When people want to identify the causes of an event, assign credit or blame, or learn from their mistakes, they often reflect on how things could have gone differently. In this kind of reasoning, one considers a counterfactual world in which some events are different from their real-world counterparts and considers what else would have changed. Researchers have recently proposed several probabilistic models that aim to capture how people do (or should) reason about counterfactuals. We present a new model and (...) show that it accounts better for human inferences than several alternative models. Our model builds on the work of Pearl (2000), and extends his approach in a way that accommodates backtracking inferences and that acknowledges the difference between counterfactual interventions and counterfactual observations. We present six new experiments and analyze data from four experiments carried out by Rips (2010), and the results suggest that the new model provides an accurate account of both mean human judgments and the judgments of individuals. (shrink)

Recent work on the interpretation of counterfactual conditionals has paid much attention to the role of causal independencies. One influential idea from the theory of Causal Bayesian Networks is that counterfactual assumptions are made by intervention on variables, leaving all of their causal non-descendants unaffected. But intervention is not applicable across the board. For instance, backtracking counterfactuals, which involve reasoning from effects to causes, cannot proceed by intervention in the strict sense, for otherwise they would be equivalent to their consequents. (...) We discuss these and similar cases, focusing on two factors which play a role in determining whether and which causal parents of the manipulated variable are affected: Speakers' need for an explanation of the hypothesized state of affairs, and differences in the ‘resilience’ of beliefs that are independent of degrees of certainty. We describe the relevant theoretical notions in some detail and provide experimental evidence that these factors do indeed affect speakers' interpretation of counterfactuals. (shrink)

Newsome ((2003). The debate between current versions of covariation and mechanism approaches to causal inference. Philosophical Psychology, 16, 87-107.) recently published a critical review of psychological theories of human causal inference. In that review, he characterized covariation and mechanism theories, the two dominant theory types, as competing, and offered possible ways to integrate them. I argue that Newsome has misunderstood the theoretical landscape, and that covariation and mechanism theories do not directly conflict. Rather, they rely on distinct sets of reliable (...) indicators of causation, and focus on different types of causation (type vs. token). There are certainly debates in the research field, but the theoretical landscape is not as fractured as Newsome suggests, and a potential unifying framework has already emerged using causal Bayes nets. Philosophical work on causal epistemology matters for psychologists, but not in the way Newsome suggests. (shrink)

Life-science phenomena are often explained by specifying the mechanisms that bring them about. The new mechanistic philosophers have done much to substantiate this claim and to provide us with a better understanding of what mechanisms are and how they explain. Although there is disagreement among current mechanists on various issues, they share a common core position and a seeming commitment to some form of scientific realism. But is such a commitment necessary? Is it the best way to go about mechanistic (...) explanation? In this article, we propose an alternative antirealist account that also fits explanatory practice in the life sciences. We pay special attention to mechanistic models, i.e. scientific models that involve a mechanism, and to the role of coherence considerations in building such models. To illustrate our points, we consider the mechanism for the action potential. 1 Introduction2 Some Core Features of Mechanistic Explanation3 Scientific Realism and Mechanistic Explanation4 Antirealist Mechanistic Explanation: The Case of the Action Potential5 Some Outstanding Issues for the Antirealist Mechanist6 Two Problems for the Realist Mechanist7 Conclusions. (shrink)

When evaluating theories of causation, intuitions should not play a decisive role, not even intuitions in flawlessly-designed thought experiments. Indeed, no coherent theory of causation can respect the typical person’s intuitions in redundancy (pre-emption) thought experiments, without disrespecting their intuitions in threat-and-saviour (switching/short-circuit) thought experiments. I provide a deductively sound argument for these claims. Amazingly, this argument assumes absolutely nothing about the nature of causation. I also provide a second argument, whose conclusion is even stronger: the typical person’s causal intuitions (...) are thoroughly unreliable. This argument proceeds by raising the neglected question: in what respects is information about intermediate and enabling variables relevant to reliable causal judgment? (shrink)

