Actual Causation
Workshop 23–24 Sept. 2010
Organizers: Wolfgang Spohn, Luke Glynn, Michael Baumgartner
Speakers: Michael Baumgartner (University of Konstanz), David Danks (Carnegie Mellon University), Isabelle Drouet (CNRS/IHPST), Luke Glynn (University of Konstanz), Ned Hall (Harvard University), Jens Harbecke (University of Witten), Christopher Hitchcock (CalTech), Franz Huber (University of Konstanz), Max Kistler (University Paris 1 Panthéon-Sorbonne), and L.A. Paul (University of North Carolina)
For many decades, counterfactual theories of causation were alone in their concern with token-level or actual causation, whereas the other theoretical frameworks primarily analyzed type-level or generic causation. In recent years, however, the overall focus has noticeably shifted towards actual causation. Among the issues that have attracted philosophers' interest are, for example, the relational properties of actual causation, its context-sensitivity, or questions concerning cases of overdetermination, preemption, and prevention. Theories of actual causation developed over the past 10 years are (unsurprisingly) as diverse as the problems they tackle. Their philosophical motivations, foundations, and theoretical credentials diverge, just as do their assessments of many exemplary causal processes. This workshop brings together some of the leading authors that have shaped the recent debates on actual causation. It provides a forum for discussion and exchange of the newest results and ideas.
Program
Thursday 23rd September
09:45–11:00 C. Hitchcock: Actual Causation and Defaults (Chair: Baumgartner)
11:00–11:15 Coffee Break
11:15–12:30 L. Glynn: Of Miracles and Interventions (Chair: Baumgartner)
12:30–14:00 Lunch Break
14:00–15:15 F. Huber: Structural Equations and Beyond (Chair: Reiss)
15:15–16:30 D. Danks: The Difficulties of Actual Causation (Chair: Reiss)
16:30–16:45 Coffee Break
16:45–18:00 M. Kistler: Actual causation and simultaneous lawful dependence (Chair: Drouet)
Friday 24th September
09:45–11:00 M. Baumgartner: A Regularity Theoretic Approach to Actual Causation (Chair: Glynn)
11:00–11:15 Coffee Break
11:15–12:30 J. Harbecke: Regularities, Counterfactuals, and Causation (Chair: Spohn)
12:30–14:00 Lunch Break
14:00–15:15 L. Paul: Causal Overdetermination (Chair: Huber)
15:15–16:30 N. Hall: In Praise of Causal Mechanisms (Chair: Huber)
16:30–16:45 Coffee Break
16:45–18:00 I. Drouet: Causal Singular Probabilities (Chair: Glynn)
Titles/Abstracts
Actual Causation and Defaults
Christopher Hitchcock
A number of authors have attempted to define actual causation using structural equation models. Arguments from a number of directions suggest that an adequate account of actual causation will have make use of information about defaults and norms. First, existing accounts are subject to straightforward counterexamples. Second, evidence from social psychology and experimental philosophy suggests that norms affect our judgments of actual causation. Third, building in information about norms and defaults allows us to capture some of the diversity in causal judgments reported in the philosophical literature. We provide a ‘graded’ account of actual causation: events are not simply judged as actual causes or not, but are graded in comparison with other candidate causes. This allows us to capture some of the subtler features of causal judgment. In addition to presenting the details of the theory, I will discuss some of the technical challenges involved, and address a number of philosophical objections.
Structural Equations and Beyond
Franz Huber
Recent accounts of actual causation are stated in terms of extended causal models. These contain two elements representing two seemingly distinct modalities: structural equations representing counterfactual structure, and ranking functions representing normality. This paper presents counterfactual models which contain just one element representing one modality: objective ranking functions representing counterfactuality. I will formulate a condition on extended causal models that relates counterfactual structure and normality and show that extended causal models satisfying this condition can be subsumed under counterfactual models. Then I will formulate another condition on extended causal models that relates counterfactuality and actuality and show that extended causal models additionally satisfying this second condition can be subsumed under counterfactual models in a conservative way. Consequently there is only one modality that is needed for an account of actual causation.
Regularities, Counterfactuals, and Causation
Jens Harbecke
As it is evident from sec. 7 of the Enquiry, Hume believed in a close connection between a regularity definition and a counterfactual definition of causation. This view stands in sharp contrast to how the relation of these approaches is conceived of today. Research on both is still carried out widely independently, and theorists on both strands have often described the respective other framework as an unsuccessful and deficient rival of their own theory. The aim of my talk is to show that regularity and counterfactual approaches should not be considered as rivals but as two highly interdependent analyses of causation. Causally interpretable regularities often presuppose the truth of certain counterfactuals, whereas causally interpretable counterfactuals typically require specific regularities to hold at the actual world. Furthermore, the two branches display striking correspondences with respect to what are commonly considered their most severe counterexamples. The interdependence and the correspondences of the two approaches suggests that Hume’s intuition was correct and that the two approaches ought to be considered as two expressions of the same underlying idea. This idea I will call “nomalous causation”. To substantiate my claims, I first collect two general observations that may be considered as supporting the orthodox distinction between the regularity and the counterfactual definition of causation. I then demonstrate that the most recent versions of counterfactual and regularity theories of causation are highly interdependent nevertheless. The interdependence is used to argue for an “inner connection” of the two approaches. I close with a proposal of how actual causation is to be thought of within this broader Humean theory of causation.
