about causal versus evidential decision theory, and its recent developments in the work of Andy Egan, through the method of some simple games based on agents' predictions of each other's actions. My main focus is on the requirement for rational agents to act in a way which is consistent over time and its implications for such games and their more realistic cousins.

In this paper I explore a version of standard (expected utility) decision theory in which the probability parameter is interpreted as an objective chance believed by agents to obtain and values of this parameter are fixed by indicative conditionals linking possible actions with possible outcomes. After reviewing some recent developments centering on the common-cause counterexamples to the standard approach, I introduce and briefly discuss the key notions in my own approach. (This approach has essentially the same results as the causal (...) approach in common-cause cases.) I then discuss the Rule of Dominance and find, in the context of the present proposal, that it cannot serve as an independent source of action justification. Turning next to Newcomb''s Problem, I argue that the much discussed issue of back-tracking counterfactuals is something of a red herring for decision theory. Once the twin distractions of back-tracking counterfactuals and Dominance Reasoning are set aside the 1-box solution emerges as a natural consequence of the present proposal. It is of interest that this proposal agrees with the causal approach in the standard common-cause examples and the expected-utility approach in the Newcomb case: one can be smart and rich and keep on smoking. (shrink)

It is often thought that the main significant difference between evidential decision theory and causal decision theory is that they recommend different acts in Newcomb-style examples (broadly construed) where acts and states are correlated in peculiar ways. However, this paper presents a class of non-Newcombian examples that evidential decision theory cannot adequately model whereas causal decision theory can. Briefly, the examples involve situations where it is clearly best to perform an act that will not influence the desired outcome. On evidential (...) decision theory—but not causal decision theory—this situation turns out to be impossible: acts that an agent does not think influence the desired outcome are never optimal. Typically, sophisticated versions of evidential decision theory emulate causal decision theoretic reasoning by (implicitly) conditioning on causal confounders, but in the kind of example considered here, this trick does not work. The upshot is that there is more to causal reasoning than has so far been appreciated. (shrink)

I argue that information is a goal-relative concept for Bayesians. More precisely, I argue that how much information is provided by a piece of evidence depends on whether the goal is to learn the truth or to rank actions by their expected utility, and that different confirmation measures should therefore be used in different contexts. I then show how information measures may reasonably be derived from confirmation measures, and I show how to derive goal-relative non-informative and informative priors given background (...) information. Finally, I argue that my arguments have important implications for both objective and subjective Bayesianism. In particular, the Uniqueness Thesis is either false or must be modified. Moreover, objective Bayesians must concede that pragmatic factors systematically influence which priors are rational, and subjective Bayesians must concede that pragmatic factors sometimes partly determine which prior distribution most accurately represents an agent's epistemic state. (shrink)

I argue that information is a goal-relative concept for Bayesians. More precisely, I argue that how much information is provided by a piece of evidence depends on whether the goal is to learn the truth or to rank actions by their expected utility, and that different confirmation measures should therefore be used in different contexts. I then show how information measures may reasonably be derived from confirmation measures, and I show how to derive goal-relative non-informative and informative priors given background (...) information. Finally, I argue that my arguments have important implications for both objective and subjective Bayesianism. In particular, the Uniqueness Thesis is either false or must be modified. Moreover, objective Bayesians must concede that pragmatic factors systematically influence which priors are rational, and subjective Bayesians must concede that pragmatic factors sometimes partly determine which prior distribution most accurately represents an agent’s epistemic state. (shrink)

There are two general classes of social conventions: conventions of coordination, and conventions of partial conflict. In coordination problems, the interests of the agents coincide, while in partial conflict problems, some agents stand to gain only if other agents unilaterally make certain sacrifices. Lewis' (1969) pathbreaking analysis of convention in terms of game theory focuses on coordination problems, and cannot accommodate partial conflict problems. In this paper, I propose a new game-theoretic definition of convention which generalizes previous game-theoretic definitions (Lewis (...) 1969, Vanderschraaf 1995), and which can be used to characterize norms of justice in partial conflict situations. I argue that the key structural property necessary for a social arrangement to be a convention is that it be conditionally self-enforcing, in the sense that: (i) each agent has a decisive reason to follow her end of the arrangement given that she expects all to do likewise, (ii) given a different set of expectations, some agents would have had a decisive reason to deviate, and (iii) these facts are common knowledge. This leads to a definition of convention as a strict correlated equilibrium (Aumann 1974) together with appropriate common knowledge conditions. Examples are given in which it is shown how this more general account of convention can be used to analyze norms of justice as well as coordination problems. It is only a general sense of common interest; which sense all the members of the society express to one another, and which induces them to regulate their conduct by certain rules. I observe, that it will be for my interest to leave another in the possession of his goods, provided he will act in the same manner with regard to me. He is sensible of a like interest in the regulation of his conduct. When this common sense of interest is mutually express'd, and is known to both, it produces a suitable resolution and behavior. And this may properly be call'd a convention or agreement betwixt us,.. (shrink)

