The standard account of counterfactuals that most philosophers endorse—Lewis's 'Analysis 1' — is wrong. The correct theory is one invented by Jonathan Bennett in 1984 which he called 'The Simple Theory'. Bennett later argued himself out of that theory and went on to champion the standard account. But those arguments fail. The Simple Theory has been right all along.

The Computational Theory of the Laws of Nature entails that the accessibility relation for nomological necessity is not symmetric or transitive. This means that nomologically possible worlds need not share our world's laws. This subverts a standard style of argument against Humean Supervenience.

Causation is defined as a relation between facts: C causes E if and only if C and E are nomologically independent facts and C is a necessary part of a nomologically sufficient condition for E. The analysis is applied to problems of overdetermination, preemption, trumping, intransitivity, switching, and double prevention. Preventing and allowing are defined and distinguished from causing. The analysis explains the direction of causation in terms of the logical form of dynamic laws. Even in a universe that is (...) deterministic in both temporal directions, not every fact must have a cause and present facts may have no future causes. (shrink)

The principle of Information Conservation or Determinism is a governing assumption of physical theory. Determinism has counterfactual consequences. It entails that if the present were different, then the future would be different. But determinism is temporally symmetric: it entails that if the present were different, the past would also have to be different. This runs contrary to our commonsense intuition that what has happened in the future depends on the past in a way the past does not depend on the (...) future. To understand how this can be so we observe that while the truth of some counterfactuals is guaranteed by the laws of logic or the laws of nature, some are not. It is among the latter contingent, counterfactuals that we find temporal asymmetry. It is this asymmetry that gives causation a temporal direction. The temporal asymmetry of these counterfactuals is explained by the fact that the dynamical laws of nature are logically irreversible functions from partial states of the world onto other partial states. (Logical reversibility is not to be confused, though it too often is, with time-reversal invariance). Though these irreversible laws are locally indeterministic, they can sum to give a globally deterministic description of the world. This combination of global determinism and local indeterminism gives rise to contingent counterfactual dependence and gives that dependence a direction. That direction is independent of the direction of entropy. The direction of contingent counterfactual dependence is time's arrow. (shrink)

A new account of the of the laws of nature based upon Algorithmic Information Theory.

Our justification for punishing the wrong doer is not that we are enacting God-like retribution. Neither do we have to argue that inflicting the punishment will make any person, living or dead, happier or better off. We punish to keep a promise to the victims: a promise made before they were victims, a promise they were entitled to ask for; that we were entitled to give and that we are now obliged to honor.

Quine taught us that the collapse of positivism entails that empirical theories are, in principle, undetermined-- not just by the available evidence-- but by all possible evidence. Without disputing that conclusion, contemporary philosophers-- exampled here by Simon Blackburn and Jerry Fodor-- have wanted to treat this as a merely abstract possibility that need not undermine our confidence in actual scientific theory and practice. I argue that there is no basis for this complacency.

Introduces 'Turing Worlds' as a device for thinking about Metaphysical Problems and uses them to examine several different theories of counter-factuals.

If there are rights there is surely a right to self-defense. But self-defense has proved very puzzling to rights theorists. The central puzzle has been called the "paradox of self-defense": If our right not to be harmed gives rise to our right to fight back, what happens to the attacker's right not to be harmed when the defender fights back? If the attacker somehow forfeits his right to self-defense because he is a bad actor, what do we say about innocent (...) threats? I argue that these problems stem from a confusion about the nature of rights. (shrink)

While there is agreement that it is the central task of semantics to give the semantic interpretation of every sentence in the language, nowhere in the linguistic literature will one find, so far as I know, a straightforward account of how a theory performs this task, or how to tell when it has been accomplished. The contrast with syntax is striking. The main Job of a modest syntax is to characterizemeaningfulness. We may have as much confidence in the correctness of (...) such a characterization as we have in the representativeness of our sample and our ability to say when particular expressions are meaningful. What clear and analogous task and test exist for semantics? (shrink)