Michael Scriven’s (1959) example of identical twins (who are said to be equal in fitness but unequal in their reproductive success) has been used by many philosophers of biology to discuss how fitness should be defined, how selection should be distinguished from drift, and how the environment in which a selection process occurs should be conceptualized. Here it is argued that evolutionary theory has no commitment, one way or the other, as to whether the twins are equally fit. This is (...) because the theory of natural selection is fundamentally about the fitnesses of traits, not the fitnesses of token individuals. A plausible philosophical thesis about supervenience entails that the twins are equally fit if they live in identical environments, but evolutionary biology is not committed to the thesis that the twins live in identical environments. Evolutionary theory is right to focus on traits, rather than on token individuals, because the fitnesses of token organisms (as opposed to their actual survivorship and degree of reproductive success) are almost always unknowable. This point has ramifications for the question of how Darwin’s theory of evolution and R. A. Fisher’s are conceptually different. (shrink)

Is all explanations causal explanation? Puzzles about barometer readings "explain" storms and shadow lengths "explaining" flagpole heights make it attractive to think so. Wesley Salmon (1984) has endorsed this causal thesis. One way to test this thesis is to assess the explanatory import of pseudo-processes. I do so by discussing the concept of heritability, which measures a pseudo-process, and one role it played in the theory of natural selection: explaining response to selection. This will show, not just that heritability has (...) heuristic or predictive utility, but that it can be explanatory. Possible responses to this case are also discussed. (shrink)

This article reviews two standard criticisms of creationism/intelligent design (ID): it is unfalsifiable, and it is refuted by the many imperfect adaptations found in nature. Problems with both criticisms are discussed. A conception of testability is described that avoids the defects in Karl Popper’s falsifiability criterion. Although ID comes in multiple forms, which call for different criticisms, it emerges that ID fails to constitute a serious alternative to evolutionary theory.

The chapter discusses the principle of conservatism and traces how the general principle is related to the specific one. This tracing suggests that the principle of conservatism needs to be refined. Connecting the principle in cognitive science to more general questions about scientific inference also allows us to revisit the question of realism versus instrumentalism. The framework deployed in model selection theory is very general; it is not specific to the subject matter of science. The chapter outlines some non-Bayesian ideas (...) that have been developed in model selection theory. The principle of conservatism, like C. Lloyd Morgan's canon, describes a preference concerning kinds of parameters. It says that a model that postulates only lower-level intentionality is preferable to one that postulates higher-level intentionality if both fit the data equally well. The model selection approach to parsimony helps explain why unification is a theoretical virtue. (shrink)

The logical empiricists said some good things about epistemology and scientific method. However, they associated those epistemological ideas with some rather less good ideas about philosophy of language. There is something epistemologically suspect about statements that cannot be tested. But to say that those statements are meaningless is to go too far. And there is something impossible about trying to figure out which of two empirically equivalent theories is true. But to say that those theories are synonymous is also to (...) go too far. My goal in this paper is not to resuscitate all these positivist ideas, but to revisit just one of them. Instrumentalism is the idea that theories are instruments for making predictions. Of course, no one would disagree that this is one of the things we use theories to do. In just the same way, no one could disagree with the emotivist claim that one of the things we do with ethical terms like "good" and "right" is to express our feelings of approval and disapproval. Instrumentalism and emotivism become contentious, and therefore interesting, when these claims are supplemented. (shrink)

There are two concepts of causes, property causation and token causation. The principle I want to discuss describes an epistemological connection between the two concepts, which I call the Connecting Principle. The rough idea is that if a token event of type Cis followed by a token event of type E, then the support of the hypothesis that the first event token caused the second increases as the strength of the property causal relation of C to E does. I demonstrate (...) the principle, illustrate its application to phylogenies, infections, and rumours, and discuss its consequence for the conceptual distinctness of causal processes from the events they connect. Although I am by no means confident that the Connecting Principle is ultimately correct, it seems to be a useful point of departure into an important aspect of the epistemology of causality. (shrink)

