My research to date has been shaped by a broad interest in the relationship between causation and counterfactuals. This has led me to my more specific work on the agency theory and the concept of causation–the subjects of much of my current research. This research has culminated in my Ph.D thesis The Trouble with Tokens: Old Problems for the New Agency Theory of Causation. For an abstract of my thesis, click here.
Past Work I conducted research into the function of counterfactuals. On the Thinking Counterfactually project, I worked on two major projects. The first concerned objective probability and metaphysical necessity. There is a venerable research program that seeks to analyse (or at least, characterise) metaphysical modality in terms of objective probability. Together with Prof. John Divers, I investigated the prospects for such a program, with an eye to arguing that either objective probability is itself irreducibly modal, or else there is little hope of giving a characterisation of modality in terms of objective probability. In either case, an analysis of the former in terms of the latter does not seem forthcoming. The second project concerned Williamson's modal epistemology. Williamson seeks to naturalise modal epistemology by subsuming modal reasoning under ordinary counterfactual reasoning. I argued that this requires an argument for the indispensability of counterfactual reasoning--an argument that Williamson does not provide. |
© Shyane Siriwardena
|
Publications
Siriwardena, S. (2018) Old problems for the agency theory of causal discourse. Erkenntnis. [link]
Works in Progress
Drafts available upon request. Please contact me by email.
Drafts available upon request. Please contact me by email.
“Counterfactuals, Causal Relevance and the Suppositional Theory” (Under review at Thought)
The suppositional theory (ST) of counterfactuals (Edgington (2004)) understands the chance of a counterfactual to be equal to the chance of its consequent conditional on its antecedent and on certain events occurring in the time between the two—in particular, those events that are causally independent of the antecedent. I present a new counterexample to ST that ostensibly threatens this causal relevance condition. I consider and reject an argument for opposing ST on these grounds. I then show that ST and its causal relevance constraint are underspecified. When the constraint is cashed out in contrastive causal terms, the resultant contrastive suppositional theory can successfully accommodate the putative counterexample. Furthermore, it can do so while preserving the original theory’s successes. The upshot of the argument is this: any counterfactual semantics that employs a causal relevance constraint must commit itself to some particular theory of causation if it is to be complete.
“Explaining Causal Concepts: Genealogies and Practical Explications”
Huw Price’s (1993 (with Peter Menzies), 1996, 2007, Forthcoming) agency theory seeks to reveal the function of our present concept of causation by way of a fictional genealogy of that concept. The genealogical fiction is meant to arrive at a concept like ours by way of a naturalistic story about human beings with our basic cognitive capacities and interests. Importantly, it is not Price’s intention to be at all revisionist about our concept of causation; the intention is instead to tell a fictional story about the genesis and evolution of some concept, and then to show that that concept is just the same as the concept of causation that we in fact employ. This constitutes the Pricean project of “philosophical anthropology.” The difficulty is this: it is not clear why a fictional genealogy should compel us to believe anything about our actual concept of causation. Thus, to the extent that fictional genealogies are methodologically dubious, it would be advantageous to the agency theory to achieve its functional account of our causal concept by way of a different method. In this paper, I seek to offer just such an account.
“Manipulationist Theories of the Concept of Causation and the Placement Problem”
I explore three distinct reasons why the agent-theoretic and interventionist concepts of causation might be taken to be unsuitable for use in scientific theory: non-reductiveness, relativity, and anthropocentricity. In each case, I demonstrate that these characteristics of the manipulationist concepts do not threaten the use of those concepts in scientific theorising. After clarifying what each of these notions mean, and in what way there are supposed to make the relevant concept of causation inconsistent with scientific theorising, I argue that where the characterisations of the concept of causation are faithful to the agency or interventionist theories, they are not problematic in the way suggested.
“Lewis and Woodward on Token Causation”
I compare the Lewisian influence theory of causation with Woodward’s manipulationist theory, and attempt to show that the latter is subject to some of the same counterexamples as the former. Woodward maintains that his theory improves on Lewis’ by avoiding many of the counterexamples that the influence theory faces. By elaborating on an example from Igal Kvart (2001), I argue that Woodward’s theory is no more successful than Lewis’ in the face of preemption cases involving effects that are insensitive to alterations in their respective causes (i.e. effects for which there are only two alterations: their occurrence, and their non-occurrence). I also run through a series of smaller objections from Woodward against Lewis’ influence theory, and offer responses on Lewis’ behalf.
The suppositional theory (ST) of counterfactuals (Edgington (2004)) understands the chance of a counterfactual to be equal to the chance of its consequent conditional on its antecedent and on certain events occurring in the time between the two—in particular, those events that are causally independent of the antecedent. I present a new counterexample to ST that ostensibly threatens this causal relevance condition. I consider and reject an argument for opposing ST on these grounds. I then show that ST and its causal relevance constraint are underspecified. When the constraint is cashed out in contrastive causal terms, the resultant contrastive suppositional theory can successfully accommodate the putative counterexample. Furthermore, it can do so while preserving the original theory’s successes. The upshot of the argument is this: any counterfactual semantics that employs a causal relevance constraint must commit itself to some particular theory of causation if it is to be complete.
“Explaining Causal Concepts: Genealogies and Practical Explications”
Huw Price’s (1993 (with Peter Menzies), 1996, 2007, Forthcoming) agency theory seeks to reveal the function of our present concept of causation by way of a fictional genealogy of that concept. The genealogical fiction is meant to arrive at a concept like ours by way of a naturalistic story about human beings with our basic cognitive capacities and interests. Importantly, it is not Price’s intention to be at all revisionist about our concept of causation; the intention is instead to tell a fictional story about the genesis and evolution of some concept, and then to show that that concept is just the same as the concept of causation that we in fact employ. This constitutes the Pricean project of “philosophical anthropology.” The difficulty is this: it is not clear why a fictional genealogy should compel us to believe anything about our actual concept of causation. Thus, to the extent that fictional genealogies are methodologically dubious, it would be advantageous to the agency theory to achieve its functional account of our causal concept by way of a different method. In this paper, I seek to offer just such an account.
“Manipulationist Theories of the Concept of Causation and the Placement Problem”
I explore three distinct reasons why the agent-theoretic and interventionist concepts of causation might be taken to be unsuitable for use in scientific theory: non-reductiveness, relativity, and anthropocentricity. In each case, I demonstrate that these characteristics of the manipulationist concepts do not threaten the use of those concepts in scientific theorising. After clarifying what each of these notions mean, and in what way there are supposed to make the relevant concept of causation inconsistent with scientific theorising, I argue that where the characterisations of the concept of causation are faithful to the agency or interventionist theories, they are not problematic in the way suggested.
“Lewis and Woodward on Token Causation”
I compare the Lewisian influence theory of causation with Woodward’s manipulationist theory, and attempt to show that the latter is subject to some of the same counterexamples as the former. Woodward maintains that his theory improves on Lewis’ by avoiding many of the counterexamples that the influence theory faces. By elaborating on an example from Igal Kvart (2001), I argue that Woodward’s theory is no more successful than Lewis’ in the face of preemption cases involving effects that are insensitive to alterations in their respective causes (i.e. effects for which there are only two alterations: their occurrence, and their non-occurrence). I also run through a series of smaller objections from Woodward against Lewis’ influence theory, and offer responses on Lewis’ behalf.