Reductionists say things like: all mental properties are physical properties; all normative properties are natural properties. I argue that the only way to resist reductionism is to deny that causation is difference making (thus making the epistemology of causation a mystery) or to deny that properties are individuated by their causal powers (thus making properties a mystery). That is to say, unless one is happy to deny supervenience, or to trivialize the debate over reductionism. To show this, I argue that (...) if properties are individuated by their causal powers then, surprisingly, properties are individuated by necessary co-exemplification. (shrink)

Explainable artificial intelligence (XAI) aims to help people understand black box algorithms, particularly of their outputs. But what are these explanations and when is one explanation better than another? The manipulationist definition of explanation from the philosophy of science offers good answers to these questions, holding that an explanation consists of a generalization that shows what happens in counterfactual cases. Furthermore, when it comes to explanatory depth this account holds that a generalization that has more abstract variables, is broader in (...) scope and/or more accurate is better. By applying these definitions and contrasting them with alternative definitions in the XAI literature I hope to help clarify what a good explanation is for AI. (shrink)

When is a cause of a cause of an effect also a cause of that effect? The right answer is either Sometimes or Always . In favour of Always , transitivity is considered by some to be necessary for distinguishing causes from redundant non-causal events. Moreover transitivity may be motivated by an interest in an unselective notion of causation, untroubled by principles of invidious discrimination. And causal relations appear to add up like transitive relations, so that the obtaining of the (...) overarching relation is not independent of the obtaining of the intermediaries. On the other hand, in favour of Sometimes , often we seem not to treat events that are very spatiotemporally remote from an effect as its causes, even when connected to the effect in question by a chain of counterfactual or chance-raising dependence. Moreover cases of double prevention provide counterexamples to causal transitivity even over short chains. According to the argument of this paper, causation is non-transitive. Transitizing causation provides no viable account of causal redundancy. An unselective approach to causation may motivate resisting the distance counterexamples to transitivity, but it does not help with double prevention, and even makes it more intractable. The strongest point in favour of transitivity is the adding up of causal relations, and this is the point that extant non-transitizing analyses have not adequately addressed. I propose a necessary condition on causation that explains the adding up phenomenon. In doing so it also provides a unifying explanation of distance and double prevention counterexamples to transitivity. (shrink)

The primary aim of this paper is to analyze the concept of degrees of causal contribution for actual events and examine the way in which it can be formally defined. This should go some way to filling out a gap in the legal and philosophical literature on causation. By adopting the conception of a cause as a necessary element of a sufficient set (the so-called NESS test) we show that the concept of degrees of causation can be given clear and (...) even empirical meaning. We then apply a game theoretical framework to derive a measure of causal contribution. Our favoured measure turns out to be a generalised version of the normalized Penrose-Banzhaf index of voting power. (shrink)

This paper examines the structure of moral responsibility for outcomes. A central feature of the analysis is a condition that we term the ‘avoidance potential’, which gives precision to the idea that moral responsibility implies a reasonable demand that an agent should have acted otherwise. We show how our theory can allocate moral responsibility to individuals in complex collective action problems, an issue that sometimes goes by the name of ‘the problem of many hands’. We also show how it allocates (...) moral responsibility in the classic Frankfurt example. (shrink)

Theories of explanation need to account for a puzzling feature of our explanatory practices: the fact that we prefer explanations that are relatively abstract but only moderately so. Contra Franklin-Hall ([2016]), I argue that the interventionist account of explanation provides a natural and elegant explanation of this fact. By striking the right balance between specificity and generality, moderately abstract explanations optimally subserve what interventionists regard as the goal of explanation, namely identifying possible interventions that would have changed the explanandum.

This paper argues that so-called “trumping preemption” is in fact overdetermination or early preemption, and is thus not a distinctive form of redundant causation. I draw a novel lesson from cases thought to be trumping: that the boundary between preemption and overdetermination should be reconsidered.