Developing a definition of group selection, and applying that definition to the dispute in the social sciences between methodological holists and methodological individualists, are the two goals of this paper. The definition proposed distinguishes between changes in groups that are due to group selection and changes in groups that are artefacts of selection processes occurring at lower levels of organization. It also explains why the existence of group selection is not implied by the mere fact that fitness values of organisms (...) are sensitive to the composition of groups. And, lastly, the definition explains why group selection need not involve selection for altruism. Group selection is thereby seen as an evolutionary force which is objectively distinct from other evolutionary forces. Applying the distinction between group and individual selection to the holism/individualism dispute has the desirable result that the dispute is not decidable a priori. This way of looking at the dispute yields a conception of individualism which is untainted by atomism and a conception of holism which is unspoiled by hypostatis. (shrink)

In their book What Darwin Got Wrong, Jerry Fodor and Massimo Piattelli-Palmarini construct an a priori philosophical argument and an empirical biological argument. The biological argument aims to show that natural selection is much less important in the evolutionary process than many biologists maintain. The a priori argument begins with the claim that there cannot be selection for one but not the other of two traits that are perfectly correlated in a population; it concludes that there cannot be an evolutionary (...) theory of adaptation. This article focuses mainly on the a priori argument. (shrink)

The task of Philosophy in Science (PinS) is to use philosophical tools to help solve scientific problems. This article describes how I stumbled into this line of work and then addressed several topics in philosophy of biology—units of selection, cladistic parsimony, robustness and trade-offs in model building, adaptationism, and evidence for common ancestry—often in collaboration with scientists. I conclude by offering advice for would-be PinS practitioners.

We argue against some of Reichenbach's claims about causal forks are incorrect. We do not see why the Second Law of Thermodynamics rules out the existence of conjunctive forks open to the past. In addition, we argue that a common effect rarely forms a conjunctive fork with its joint causes, but it sometimes does. Nevertheless, we think there is something to be said for Reichenbach's idea that forks of various kinds are relevant to explaining why we know more about the (...) past than about the future. And entropy turns out to play a role in the explanation, though not the determining role that Reichenbach envisions. (shrink)

Maximum Parsimony (MP) and Maximum Likelihood (ML) are two methods for evaluating which phlogenetic tree is best supported by data on the characteristics of leaf objects (which may be species, populations, or individual organisms). MP has been criticized for assuming that evolution proceeds parsimoniously -- that if a lineage begins in state i and ends in state j, the way it got from i to j is by the smallest number of changes. MP has been criticized for needing to assume (...) some model or other of the evolutionary process even though biologists often do not know what that process really is. This paper critically evaluates both criticisms. (shrink)

Do traits evolve because they are good for the group, or do they evolve because they are good for the individual organisms that have them? The question is whether groups, rather than individual organisms, are ever “units of selection.” My exposition begins with the 1960’s, when the idea that traits evolve because they are good for the group was criticized, not just for being factually mistaken, but for embodying a kind of confused thinking that is fundamentally at odds with the (...) logic that Darwin’s theory requires. A counter-movement has arisen since the 1960’s, called multi-level selection theory, according to which selection acts at multiple levels, including the level of the group. After discussing the 1960’s attack on group selection and the concept’s subsequent revival, I examine Darwin’s views on the subject. I discuss what Darwin says about four examples: human morality, the barbed stinger of the honeybee, neuter workers in species of social insect, and the sterility of many interspecies hybrids. I argue that Darwin defended hypotheses of group selection in the first three problems, but rejected it in the fourth. I also discuss Darwin’s general views about the role of group selection in evolution. (shrink)

Carl Hempel (1965) argued that probabilistic hypotheses are limited in what they can explain. He contended that a hypothesis cannot explain why E is true if the hypothesis says that E has a probability less than 0.5. Wesley Salmon (1971, 1984, 1990, 1998) and Richard Jeffrey (1969) argued to the contrary, contending that P can explain why E is true even when P says that E’s probability is very low. This debate concerned noncontrastive explananda. Here, a view of contrastive causal (...) explanation is described and defended. It provides a new limit on what probabilistic hypotheses can explain; the limitation is that P cannot explain why E is true rather than A if P assign E a probability that is less than or equal to the probability that P assigns to A. The view entails that a true deterministic theory and a true probabilistic theory that apply to the same explanandum partition are such that the deterministic theory explains all the true contrastive propositions constructable from that partition, whereas the probabilistic theory often fails to do so. (shrink)