Causal eliminativists maintain that all causal talk is false. The prospects for such a view seem to be stymied by an indispensability argument, charging that any agent must distinguish between effective and ineffective strategies, and that such a distinction must commit that agent to causal notions. However, this argument has been under-explored. The contributions of this paper are twofold: first, I provide a thorough explication of the indispensability argument and the various ways it might be defended. Second, I point to (...) an important limitation in the argument and suggest that it does not give us sufficient reason to reject eliminativism. In support of this last claim, I show that the distinction between effective and ineffective strategies could perfectly well be grounded in a counterfactual rather than a causal decision theory and argue that there are fully adequate explanations of how we could come to make the requisite counterfactual judgments that need not invoke causal concepts. (shrink)

I describe a thought experiment in which an agent must choose between suffering a greater pain in the past or a lesser pain in the future. This case demonstrates that the ‘temporal value asymmetry’ – our disposition to attribute greater significance to future pleasures and pains than to past – can have consequences for the rationality of actions as well as attitudes. This fact, I argue, blocks attempts to vindicate the temporal value asymmetry as a useful heuristic tied to the (...) asymmetry of causation. Since the two standard arguments for the rationality of the temporal value asymmetry appeal to causal asymmetry and the passage of time respectively, the failure of the causal asymmetry explanation suggests that the B-theory, which rejects temporal passage, has substantial revisionary implications concerning our attitudes toward past and future experience. (shrink)

This paper focuses on the view that rationality requires that our credences be regular. I go through different formulations of the requirement, and show that they face several problems. I then formulate a version of the requirement that solves most of, if not all, these problems. I conclude by showing that an argument thought to support the requirement as traditionally formulated actually does not; if anything, the argument, slightly modified, supports my version of the requirement.Send article to KindleTo send this (...) article to your Kindle, first ensure [email protected] is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle. Find out more about sending to your Kindle. Note you can select to send to either the @free.kindle.com or @kindle.com variations. ‘@free.kindle.com’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply. Find out more about the Kindle Personal Document Service.REGULARITY REFORMULATEDVolume 9, Issue 4Weng Hong TangDOI: https://doi.org/10.1017/epi.2012.23Your Kindle email address Please provide your Kindle [email protected]@kindle.com Available formats PDF Please select a format to send. By using this service, you agree that you will only keep articles for personal use, and will not openly distribute them via Dropbox, Google Drive or other file sharing services. Please confirm that you accept the terms of use. Cancel Send ×Send article to Dropbox To send this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about sending content to Dropbox. REGULARITY REFORMULATEDVolume 9, Issue 4Weng Hong TangDOI: https://doi.org/10.1017/epi.2012.23Available formats PDF Please select a format to send. By using this service, you agree that you will only keep articles for personal use, and will not openly distribute them via Dropbox, Google Drive or other file sharing services. Please confirm that you accept the terms of use. Cancel Send ×Send article to Google Drive To send this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about sending content to Google Drive. REGULARITY REFORMULATEDVolume 9, Issue 4Weng Hong TangDOI: https://doi.org/10.1017/epi.2012.23Available formats PDF Please select a format to send. By using this service, you agree that you will only keep articles for personal use, and will not openly distribute them via Dropbox, Google Drive or other file sharing services. Please confirm that you accept the terms of use. Cancel Send ×Export citation Request permission. (shrink)

Examples involving common causes — most prominently, examples involving genetically influenced choices — are analytically equivalent not to standard Newcomb Problems — in which the Predictor genuinely predicts the agent's decision — but to non-standard Newcomb Problems — in which the Predictor guarantees the truth of her predictions by interfering with the agent's decision to make the agent choose as it was predicted she would. When properly qualified, causal and epistemic decision theories diverge only on standard — not on non-standard (...) — Newcomb Problems, and thus not on examples involving common causes. (shrink)