Philosophical works on actual causation make wide use of thought experiments. The principal aim of this paper is to show how thought experiments are used in the contemporary debate over actual causation and to discuss their role in relation to formal approaches in terms of causal models. I claim that a recourse to thought experiments is not something old fashioned or superseded by abstract models, but it is useful to interpret abstract models themselves and to use our intuitions to judge (...) the results of the model. Recent research on actual causation has stressed the importance of integrating formal models with some notion of normality; I suggest that thought experiments can be useful in eliciting intuitions where normality is not intended in a statistical sense. The first expository part gives a short presentation of the notion of actual causation, summarising some typical problems of counterfactual approaches and how they are treated in causal and structural models. The second part focuses on the problems of model isomorphism and criticises some radical ideas opposing the role of thought experiments, claiming that they may also be of use in evaluating formal models. (shrink)

Philosophical works on actual causation make wide use of thought experiments. The principal aim of this paper is to show how thought experiments are used in the contemporary debate over actual causation and to discuss their role in relation to formal approaches in terms of causal models. I claim that a recourse to thought experiments is not something old fashioned or superseded by abstract models, but it is useful to interpret abstract models themselves and to use our intuitions to judge (...) the results of the model. Recent research on actual causation has stressed the importance of integrating formal models with some notion of normality; I suggest that thought experiments can be useful in eliciting intuitions where normality is not intended in a statistical sense. The first expository part gives a short presentation of the notion of actual causation, summarising some typical problems of counterfactual approaches and how they are treated in causal and structural models. The second part focuses on the problems of model isomorphism and criticises some radical ideas opposing the role of thought experiments, claiming that they may also be of use in evaluating formal models. (shrink)

Philosophical works on actual causation make wide use of thought experiments. The principal aim of this paper is to show how thought experiments are used in the contemporary debate over actual causation and to discuss their role in relation to formal approaches in terms of causal models. I claim that a recourse to thought experiments is not something old fashioned or superseded by abstract models, but it is useful to interpret abstract models themselves and to use our intuitions to judge (...) the results of the model. Recent research on actual causation has stressed the importance of integrating formal models with some notion of normality; I suggest that thought experiments can be useful in eliciting intuitions where normality is not intended in a statistical sense. The first expository part gives a short presentation of the notion of actual causation, summarising some typical problems of counterfactual approaches and how they are treated in causal and structural models. The second part focuses on the problems of model isomorphism and criticises some radical ideas opposing the role of thought experiments, claiming that they may also be of use in evaluating formal models. (shrink)

This article introduces comparative process tracing as a two-step methodological approach that combines theory, chronology, and comparison. For each studied case, the processes leading “from A to B” are reconstructed and analyzed in terms of ideal-type social mechanisms and then compared by making use of the identified mechanisms and ideal-type periodization. Central elements of CPT are path dependence, critical junctures and focal points, social mechanisms, context, periodization, and counterfactual analysis. The CPT approach is described, discussed, and compared with more formal (...) and deterministic forms of process tracing, which are found to be less fruitful for systematic comparison. (shrink)