Los filósofos han tendido a discutir el esencialismo como si fuera una doctrina global, una filosofía que, por alguna razón uniforme, debiera ser adoptada por todas las ciencias o por ninguna. Popper (1972) ha adoptado una postura global negativa, porque ve al esencialismo como un obstáculo fundamental para la racionalidad científica. También Quine (1953b, 1960), por una combinación de motivos semánticos y epistemológicos, quiere desterrar el esencialismo de la totalidad del discurso científico. Sin embargo, en fechas más recientes, Putnam (1975) (...) y Kripke (1972) han propugnado doctrinas esencialistas y han afirmado que es tarea de cada ciencia investigar las propiedades esenciales de sus clases naturales constitutivas. (shrink)

Conway Morris argues against Stephen Jay Gould's argument that the history of life is radically contingent by describing the abundance of convergences, wherein different lineages starting in different states, arrive at the same adaptations. A standard example is the evolution of the camera eye. This review assesses the validity of Conway Morris' argument.

All extant purely probabilistic measures of explanatory power satisfy the following technical condition: if Pr(E | H1) > Pr(E | H2) and Pr(E | ∼H1) < Pr(E | ∼H2), then H1’s explanatory power with respect to E is greater than H2’s explanatory power with respect to E. We argue that any measure satisfying this condition faces three serious problems—the Problem of Temporal Shallowness, the Problem of Negative Causal Interactions, and the Problem of Nonexplanations. We further argue that many such measures (...) face a fourth problem—the Problem of Explanatory Irrelevance. (shrink)

We argue elsewhere that explanatoriness is evidentially irrelevant . Let H be some hypothesis, O some observation, and E the proposition that H would explain O if H and O were true. Then O screens-off E from H: Pr = Pr. This thesis, hereafter “SOT” , is defended by appeal to a representative case. The case concerns smoking and lung cancer. McCain and Poston grant that SOT holds in cases, like our case concerning smoking and lung cancer, that involve frequency (...) data. However, McCain and Poston contend that there is a wider sense of evidential relevance—wider than the sense at play in SOT—on which explanatoriness is evidentially relevant even in cases involving frequency data. This is their main point, but they also contend that SOT does not hold in certain cases not involving frequency data. We reply to each of these points and conclude with some general remarks on screening-off as a test of evidential relevance. (shrink)

We argued that explanatoriness is evidentially irrelevant in the following sense: Let H be a hypothesis, O an observation, and E the proposition that H would explain O if H and O were true. Then our claim is that Pr = Pr. We defended this screening-off thesis by discussing an example concerning smoking and cancer. Climenhaga argues that SOT is mistaken because it delivers the wrong verdict about a slightly different smoking-and-cancer case. He also considers a variant of SOT, called (...) “SOT*”, and contends that it too gives the wrong result. We here reply to Climenhaga’s arguments and suggest that SOT provides a criticism of the widely held theory of inference called “inference to the best explanation”. (shrink)

Sober [2011] argues that some causal statements are a priori true and that a priori causal truths are central to explanations in the theory of natural selection. Lange and Rosenberg [2011] criticize Sober's argument. They concede that there are a priori causal truths, but maintain that those truths are only ‘minimally causal’. They also argue that explanations that are built around a priori causal truths are not causal explanations, properly speaking. Here we criticize both of Lange and Rosenberg's (...) claims. (shrink)

The disjunction problem and the distality problem each presents a challenge that any theory of mental content must address. Here we consider their bearing on purely probabilistic causal theories. In addition to considering these problems separately, we consider a third challenge—that a theory must solve both. We call this “the hard problem.” We consider 8 basic ppc theories along with 240 hybrids of them, and show that some can handle the disjunction problem and some can handle the distality problem, but (...) none can handle the hard problem. This is our main result. We then discuss three possible responses to that result, and argue that though the first two fail, the third has some promise. (shrink)

