An orthodox view says that the most primitive kind causation involves one event causing another event. I argue against this view. Instead, the most primitive kind of causation is “thing causation”, which involves a thing causing a thing to do something. I give a definition of event causation from thing causation. Appealing to the phenomenon of “fine-grained” causation, I argue that thing causation is not definable from event causation.
Consider some basketball fans watching a game, cheering for their team, and getting very worked up. Why do they care so much about whether a ball goes through one hoop rather than another? Are they attuned to some great significance of the outcome of the game? Or do they vastly overestimate its importance? Or is there some other way of making sense of their investment in the outcome? Or are they simply irrational? I argue that the answer to all these questions is “no”. Fans usually don’t believe that the outcome of a game matters; instead, they pretend it matters, and this causes them to have the phenomenology of caring about the outcome. Having the phenomenology of caring is not the same as caring. The same, I argue, is true of players of games, and this is a central feature of games: if you want to play a game, or spectate enthusiastically, you pretend that the outcome of a game is important. I develop this, making use of machinery that Kendall Walton deploys in trying to solve the paradoxes of fiction, and carrying out his idea that it be applied to sports. I conclude with some general observations about the ethics and metaphysics of caring.
A popular principle about grounding, "Internality", says that if A grounds B, then necessarily, if A and B obtain, then A grounds B. I argue that Internality is false. Its falsity reveals a distinctive, new kind of explanation, which I call "ennobling". Its falsity also entails that every previously proposed theory of what grounds grounding facts is false. I construct a new theory.
The classic question about supererogatory acts—good deeds beyond the call of duty—is: why aren’t they obligatory? The classic answer: they are costly to the agent, who may give her own interests special weight. We argue that this “Cost View” faces counterexamples and mystifies the fact that supererogatory acts aren’t wrong. We then introduce the “Rights View,” which may give better explanations than the Cost View, but which must reckon with hard cases of its own.
I argue that standard versions of presentism, in combination with certain common views in metaphysics, such as a weak version of haecceitism, entail that both the future and the past are “open”. Whereas presentists typically embrace an open future, the idea that the past is open is highly counterintuitive. I argue that the fault here lies with presentism, rather than haecceitism and the other views, and I take this to constitute a novel objection to presentism. The conclusion of my argument is not simply that presentism is false, but also that presentism does not reflect our naive, commonsense picture of time.
I defend causation by agents from C. D. Broad’s influential argument against agent causation. Broad says that a cause must determine whether and when its effect occurs, and agents do not determine whether and when their purported effects occur, so they do not cause them. I argue that Broad’s argument is not sound. If it were sound, then it would pose an equal threat to event causation; for there is no sense in which events, but not agents, determine their purported effects. I show that Broad’s mistake has persisted through the subsequent literature: a number of more recent philosophers, including Donald Davidson, Carl Ginet, and Kadri Vihvelin, similarly demand from agent-causes a kind of nomic force or explanatory relevance that no event-cause enjoys.