Overlappings: Probability-raising without causation

Australasian Journal of Philosophy

Volumn 78, Issue 1, 2000, Pages 40-46

  Schaffer, Jonathan ^a  

^a ST FRANCIS XAVIER UNIVERSITY   (Canada)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 34249737845     PISSN: 00048402     EISSN: None     Source Type: Journal    
DOI: 10.1080/00048400012349331     Document Type: Article

References (14)

1. Patrick Suppes, A Probabilistic Theory of Causality (Amsterdam: North-Holland, 1970).
2. David Lewis, Philosophical Papers volume 2 (Oxford: Oxford University Press, 1986). Note that this is only a sufficiency claim ('C causes E if...'): Lewis weakens this condition to try to capture necessity (especially in light of preemption) by taking its ancestral, and later by taking the ancestral of counterfactual-dependence or quasi-dependence. These revisions merely allow more things to count as causes, and so couldn't help discount the relevance of the noncause to the overlapping projected effect. Also Lewis actually requires the stronger condition that had C not occurred then the chance of E would have been less by a large factor. But set the projected effectiveness of the cause in overlapping cases (e.g., Merlin's spell) as low in the (0,1) interval as you like, and the projected effectiveness of the noncause (e.g., Morgana's spell) as high in (0,1) as you like, and you can get counterfactual-chance dependence to an arbitrarily large factor.
3. Huw Price, 'Agency and Probabilistic Causality', British Journal for the Philosophy of Science 42 (1991), pp. 157-176.
4. Also Huw Price and Peter Menzies, 'Causation as a Secondary Quality', British Journal for the Philosophy of Science 44 (1993), pp. 187-203.
5. There are two further challenges to the sufficiency of probability-raising for causation in the literature which are orthogonal to the issues raised here: causal asymmetry, in which C is a probability-raiser of E because C is an effect of, or correlate from a common cause with, E; and trivial relevance, in which C is a probability-raiser of E because C and E stand in analytic or mereological entailment relations (the trivial relevance challenge is due to Jaegwon Kim, 'Causes and Counterfactuals', Journal of Philosophy 70 (1973), pp. 570-572.) The standard reply to the asymmetry challenge is to give an independent account of the causal arrow to conjoin with or integrate into the probability-raising relation (for Suppes, the temporal arrow plus the screening condition; for Lewis, the arrow of overdetermination as it grounds the distinction between standard and backtracking counterfactuals; for Price, the subjective arrow projected from the agent's perspective; a number of other constructions are available
6. (See Daniel Hausman's Causal Asymmetries (Cambridge: Cambridge University Press, 1998) for a comprehensive discussion.) The standard reply to the trivial relevance challenge is to give an independent theory of events designed to rule out these cases. (See Lewis, op. cit. for a discussion along these lines.) Thus, strictly speaking, what is assumed in Suppes, Lewis, and Price is that probability-raising between actual, distinct events, plus causal priority, suffices for causation.
7. Peter Menzies, 'Probabilistic Causation and Causal Processes: A Critique of Lewis', Philosophy of Science 56 (1989), pp. 642-663.
8. D. H. Mellor, The Facts of Causation (London: Routledge, 1995).
9. Ellery Eells, Probabilistic Causality (Cambridge: Cambridge University Press. 1991).
10. Igal Kvart, 'Cause and Some Positive Causal Impact' in James Tomberlin (ed.), Philosophical Perspectives 11: Mind, Causation, and World (Oxford: Basil Blackwell, 1997), pp. 401-432.
11. Ned Hall, 'Two Concepts of Causation' (forthcoming).
12. Deborah Rosen, 'In Defense of a Probabilistic Theory of Causality', Philosophy of Science 45 (1978), pp. 604-613;
13. Paul Humphreys, 'Cutting the Causal Chain', Pacific Philosophical Quarterly 61 (1980), pp. 305-314. It is worth asking if there are cases of genuinely causal probability-raising that disappear under precision. How about this: I bet you the die will land with 1-3 showing, and slip you a trick die with faces 1,2,2,3,4,6. The die lands 1-3. I win. Intuitively, my giving you the trick die caused my winning, as can be seen probabilistically: p(die lands 1-3|trick die) > p (die lands l-3|normal die), and as would surely be echoed in your protests. But wait: the die actually landed 1, and p(die lands l|trick die) = p(die lands l|normal die).
14. Michael Tooley, Causation: A Realist Approach (Oxford: Clarendon Press, 1987).