Machine learning (ML) techniques have become pervasive across a range of different applications, and are now widely used in areas as disparate as recidivism prediction, consumer credit-risk analysis, and insurance pricing. Likewise, in the physical world, ML models are critical components in autonomous agents such as robotic surgeons and self-driving cars. Among the many ethical dimensions that arise in the use of ML technology in such applications, analyzing morally permissible actions is both immediate and profound. For example, there is the (...) potential for learned algorithms to become biased against certain groups. More generally, in so much that the decisions of ML models impact society, both virtually (e.g., denying a loan) and physically (e.g., driving into a pedestrian), notions of accountability, blame and responsibility need to be carefully considered. In this article, we advocate for a two-pronged approach ethical decision-making enabled using rich models of autonomous agency: on the one hand, we need to draw on philosophical notions of such as beliefs, causes, effects and intentions, and look to formalise them, as attempted by the knowledge representation community, but on the other, from a computational perspective, such theories need to also address the problems of tractable reasoning and (probabilistic) knowledge acquisition. As a concrete instance of this tradeoff, we report on a few preliminary results that apply (propositional) tractable probabilistic models to problems in fair ML and automated reasoning of moral principles. Such models are compilation targets for certain types of knowledge representation languages, and can effectively reason in service some computational tasks. They can also be learned from data. Concretely, current evidence suggests that they are attractive structures for jointly addressing three fundamental challenges: reasoning about possible worlds + tractable computation + knowledge acquisition. Thus, these seems like a good starting point for modelling reasoning robots as part of the larger ecosystem where accountability and responsibility is understood more broadly. (shrink)

Pearl opened the door to formally defining actual causation using causal models. His approach rests on two strategies: first, capturing the widespread intuition that X = x causes Y = y iff X = x is a Necessary Element of a Sufficient Set for Y = y, and second, showing that his definition gives intuitive answers on a wide set of problem cases. This inspired dozens of variations of his definition of actual causation, the most prominent of which are due (...) to Halpern & Pearl. Yet all of them ignore Pearl’s first strategy, and the second strategy taken by itself is unable to deliver a consensus. This paper offers a way out by going back to the first strategy: it offers six formal definitions of causal sufficiency and two interpretations of necessity. Combining the two gives twelve new definitions of actual causation. Several interesting results about these definitions and their relation to the various Halpern & Pearl definitions are presented. Afterwards the second strategy is evaluated as well. In order to maximize neutrality, the paper relies mostly on the examples and intuitions of Halpern & Pearl. One definition comes out as being superior to all others, and is therefore suggested as a new definition of actual causation. (shrink)

In recent years, the debate on the problem of causal exclusion has seen an ‘interventionist turn’. Numerous non-reductive physicalists (e.g. Shapiro and Sober 2007) have argued that Woodward's (2003) interventionist theory of causation provides a means to empirically establish the existence of non-reducible mental-to-physical causation. By contrast, Baumgartner (2010) has presented an interventionist exclusion argument showing that interventionism is in fact incompatible with non-reductive physicalism. In response, a number of revised versions of interventionism have been suggested that are compatible with (...) non-reductive physicalism. The first part of this paper reconstructs the definitional details of these modified interventionist theories. The second part investigates whether the modification proposed in Woodward (2011) is not only compatible with, but moreover supports non-reductive physicalism. In particular, it is examined whether that newest variant of interventionism allows for empirically resolving the problem of causal exclusion as envisaged by Shapiro, Sober and others. (shrink)

An actual cause of some token effect is itself a token event that helped to bring about that effect. The notion of an actual cause is different from that of a potential cause – for example a pre-empted backup – which had the capacity to bring about the effect, but which wasn't in fact operative on the occasion in question. Sometimes actual causes are also distinguished from mere background conditions: as when we judge that the struck match was a cause (...) of the fire, while the presence of oxygen was merely part of the relevant background against which the struck match operated. Actual causation is also to be distinguished from type causation: actual causation holds between token events in a particular, concrete scenario; type causation, by contrast, holds between event kinds in scenario kinds. (shrink)

We argue that explainable artificial intelligence (XAI), specifically reason-giving XAI, often constitutes the most suitable way of ensuring that someone can properly be held responsible for decisions that are based on the outputs of artificial intelligent (AI) systems. We first show that, to close moral responsibility gaps (Matthias 2004), often a human in the loop is needed who is directly responsible for particular AI-supported decisions. Second, we appeal to the epistemic condition on moral responsibility to argue that, in order to (...) be responsible for her decision, the human in the loop has to have an explanation available of the system’s recommendation. Reason explanations are especially well-suited to this end and we examine whether – and how – it might be possible to make such explanations fit with AI systems. We support our claims by focusing on a case of disagreement between human in the loop and AI system. (shrink)