Decision theory requires agents to assign probabilities to states of the world and utilities to the possible outcomes of different actions. When agents commit to having the probabilities and/or utilities in a decision problem defined by objective features of the world, they may find themselves unable to decide which actions maximize expected utility. Decision theory has long recognized that work-around strategies are available in special cases; this is where dominance reasoning, minimax, and maximin play a role. Here we describe a (...) different work around, wherein a rational decision about one decision problem can be reached by ‘interpolating’ information from another problem that the agent believes has already been rationally solved. (shrink)

Is there some general reason to expect organisms that have beliefs to have false beliefs? And after you observe that an organism occasionally occupies a given neural state that you think encodes a perceptual belief, how do you evaluate hypotheses about the semantic content that that state has, where some of those hypotheses attribute beliefs that are sometimes false while others attribute beliefs that are always true? To address the first of these questions, we discuss evolution by natural selection and (...) show how organisms that are risk-prone in the beliefs they form can be fitter than organisms that are risk-free. To address the second question, we discuss a problem that is widely recognized in statistics – the problem of over-fitting – and one influential device for addressing that problem, the Akaike Information Criterion (AIC). We then use AIC to solve epistemological versions of the disjunction and distality problems, which are two key problems concerning what it is for a belief state to have one semantic content rather than another. (shrink)

We argue in Roche and Sober (2013) that explanatoriness is evidentially irrelevant in that Pr(H | O&EXPL) = Pr(H | O), where H is a hypothesis, O is an observation, and EXPL is the proposition that if H and O were true, then H would explain O. This is a “screening-off” thesis. Here we clarify that thesis, reply to criticisms advanced by Lange (2017), consider alternative formulations of Inference to the Best Explanation, discuss a strengthened screening-off thesis, and consider (...) how it bears on the claim that unification is evidentially relevant. (shrink)

In their 2010 book, Biology’s First Law, D. McShea and R. Brandon present a principle that they call ‘‘ZFEL,’’ the zero force evolutionary law. ZFEL says (roughly) that when there are no evolutionary forces acting on a population, the population’s complexity (i.e., how diverse its member organisms are) will increase. Here we develop criticisms of ZFEL and describe a different law of evolution; it says that diversity and complexity do not change when there are no evolutionary causes.

According to Michael Friedman’s theory of explanation, a law X explains laws Y1, Y2, …, Yn precisely when X unifies the Y’s, where unification is understood in terms of reducing the number of independently acceptable laws. Philip Kitcher criticized Friedman’s theory but did not analyze the concept of independent acceptability. Here we show that Kitcher’s objection can be met by modifying an element in Friedman’s account. In addition, we argue that there are serious objections to the use that Friedman makes (...) of the concept of independent acceptability. (shrink)

In Ockham's Razors: A User's Guide, Elliott Sober argues that parsimony considerations are epistemically relevant on the grounds that certain methods of model selection, such as the Akaike Information Criterion, exhibit good asymptotic behaviour and take the number of adjustable parameters in a model into account. I raise some worries about this form of argument.

on ). Advice about how to move forward on the mindreading debate, particularly when it comes to overcoming the logical problem, is much needed in comparative psychology. In chapter 4 of his book Ockham’s Razors, Elliott Sober takes on the task by suggesting how we might uncover the mechanism that mediates between the environmental stimuli that is visible to all, and chimpanzee social behavior. I argue that Sober's proposed method for deciding between the behaivor-reading and mindreading hypotheses (...) fails given the nature of each of those hypotheses. I argue that the behavior-reading hypothesis that Povinelli and colleagues propose is so rich and robust that it is going to make predictions that are behaviorally indiscernible from the mindreading hypothesis. Further, I argue that the logical problem artificially separates one’s knowledge of behavior and one’s knowledge of mind. If we reject this form of dualism, the logical problem doesn’t arise. (shrink)