A Regularity Theoretic Approach to Actual Causation
Michael Baumgartner
Danks et al. (2010) justifiably doubt that, in light of the unmanageable amount of possible counterexamples and of the muddy intuitive background against which (‘Socratic’) theories of actual causation are typically assessed, an entirely satisfactory theory will ever be available. Notwithstanding these dark prospects, accounts of actual causation are currently flourishing like never before—just to mention three examples: the blueprint theory developed by Hall & Paul (2003), Hitchcock’s (2001, 2007) structural equations account, or Hall’s (2007) default/deviant theory (cf. also Hitchcock 2009).
In this talk, I add another (old) candidate theory to the list. I re-animate a regularity theoretic approach to actual causation. I do not claim that all problem cases can be satisfactorily captured within this regularity theoretic framework, but I argue that a sufficiently sophisticated regularity theory performs at least as well when it comes to reproducing causal dependencies on token level as do the more recent accounts. Moreover, while the latter draw on highly intricate or context-sensitive conceptual resources, as explicitly nonforetracking counterfactuals or a distinction between default and deviant values of variables, the regularity theory I propose only resorts to material conditionals, to standard Boolean minimization procedures, and to non-modal and context-independent conditions that regulate the behavior of causal models under model expansions.
In Praise of Causal Mechanisms
Ned Hall
Consider two theses about causation: (1) Causes are connected to their effects by way of mediating causal mechanisms or processes. (2) Scientific inquiry aims (at least in part) at discerning and describing the causal structure of our world. Some of the best contemporary work on causation claims—often implicitly, but sometimes quite explicitly—that, in giving an account of causation, we should sacrifice (1) for the sake of producing an account that makes the best sense of (2). I will first try to show why this claim is quite attractive, and then obstreperously argue against it: I will aim to show that the best way to make sense of (2) is, in fact, by means of an account of causal structure that fully vindicates (1).
Of Miracles and Interventions
Luke Glynn
Lewis claims that, for the purposes of delivering a counterfactual analysis of causation, counterfactuals of the form ‘if c hadn’t occurred, then e wouldn’t have occurred’ should be evaluated with respect to those possible worlds in which the non-occurrence of c is realised by a small miracle occurring shortly beforehand. Woodward disagrees. According to him, such counterfactuals ought to be evaluated with respect to those worlds in which c’s non-occurrence is realised by an intervention on whether-or-not c occurs with respect to e. The notion of an intervention, unlike that of a miracle, is causal and so Woodward’s analysis of causation is non-reductive. As one might expect, Woodward claims compensating advantages for his account. In this paper, I argue that these advantages can be had without the appeal to interventions and the consequent sacrifice of potential reductivity.
The Difficulties of Actual Causation
David Danks
Causal Overdetermination
Laurie Paul
Causal Singular Probabilities
Isabelle Drouet
Causal interpretations of singular probabilities in the line of Popper's propensity theory are traditionally rejected on the ground that usual probability axioms do not exhibit the same kind of asymmetry as causal relations. This has led to the widely shared view that if there exists a deep conceptual connection between singular probability and actual causation, the connection must consist in a probabilistic theory of actual causation. Recently, such a theory has been put forward by Luke Glynn. Starting from a criticism of the traditional argument against causal interpretations of singular probabilities, I envisage the possibility of causally interpreting singular probabilities. In the light of Glynn's account, it is investigated whether the apparent difficulties of the project are specific to it, or rather generically affect any theory that links singular probabilities to causation.
Actual causation and simultaneous lawful dependence
Max Kistler
Recent accounts of actual causation in terms of interventions seem to be unable to cope with simultaneous lawful dependence. Many physical laws have the form of equations according to which the values of two or more variables are mutually and simultaneously dependent on each other. This raises the problem that interventionist conditions are not sufficient for causation. I look at five ways of solving this problem. A second problem stems from the fact that it seems to be impossible to intervene (in the technical sense of Woodward 2003) on variables that are related by simultaneous lawful dependence or supervenience. I look at six ways of solving this problem. I come to the conclusion that recent theories intended to account for actual causation really account for a different relation: lawful dependence.