The majority of the currently flourishing theories of actual causation are located in a broadly counterfactual framework that draws on structural equations. In order to account for cases of symmetric overdeterminiation and preemption, these theories resort to rather intricate analytical tools, most of all, to what Hitchcock has labeled explicitly nonforetracking counterfactuals. This paper introduces a regularity theoretic approach to actual causation that only employs material conditionals, standard Boolean minimization procedures, and a stability condition that regulates the behavior of causal (...) models under model expansions. Notwithstanding its lightweight analytical toolbox, this regularity theory performs at least as well as the structural equations accounts with their heavy appliances. (shrink)

Team semantics is a highly general framework for logics which describe dependencies and independencies among variables. Typically, the dependencies considered in this context are properties of sets of configurations or data records. We show how team semantics can be further generalized to support languages for the discussion of interventionist counterfactuals and causal dependencies, such as those that arise in manipulationist theories of causation. We show that the “causal teams” we introduce in the present paper can be used for modelling some (...) classical counterfactual scenarios which are not captured by the usual causal models. We then analyse the basic properties of our counterfactual languages and discuss extensively the differences with respect to the Lewisian tradition. (shrink)

The well-known formal semantics of conditionals due to Stalnaker Studies in logical theory, Blackwell, Oxford, 1968), Lewis, and Gärdenfors The logic and 1140 epistemology of scientific change, North-Holland, Amsterdam, 1978, Knowledge in flux, MIT Press, Cambridge, 1988) all fail to distinguish between trivially and nontrivially true indicative conditionals. This problem has been addressed by Rott :345–370, 1986) in terms of a strengthened Ramsey Test. In this paper, we refine Rott’s strengthened Ramsey Test and the corresponding analysis of explanatory relations. We (...) show that our final analysis captures the presumed asymmetry between explanans and explanandum much better than Rott’s original analysis. (shrink)

The present paper aims to complement causal model approaches to causal explanation by Woodward [15], Halpern and Pearl [5], and Strevens [14]. It centres on a strengthened Ramsey Test of conditionals: α ≫ γ iff, after sus- pending judgment about α and γ, an agent can infer γ from the supposition of α. It has been shown by Andreas and Gu ̈nther [1] that such a conditional can be used as starting point of an analysis of ‘because’ in natural language. (...) In what follows, we shall refine this analysis so as to yield a fully fledged account of causal explanation. (shrink)

In this paper, we analyse actual causation in terms of production. The latter concept is made precise by a strengthened Ramsey Test semantics of conditionals: \ iff, after suspending judgement about A and C, C is believed in the course of assuming A. This test allows us to verify or falsify that an event brings about another event. Complementing the concept of production by a weak condition of difference-making gives rise to a full-fledged analysis of causation.

We define a formal semantics of conditionals based on _normatively ideal worlds_. Such worlds are described informally by Armgardt (Gabbay D, Magnani L, Park W, Pietarinen A-V (eds) Natural arguments: a tribute to john woods, College Publications, London, pp 699–708, 2018) to address well-known problems of the counterfactual approach to causation. Drawing on Armgardt’s proposal, we use iterated conditionals in order to analyse causal relations in scenarios of multi-agent interaction. This results in a refined counterfactual approach to causal responsibility in (...) legal contexts, which solves overdetermination problems in an intuitively accessible manner. (shrink)

We aim to devise a Ramsey test analysis of actual causation. Our method is to define a strengthened Ramsey test for causal models. Unlike the accounts of Halpern and Pearl ([2005]) and Halpern ([2015]), the resulting analysis deals satisfactorily with both over- determination and conjunctive scenarios.

In this paper, we propose a regularity theory of causation. The theory aims to be reductive and to align with our pre‐theoretic understanding of the causal relation. We show that our theory can account for a wide range of causal scenarios, including isomorphic scenarios, omissions, and scenarios which suggest that causation is not transitive.

