Classic studies of explanation, such as those of Hempel and Bromberger, took it for granted that an explanation-seeking question of the form "Why P?" should be understood as asking about the proposition P. This view has been recently challenged by Bas van Fraassen and Alan Garfinkel. They acknowledge that some questions have the surface form "Why P?", but they hold that a correct reading for why-questions should take the form "Why P (rather than Q)?", where Q is a contrasting alternative. (...) This contrast theory is discussed here. It is argued that, properly understood, the contrast theory and the propositional approach can and should give equivalent readings to why-questions. (shrink)

William Lane Craig has defended a modern First Cause argument based on 1) a principle of universal causality and 2) the claim that the universe must have had a beginning. But 1) is susceptible to counter examples from quantum theory. Moreover, Craig’s defense of 2) is open to serious question. He claims that an actual infinity (of time) is impossible; he also claims that 2) is in fact supported by big bang theory. I argue that both of these claims are (...) mistaken, and that in consequence we have no particular reason to suppose that 2) is true. I conclude that the First Cause argument fails, but I suggest that a weaker inductive argument might be worth a try. (shrink)

SummaryIt is argued, contrary to Nelson Goodman, that confirmability is not a semantical property possessed by some hypotheses. Instead, hypotheses are confirmed or disconfirmed on the basis of all relevant information, not just postivie or negative instances.

Summary“There are some types of reasoning which are acceptable in a given situation but not justifiable according to the rules of formal logic. This sort of reasoning seems to depend on a judgment about what the speaker knows along with an Assumption of Maximum Information, that if the speaker is serious he is making the logically strongest statement he knows to be true. Because such reasoning can be informally correct, formal logic should be understood as establishing rules not for all (...) acts of inference but only for those which do not depend on contingent judgments about the context of an assertion.”RésuméIl y a des types de raisonnement qui sont acceptables dans une situation donnée, mais qui ne sont pas justifiables selon les règles de la logique formelle. Cette sorte de raisonnement semble dépendre d'un jugement sur ce que sait celui qui parle, ainsi que d'un Postulat d'Information Maximale selon lequel, si le locuteur est sérieux, il choisit l'énoncé le plus fort logiquement parmi ceux qu'il sait être vrais. Un tel raisonnement étant informellement correct, la logique formelle devrait être conçue comme édictant des règles non pas pour tous les actes d'inférence, mais seulement pour ceux qui ne dépendent pas de jugements contingents quant au contexte d'une assertion.Zusammenfassung〈Es gibt Argumentationsweisen, die in einer bestimmten Situation annehmbar sind, obschon sie sich auf Grund der Regeln der formalen Logik nicht rechtfertigen lassen. Sie hängen ab von einem Urteil über das, was der Redende weiss, wie auch von der Annahme maximaler Information, die besagt, dass er, wenn er im Ernst spricht, auch die starkste Aussage macht, von der er weiss, dass sie wahr ist. Weil es korrekte informelle Folgerungen dieser Art gibt, kann man sagen, dass die formate Logik nicht Regeln fiir aile Schlussfolgerungen liefert, sondern nur für solche, die nicht von kontingenten Urteilen über den Aeusserungskontext abhängen.〉. (shrink)