Jim Joyce has argued that David Lewis’s formulation of causal decision theory is inadequate because it fails to apply to the “small world” decisions that people face in real life. Meanwhile, several authors have argued that causal decision theory should be developed such that it integrates the interventionist approach to causal modeling because of the expressive power afforded by the language of causal models, but, as of now, there has been little work towards this end. In this paper, I propose (...) a variant of Lewis’s causal decision theory that is intended to meet both of these demands. Specifically, I argue that Lewis’s causal decision theory can be rendered applicable to small world decisions if one analyzes his dependency hypotheses as causal hypotheses that depend on the interventionist causal modeling framework for their semantics. I then argue that this interventionist variant of Lewis’s causal decision theory is preferable to interventionist causal decision theories that purportedly generalize Lewis’s through the use of conditional probabilities. This is because Lewisian interventionist decision theory captures the causal decision theorist’s conviction that any correlation between what the agent does and cannot cause should be irrelevant to the agent’s choice, while purported generalizations do not. (shrink)

Does the desirability of a proposition depend on whether it is true? Not according to the Invariance assumption, held by several notable philosophers. The Invariance assumption plays an important role in David Lewis’ famous arguments against the so-called Desire-as-Belief thesis (DAB), an anti-Humean thesis according to which a rational agent desires a proposition exactly to the degree that she believes the proposition to be desirable. But the assumption is of interest independently of Lewis’ arguments, for instance since both Richard Jeffrey (...) and James Joyce make the assumption (or, strictly speaking, accept a thesis that implies Invariance) in their influential books on decision theory. The main claim to be defended in this paper is that Invariance is incompatible with certain assumptions of decision theory. I show that the assumption fails on the most common interpretations of desirability and/or choice-worthiness found in decision theory. I moreover show that Invariance is inconsistent with Richard Jeffrey’s decision theory, on which Lewis’ arguments against DAB are based. Finally, I show that Invariance contradicts how we in general do and should think about conditional desirability. (shrink)

In this paper, I use interventionist causal models to identify some novel Newcomb problems, and subsequently use these problems to refine existing interventionist treatments of causal decision theory. The new Newcomb problems that make trouble for existing interventionist treatments involve so-called ‘exotic choice’—that is, decision-making contexts where the agent has evidence about the outcome of her choice. I argue that when choice is exotic, the interventionist can adequately capture causal decision-theoretic reasoning by introducing a new interventionist approach to updating on (...) exotic evidence. But I also argue that this new updating procedure is principled only if the interventionist trades in the typical interventionist conception of choice for an alternative Ramseyan conception. I end by arguing that the guide to exotic choice developed here may, despite its name, be useful in some everyday contexts. (shrink)

The paper will show how one may rationalize one-boxing in Newcomb's problem and drinking the toxin in the Toxin puzzle within the confines of causal decision theory by ascending to so-called reflexive decision models which reflect how actions are caused by decision situations (beliefs, desires, and intentions) represented by ordinary unreflexive decision models.

The crucial premise of the standard argument for two-boxing in Newcomb's problem, a causal dominance principle, is false. We present some counterexamples. We then offer a metaethical explanation for why the counterexamples arise. Our explanation reveals a new and superior argument for two-boxing, one that eschews the causal dominance principle in favor of a principle linking rational choice to guidance and actual value maximization.

Consequentialists often assume rational monism: the thesis that options are always made rationally permissible by the maximization of the selfsame quantity. This essay argues that consequentialists should reject rational monism and instead accept rational pluralism: the thesis that, on different occasions, options are made rationally permissible by the maximization of different quantities. The essay then develops a systematic form of rational pluralism which, unlike its rivals, is capable of handling both the Newcomb problems that challenge evidential decision theory and the (...) unstable problems that challenge causal decision theory. (shrink)

The world is said to contain crystal balls whenever the present carries news of the as-yet-undetermined parts of the future. Many philosophers believe that crystal balls are metaphysically possible. In this essay, I argue that they are not. Whether crystal balls are possible matters, for at least two reasons. The first is epistemological. According to a simple, user-friendly chance norm for credence, which I call the Present Principle, agents are rationally required to conform their credences to their expectations of the (...) present chances, deferring to the present chances as they would to an expert. I would like to defend the Present Principle since its truth would do much to simplify the relation between chance and credence. But the Present Principle is counterexample-free, and hence defensible, only if crystal balls are impossible. The second reason is decision-theoretic. The problem of crystal balls is one of the main objections to causal decision theory. But crystal ball cases can be counterexample to causal decision theory only if crystal balls are possible, and, as I argue, crystal balls are not possible. (shrink)

Seeking a decision theory that can handle both the Newcomb problems that challenge evidential decision theory and the unstable problems that challenge causal decision theory, some philosophers recently have turned to ‘graded ratifiability’. However, the graded ratifiability approach to decision theory is, despite its virtues, unsatisfactory; for it conflicts with the platitude that it is always rationally permissible for an agent to knowingly choose their best option.