A primeira tese de Sober é que não podemos agir livremente, a não ser que o Argumento da Causalidade ou o Argumento da Inevitabilidade tenham alguma falha. O Argumento da Causalidade é o seguinte: nossos estados mentais causam movimentos corporais; mas nossos estados mentais são causados por fatores do mundo físico. Nossa personalidade pode ser reconduzida à nossa experiência e à nossa genética. E tanto a experiência quanto a genética foram causados por itens do mundo físico. Assim, o meio (...) ambiente e os genes são os causadores de nossas crenças e desejos. E estes, por sua vez, causam o nosso comportamento. Como, em última instância, não escolhemos nem os nossos genes e nem o meio ambiente no qual adquirimos as nossas experiências, também não escolhemos o nosso comportamento: ele é causado por fatores além do nosso controle; isso nos faz não ser livres. E o Argumento da Inevitabilidade é exposto por Sober assim: se uma ação foi praticada livremente, então deve ter sido possível ao agente agir de outra forma. Mas, dado que as causas de nossas ações são as nossas crenças e desejos, não poderíamos ter agido diferentemente de como elas nos determinam a agir. (shrink)

Jaegwon Kim’s supervenience/exclusion argument attempts to show that non-reductive physicalism is incompatible with mental causation. This influential argument can be seen as relying on the following principle, which I call “the piggyback principle”: If, with respect to an effect, E, an instance of a supervenient property, A, has no causal powers over and above, or in addition to, those had by its supervenience base, B, then the instance of A does not cause E (unless A is identical with B). In (...) their “Epiphenomenalism: The Dos and the Don’ts,” Larry Shapiro and Elliott Sober employ a novel empirical approach to challenge the piggyback principle. Their empirical approach pulls from the experiments of August Weismann regarding the inheritance of acquired characteristics. Through an examination of Weismann’s experiments, Shapiro and Sober extract lessons in reasoning about the epiphenomenalism of a property. And according to these empirically drawn lessons, the piggyback principle is a don’t. My primary aim in this paper is to defend the piggyback principle against Shapiro and Sober’s empirical approach. (shrink)

Early in the eleventh of his Fifteen Sermons, Joseph Butler advances his best-known argument against psychological hedonism. Elliott Sober calls that argument Butler’s stone, and famously objects to it. I consider whether Butler’s stone has philosophical value. In doing so I examine, and reject, two possible ways of overcoming Sober’s objection, each of which has proponents. In examining the first way I discuss Lord Kames’s version of the stone argument, which has hitherto escaped scholarly attention. Finally, I (...) show that Butler’s stone does something important, which I have not found previously discussed. Butler’s stone blocks an inference, persuasive to many people, which purports to show that we intrinsically desire only pleasure. (shrink)

The epistemic status of Natural Selection has intrigued to biologists and philosophers since the very beginning of the theory to our present times. One prominent contemporary example is Elliott Sober, who claims that Natural Selection, and some other theories in biology, and maybe in economics, are peculiar in including explanatory models/conditionals that are a priori in a sense in which explanatory models/conditionals in Classical Mechanics and most other standard theories are not. In this paper, by analyzing what we (...) take to be the four possible interpretations of Sober’s claim, we argue that, terminological preferences aside, the possible senses in which explanatory models in Natural Selection can qualify, or include elements that can qualify, as a priori, also apply to Classical Mechanics and other standard, highly unified theories. (shrink)

Where E is the proposition that [If H and O were true, H would explain O], William Roche and Elliot Sober have argued that P(H|O&E) = P(H|O). In this paper I argue that not only is this equality not generally true, it is false in the very kinds of cases that Roche and Sober focus on, involving frequency data. In fact, in such cases O raises the probability of H only given that there is an explanatory connection between (...) them. (shrink)

The optimality approach to modeling natural selection has been criticized by many biologists and philosophers of biology. For instance, Lewontin (1979) argues that the optimality approach is a shortcut that will be replaced by models incorporating genetic information, if and when such models become available. In contrast, I think that optimality models have a permanent role in evolutionary study. I base my argument for this claim on what I think it takes to best explain an event. In certain contexts, optimality (...) and game-theoretic models best explain some central types of evolutionary phenomena. ‡Thanks to Michael Friedman, Helen Longino, Michael Weisberg, and especially Elliott Sober for comments on earlier drafts of this paper. †To contact the author, please write to: Department of Philosophy, Stanford University, Stanford, CA 94305-2155; e-mail: [email protected]. (shrink)