In the recent literature on causal and non-causal scientific explanations, there is an intuitive assumption according to which an explanation is non-causal by virtue of being abstract. In this context, to be ‘abstract’ means that the explanans in question leaves out many or almost all causal microphysical details of the target system. After motivating this assumption, we argue that the abstractness assumption, in placing the abstract and the causal character of an explanation in tension, is misguided in ways that are (...) independent of which view of causation or causal explanation one takes to be most accurate. On major accounts of causation, as well as on major accounts of causal explanation, the abstractness of an explanation is not sufficient for it being non-causal. That is, explanations are not non-causal by dint of being abstract. (shrink)

Philosophers have already recognized the importance of causal preemption involving “positive” events. First, preemption with positive events raises problems for counterfactual theories of causation. Second, theories of moral and legal responsibility rely heavily on the concept of causation, so accurately assessing responsibility in preemption cases requires correctly assessing their causal structure. However, philosophers have not discussed preemption involving “negative” events or omissions. This paper argues that cases of preemptive omissions exist and have important implications for theories of causation and for (...) moral and legal responsibility. Of theoretical importance, the alterations made to counterfactual theories of causation to address preemption with positive events do not seem to work for accommodating preemptive omissions. Of practical importance, there have been actual legal cases involving preemptive omissions, and at least one such case was, this paper contends, decided incorrectly on erroneous causal grounds. This paper identifies what must happen for preemptive omissions to obtain. It then argues for the existence of preemptive omissions by constructing a series of cases and drawing structural parallels between preemption cases with positive events and cases with omissions. It ultimately presents a formula for generating preemptive omissions and examines both why “traditional” methods of generating preemption fail for omissions and why the proposed method avoids such concerns. (shrink)

Causal reasoning and game-theoretic reasoning are fundamental topics in artificial intelligence, among many other disciplines: this paper is concerned with their intersection. Despite their importance, a formal framework that supports both these forms of reasoning has, until now, been lacking. We offer a solution in the form of (structural) causal games, which can be seen as extending Pearl's causal hierarchy to the game-theoretic domain, or as extending Koller and Milch's multi-agent influence diagrams to the causal domain. We then consider three (...) key questions: i) How can the (causal) dependencies in games - either between variables, or between strategies - be modelled in a uniform, principled manner? ii) How may causal queries be computed in causal games, and what assumptions does this require? iii) How do causal games compare to existing formalisms? To address question i), we introduce mechanised games, which encode dependencies between agents' decision rules and the distributions governing the game. In response to question ii), we present definitions of predictions, interventions, and counterfactuals, and discuss the assumptions required for each. Regarding question iii), we describe correspondences between causal games and other formalisms, and explain how causal games can be used to answer queries that other causal or game-theoretic models do not support. Finally, we highlight possible applications of causal games, aided by an extensive open-source Python library. (shrink)

Both philosophers and psychologists have argued for the existence of distinct kinds of explanations, including teleological explanations that cite functions or goals, and mechanistic explanations that cite causal mechanisms. Theories of causation, in contrast, have generally been unitary, with dominant theories focusing either on counterfactual dependence or on physical connections. This paper argues that both approaches to causation are psychologically real, with different modes of explanation promoting judgments more or less consistent with each approach. Two sets of experiments isolate the (...) contributions of counterfactual dependence and physical connections in causal ascriptions involving events with people, artifacts, or biological traits, and manipulate whether the events are construed teleologically or mechanistically. The findings suggest that when events are construed teleologically, causal ascriptions are sensitive to counterfactual dependence and relatively insensitive to the presence of physical connections, but when events are construed mechanistically, causal ascriptions are sensitive to both counterfactual dependence and physical connections. The conclusion introduces an account of causation, an "exportable dependence theory," that provides a way to understand the contributions of physical connections and teleology in terms of the functions of causal ascriptions. (shrink)

