1.1 Why a counterfactual analysis is better than its alternatives as a general theory of (singular) causation?

Rival accounts of (singular) causation:

• Acquaintance-based accounts: Causal relations are directly perceptible. And our knowledge about them comes from direct perception alone.
• The “Canberra plan” for causation: Causal relations are whatever relations that fill in those conceptual roles according to our folk theory of causation. These relations may differ in different possible worlds.
• The deductive-nomological analysis: C causes E iff C is a member of a set of conditions jointly sufficent for E.
• Counterfactual analyses

Many rival theories, such as acquaintance-based theories or the “Canberra plan” for causation, face the two problems.

• The miscellany objection: There appear to be many actual and possible variants of singular causation, and a general theory of singular causation should revealed what they have in common.
  • There are causal relations not directly perceptible and we still have knowledge of their existence. E.g.
    • The causal mechanism by which low carbohydrate in the dinner causes low blood sugar or that by which a congressman’s behavior caused his own defeat
    • unobservable causation through magic (e.g. in Harry Potter)
  • Acquaintance-based theories can at most account for some, but not all, of them. And the “Canberra plan” for causation tells us nothing about what all actual variants of singular causation share in common.
  • Note that this assumes causal monism (at least w.r.t. the concept of causation).
• The missing relata objection: Besides causation of positive events by positive events, there are also “causation by absences, causation of absences, and causation via absences as intermediate steps”. A general theory of singular causation should allow absences as suitable causal relata.
  • Acquaintance-based theories and the “Canberra plan” for causation cannot do this job.
  • ME: Really?

1.2 Probabilistic causation

Whether part of, or even all, causal relations are indeterministic is a contingent matter to be studied by theoretical physics.

• Different interpretations of quantum mechanics take different stances on this issue.

Probabilistic version of counterfactual analysis: “C occurs, E has a certain chance (objective single-case probability) of occurring, and as it happens E does occur; but without C, E’s chance would have been less.”

• Problem: Event A may raise the probability of event B without “causing” it (in the sense that we intuitively do not think A causes B in this case). It is hard to find a noncircular way to distinguish causal instances of probability-raising from non-causal ones (noncircular in the sense of not appealing to causal notions).

  One terrorist places an unreliable bomb—a genuinely indeterministic device—on Flight 13; another terrorist places an unreliable bomb on Flight 17. As it happens, the bomb on Flight 13 goes off and the bomb on Flight 17 doesn’t. The Age runs a headline: ‘‘Airline bomb disaster.’’ The headline would have been just the same if it had been the bomb on Flight 17 that went off, or if it had been both. So the bomb on Flight 17 raised the probability of the headline, but certainly didn’t cause it. We want to say that the raising that counts is the raising of the probability of the causal chain of events and absences whereby the effect was actually caused. Raising the probability of some unactualized alternative causal chain leading to the same effect doesn’t count. But it would be circular to say it that way within an analysis of causation.