An influential argument due to Elliott Sober, subsequently strengthened by Denis Walsh and Joel Pust, moves from plausible premises to the bold conclusion that natural selection cannot explain the traits of individual organisms. If the argument were sound, the explanatory scope of selection would depend, surprisingly, on metaphysical considerations concerning origin essentialism. I show that the Sober-Walsh-Pust argument rests on a flawed counterfactual criterion for explanatory relevance. I further show that a more defensible criterion for explanatory relevance (...) recently proposed by Michael Strevens lends support to the view that natural selection can be relevant to the explanation of individual traits. (shrink)

Roughly, psychological egoism is the thesis that all of a person's intentional actions are ultimately self-interested in some sense; psychological altruism is the thesis that some of a person's intentional actions are not ultimately self-interested, since some are ultimately other-regarding in some sense. C. Daniel Batson and other social psychologists have argued that experiments provide support for a theory called the "empathy-altruism hypothesis" that entails the falsity of psychological egoism. However, several critics claim that there are egoistic explanations of the (...) data that are still not ruled out. One of the most potent criticisms of Batson comes from Elliott Sober and David Sloan Wilson. I argue for two main theses in this paper: (1) we can improve on Sober and Wilson’s conception of psychological egoism and altruism, and (2) this improvement shows that one of the strongest of Sober and Wilson's purportedly egoistic explanations is not tenable. A defense of these two theses goes some way toward defending Batson‘s claim that the evidence from social psychology provides sufficient reason to reject psychological egoism. (shrink)

The conflation of two fundamentally distinct issues has generated serious confusion in the philosophical and biological literature concerning the units of selection. The question of how a unit of selection of defined, theoretically, is rarely distinguished from the question of how to determine the empirical accuracy of claims--either specific or general--concerning which unit(s) is undergoing selection processes. In this paper, I begin by refining a definition of the unit of selection, first presented in the philosophical literature by William Wimsatt, which (...) is grounded in the structure of natural selection models. I then explore the implications of this structural definition for empirical evaluation of claims about units of selection. I consider criticisms of this view presented by Elliott Sober--criticisms taken by some (for example, Mayo and Gilinsky 1987) to provide definitive damage to the structuralist account. I shall show that Sober has misinterpreted the structuralist views; he knocks down a straw man in order to motivate his own causal account. Furthermore, I shall argue, Sober's causal account is dependent on the structuralist account that he rejects. I conclude by indicating how the refined structural definition can clarify which sorts of empirical evidence could be brought to bear on a controversial case involving units of selection. (shrink)

This paper defends the position that the supposed gap between biological altruism and psychological altruism is not nearly as wide as some scholars (e.g., Elliott Sober) insist. Crucial to this defense is the use of James Mark Baldwin's concepts of “organic selection”and “social heredity” to assist in revealing that the gap between biological and psychological altruism is more of a small lacuna. Specifically, this paper argues that ontogenetic behavioral adjustments, which are crucial to individual survival and reproduction, are (...) also crucial to species survival. In particular, it is argued that human psychological altruism is produced and maintained by various sorts of mimicry and self-reflection in the aid of both individual and species survival. The upshot of this analysis is that it is possible to offer an account of psychological altruism that is closelytethered to biological altruism without reducing entirely the former to thelatter. (shrink)

Evidential holism begins with something like the claim that “it is only jointly as a theory that scientific statements imply their observable consequences.” This is the holistic claim that Elliott Sober tells us is an “unexceptional observation”. But variations on this “unexceptional” claim feature as a premise in a series of controversial arguments for radical conclusions, such as that there is no analytic or synthetic distinction that the meaning of a sentence cannot be understood without understanding the whole (...) language of which it is a part and that all knowledge is empirical knowledge. This paper is a survey of what evidential holism is, how plausible it is, and what consequences it has. Section 1 will distinguish a range of different holistic claims, Sections 2 and 3 explore how well motivated they are and how they relate to one another, and Section 4 returns to the arguments listed above and uses the distinctions from the previous sections to identify holism's role in each case. (shrink)