Engineering failure investigations seek to reconstruct the actual causes of major engineering failures. The investigators need to establish the existence of certain past events and the actual causal relationships that these events bear to the failures in question. In this paper, I examine one method for reconstructing the actual causes of failure events, which I call "feature dependence". The basic idea of feature dependence is that some features of an event are informative about the features of its causes; therefore, the (...) investigators can use the features of a known failure event to reconstruct details of its causes. I make explicit the structure of feature dependence and the evidential basis of its key premises, and show how feature dependence works in the investigation of the American Airlines Flight 191 accident. (shrink)

How should we reason in science? Jan Sprenger and Stephan Hartmann offer a refreshing take on classical topics in philosophy of science, using a single key concept to explain and to elucidate manifold aspects of scientific reasoning. They present good arguments and good inferences as being characterized by their effect on our rational degrees of belief. Refuting the view that there is no place for subjective attitudes in 'objective science', Sprenger and Hartmann explain the value of convincing evidence in terms (...) of a cycle of variations on the theme of representing rational degrees of belief by means of subjective probabilities (and changing them by Bayesian conditionalization). In doing so, they integrate Bayesian inference—the leading theory of rationality in social science—with the practice of 21st century science. Bayesian Philosophy of Science thereby shows how modeling such attitudes improves our understanding of causes, explanations, confirming evidence, and scientific models in general. It combines a scientifically minded and mathematically sophisticated approach with conceptual analysis and attention to methodological problems of modern science, especially in statistical inference, and is therefore a valuable resource for philosophers and scientific practitioners. (shrink)

Halpern and Hitchcock appealed to the normality of witness worlds to solve the problem of isomorphism in the Halpern-Pearl definition of actual causality. This paper first proposes a new isomorphism counterexample, called “bogus permission,” to show that their approach is unsuccessful. Then, to solve the problem of isomorphism, I propose a new improvement over the Halpern-Pearl definition by introducing default worlds. Finally, I demonstrate that my new definition can resolve more potential counterexamples than similar approaches in the current literature, including (...) the Lewisian causal dependence, Menziesian causal dependence, and modified version of the Halpern-Pearl definition. Some other advantages of my definition are also discussed. (shrink)

It has recently been argued that normative considerations play an important role in causal cognition. For instance, when an agent violates a moral rule and thereby produces a negative outcome, she will be judged to be much more of a cause of the outcome, compared to someone who performed the same action but did not violate a norm. While there is a substantial amount of evidence reporting these effects, it is still a matter of debate how this evidence is to (...) be interpreted. In this paper, we engage with the three most influential classes of explanations, namely, (a) the Norm‐Sensitive Cognitive Process View, (b) the Normative Concept View, and (c) the Pragmatics View. We will outline how these theories explain the empirical results and in what ways they differ. We conclude with a reflection on how well these strategies do overall and what questions they still leave unanswered. (shrink)

Imagine you and your friend Pierre agreed on meeting each other at a café, but he does not show up. What is the difference between a friend’s not showing up meeting? and any other person not coming? In some sense, all people who did not come show the same kind of behaviour, but most people would be willing to say that the absence of a friend who you expected to see is different in kind. In this paper, I will spell (...) out this difference by investigating laypeople’s conceptualisation of absences of actions in four experiments. In languages such as German, French, Italian, or Polish, people consider a friend’s not coming an omission. Any other person’s not coming, in contrast, is not considered an omission at all, but just a mere nothing. This use of the term omission differs from the usage in English, where ‘omission’ refers to all kinds of absences. In addition, ‘omission’ is not even an everyday term, but invented by philosophers for the sake of philosophical investigation. In other languages, ‘omission’ is part of an everyday vocabulary. Finally, I will discuss how this folk concept of omission could be made fruitful for philosophical questions. (shrink)

Explanations are backed by many different relations: causation, grounding, and arguably others too. But why are these different relations capable of backing explanations? In virtue of what are they explanatory? In this paper, I propose and defend a monistic account of explanation-backing relations. On my account, there is a single relation which backs all cases of explanation, and which explains why those other relations are explanation-backing.

