Summary |
The chance facts appear to outrun what actually happens, and to involve constraints on what could happen, or on what will probably happen. Humeanism is – roughly – the thought that all such facts can be reduced to facts about what does in fact happen. The most influential characterisation of the view is David Lewis's thesis of Humean supervenience: that all matters of contingent fact supervene on the distribution of qualitative properties in space-time. In the case of chance, Lewis suggested that chanciness reduces to actual, occurrent patterns in the world. So for instance, according to a Humean, a coin's being fair reduces to some sort of fact about the actual ways in which that coin (or similar coins) fall. Non-Humeans complain that a Humean analysis of chance does not do justice to the modal character of chances. In addition to this central concern, there is a particularly acute objection raised against Humeanism: the so called "Bug". Chances, according to a Humean, are both grounded in patterns of actual events but also provide chances for events which would falsify those same chance facts. Consequently, it appears that we need to believe that such deviant futures have some chance of occurring, but also have no chance of occurring. Working out a bridge between chance-facts and what we should believe that avoids this contradiction is a central feature of the contemporary Humean project. Because non-Humeans are frequently propensity theorists, and because Humeans are frequently frequentists, these PhilPapers categories are also relevant. |