Whether we’re reading H.G. Wells, Robert Heinlein, Ray Bradbury, or Kurt Vonnegut, time travel is a wonderful narrative trick, freeing a story from the normal ‘one damn thing after another’ progression of time. But many philosophers claim it can never be more than that because backwards causation in general, and time travel in particular, are logically impossible.

In this paper I examine one type of argument commonly given for this disappointing conclusion: the time travel paradoxes. Happily for science fiction fans, these arguments fall far short of showing what they are intended to show. Why they fail can be better understood in the light of an analogy between these arguments and some arguments libertarians offer against determinism.

Contemporary arguments for scientific realism are typically based on some form of inference to the best explanation. Sometimes such arguments concern the methods of science: given the success of scientific methodology, realism offers the best explanation of this success. Sometimes such arguments concern the content of scientific theories: given observed regularities in nature, explanations must be given of those regularities; the best such explanations will be realist. One forceful explanationist argument about the content of science can be based on Hans Reichenbach’s Principle of the Common Cause (CC). Roughly, CC says that whenever an enduring correlation between two events is observed, there must be some preceding event which is the common cause of the two, and is the explanation for the correlation. Rigid adherence to this demand for a common cause, however, supposedly goes beyond the instrumentalist explanatory resources available to the antirealist, and hence CC is taken to be a sufficient condition for realism. Further, one might claim that any argument that realism has greater explanatory power than antirealism depends either explicitly or implicitly on CC, and hence CC is also a necessary condition for realism.

The focus here will be on democracy as a social good. Today people want to live in a democratic society. But why do they see this as a good? Exploring this question will lead us to a justification of democracy. I shall not be concerned primarily with justifying full or ideal democracy; for it is important to show the positive effects of even modest democratic gains. I do not think one can justify democracy in the more demanding sense of showing it has consequences that any rational being would desire. So I shall be content to give what might seem merely a social, or group-based, rather than a purely rational justification of democracy.

John Martin Fischer has recently proposed that actions and omissions are asymmetric with respect to the requirement of alternative possibilities for moral responsibility: whereas moral responsibility for an action does not require freedom to refrain from performing the action, moral responsibility for failure to perform an action does require freedom to perform the action. In what follows, I first critically assess Fischer's asymmetry principle. In arguing against the principle, I raise some concerns about Fischer's association of responsibility with control. I then motivate a riddle regarding omissions: some cases appear to show that a person is not responsible for failing to bring about something in virtue of the fact that the person could not bring about that thing. Other cases, though, seemingly show that a person is responsible for failing to bring about something even though the person could not bring about that thing. What explains the asymmetry in responsibility attributions in these cases involving omissions? Third, I consider some answers to this riddle and explain why they are inadequate. Finally, I sketch my own answer.

The Principle of the Identity of Indiscernibles (PH), commonly attributed to Leibniz, has given rise to much discussion and debate. Thus philosophers have argued over how it should be formulated, whether it is (necessarily) true, and what, if any, metaphysical consequences it has.

It is not my intention to add to these discussions here, having done so elsewhere. Rather, I intend to introduce and defend a closely related principle which I shall, for want of a better name, refer to as The Principle of the Identity of Necessary Similarity’ (PINS).

In section II, I briefly recapitulate some of the distinctions and other relevant points which are customarily made in connection with (PH). This is all too familiar material, but it is necessary in order to provide a general framework of concepts in terms of which we can discuss (PINS) in an intelligible fashion.