Some have argued that our intuitive reactions to a number of cases of moral responsibility can only be preserved at the expense of a unified account of moral responsibility for acts and omissions. I argue against this conclusion, proposing that a plausible condition on responsibility, the Causal Condition can, when properly elaborated, justify the relevant intuitive data.

Necessity and sufficiency are the building blocks of all successful explanations. Yet despite their importance, these notions have been conceptually underdeveloped and inconsistently applied in explainable artificial intelligence, a fast-growing research area that is so far lacking in firm theoretical foundations. In this article, an expanded version of a paper originally presented at the 37th Conference on Uncertainty in Artificial Intelligence, we attempt to fill this gap. Building on work in logic, probability, and causality, we establish the central role of (...) necessity and sufficiency in XAI, unifying seemingly disparate methods in a single formal framework. We propose a novel formulation of these concepts, and demonstrate its advantages over leading alternatives. We present a sound and complete algorithm for computing explanatory factors with respect to a given context and set of agentive preferences, allowing users to identify necessary and sufficient conditions for desired outcomes at minimal cost. Experiments on real and simulated data confirm our method’s competitive performance against state of the art XAI tools on a diverse array of tasks. (shrink)

We present a probabilistic extension to active path analyses of token causation (Halpern & Pearl 2001, forthcoming; Hitchcock 2001). The extension uses the generalized notion of intervention presented in (Korb et al. 2004): we allow an intervention to set any probability distribution over the intervention variables, not just a single value. The resulting account can handle a wide range of examples. We do not claim the account is complete --- only that it fills an obvious gap in previous active-path approaches. (...) It still succumbs to recent counterexamples by Hiddleston (2005), because it does not explicitly consider causal processes. We claim three benefits: a detailed comparison of three active-path approaches, a probabilistic extension for each, and an algorithmic formulation. (shrink)

A cause is proportional to its effect when, roughly speaking, it is at the right level of detail. There is a lively debate about whether proportionality is a necessary condition for causation. One of the main arguments against a proportionality constraint on causation is that many ordinary and seemingly perfectly acceptable causal claims cite causes that are not proportional to their effects. In this paper, I suggest that proponents of a proportionality constraint can respond to this objection by developing an (...) idea that is present in Yablo’s early work on proportionality, but which has strangely been ignored by both Yablo and others in the subsequent debate. My suggestion is that proportionality—and, indeed, causation itself—is relative to a domain of events. At the metaphysical level, this means that the causal relation has an extra relatum—namely, a domain of events. At the level of language, it introduces a new way in which causal claims are context-sensitive: what is expressed by a causal claim depends on the contextually relevant domain of events. As I argue, this suggestion allows us to accommodate the truth of ordinary causal claims while extending the explanatory benefits of a proportionality constraint. (shrink)

I argue that there are some causal relata for which it is indeterminate whether one caused the other. Positing indeterminacy in causation helps us defend contested principles in the logic of causation and makes possible new ways of thinking about the theoretical impact of symmetric causal overdetermination. I close by discussing amendments of current theories of causation that would help explain causal indeterminacy.

I argue that there are some causal relata for which it is indeterminate whether one caused the other. Positing indeterminacy in causation helps us defend contested principles in the logic of causation and makes possible new ways of thinking about the theoretical impact of symmetric causal overdetermination. I close by discussing amendments of current theories of causation that would help explain causal indeterminacy.

I argue that there are some causal relata for which it is indeterminate whether one caused the other. Positing indeterminacy in causation helps us defend contested principles in the logic of causation and makes possible new ways of thinking about the theoretical impact of symmetric causal overdetermination. I close by discussing amendments of current theories of causation that would help explain causal indeterminacy.