Few scientists are conscious of the distinc- tion between the logic of what they write and the rhetoric of how they write it. This is because we are taught to write scientific papers and books from a third-person per- spective, using as impersonal (and, almost inevitably, boring [1]) a style as possible. The first chapter in Elliott Sober’s new book examines the difference between Darwin’s logic and his rhetoric in The Origin, and manages to teach some interesting and (...) in- sightful historical and philosophical lessons while doing so. (shrink)

A brief introduction to the life and work of William Paley, including a discussion of the structure of his famous design argument.

Causalists about explanation claim that to explain an event is to provide information about the causal history of that event. Some causalists also endorse a proportionality claim, namely that one explanation is better than another insofar as it provides a greater amount of causal information. In this chapter I consider various challenges to these causalist claims. There is a common and influential formulation of the causalist requirement – the ‘Causal Process Requirement’ – that does appear vulnerable to these anti-causalist challenges, (...) but I argue that they do not give us reason to reject causalism entirely. Instead, these challenges lead us to articulate the causalist requirement in an alternative way. This alternative articulation incorporates some of the important anti-causalist insights without abandoning the explanatory necessity of causal information. For example, proponents of the ‘equilibrium challenge’ argue that the best available explanations of the behaviour of certain dynamical systems do not appear to provide any causal information. I respond that, contrary to appearances, these equilibrium explanations are fundamentally causal, and I provide a formulation of the causalist thesis that is immune to the equilibrium challenge. I then show how this formulation is also immune to the ‘epistemic challenge’ – thus vindicating (a properly formulated version of) the causalist thesis. (shrink)

ABSTRACT According to comparativism, comparative confidence is more fundamental than absolute confidence. In two recent AJP papers, Stefánsson has argued that comparativism is capable of explaining interpersonal confidence comparisons. In this paper, I will argue that Stefansson’s proposed explanation is inadequate; that we have good reasons to think that comparativism cannot handle interpersonal comparisons; and that the best explanation of interpersonal comparisons requires thinking about confidence in a fundamentally different way than that which comparativists propose: specifically, we should think of (...) confidence as a dimensionless quantity. (shrink)

Some worry that advanced artificial agents may resist being shut down. The Incomplete Preferences Proposal (IPP) is an idea for ensuring that doesn't happen. A key part of the IPP is using a novel 'Discounted REward for Same-Length Trajectories (DREST)' reward function to train agents to (1) pursue goals effectively conditional on each trajectory-length (be 'USEFUL'), and (2) choose stochastically between different trajectory-lengths (be 'NEUTRAL' about trajectory-lengths). In this paper, we propose evaluation metrics for USEFULNESS and NEUTRALITY. We use a (...) DREST reward function to train simple agents to navigate gridworlds, and we find that these agents learn to be USEFUL and NEUTRAL. Our results thus suggest that DREST reward functions could also train advanced agents to be USEFUL and NEUTRAL, and thereby make these advanced agents useful and shutdownable. (shrink)

Many philosophers are attracted to a complaints-based theory of the procreation asymmetry, according to which creating a person with a bad life is wrong (all else equal) because that person can complain about your act, whereas declining to create a person who would have a good life is not wrong (all else equal) because that person never exists and so cannot complain about your act. In this paper, I present two problems for such theories: the problem of impairable-life acceptance and (...) an especially acute version of the problem of improvable-life avoidance. I explain how these problems afflict two recent complaints-based theories of the procreation asymmetry, from Joe Horton and Abelard Podgorski. (shrink)

I explain and motivate the shutdown problem: the problem of creating artificial agents that (1) shut down when a shutdown button is pressed, (2) don’t try to prevent or cause the pressing of the shutdown button, and (3) otherwise pursue goals competently. I then propose a solution: train agents to have incomplete preferences. Specifically, I propose that we train agents to lack a preference between every pair of different-length trajectories. I suggest a way to train such agents using reinforcement learning: (...) we give the agent lower reward for repeatedly choosing same-length trajectories. (shrink)