Full Program »
Causal Selection, Causal Significance, and Realism about Effect SizeEvents are typically the result of many factors. A fire may depend on the striking of a match, the presence of oxygen, etc. However, causal explanations typically cite only a few such factors as causes. The striking is cited as the cause of the fire, while the presence of oxygen is not. This phenomenon of preferring some causally relevant factors is known as ‘causal selection’. Among philosophers of science, a longstanding consensus holds that all causes are ontologically and hence objectively on par. This ‘causal parity’ thesis apparently follows from the fact that causes entail counterfactual dependence. The striking caused the fire because if the match hadn’t been struck, the fire wouldn’t have lit. But the same relationship holds between the oxygen and the fire. Proponents of causal parity conclude that causal selection must thus be arbitrary, pragmatic, or interest-relative. See for example Mill (1874) and Lewis (1973). In response to this view, a recent contingent of philosophers have sought to show that there are objective reasons for causal selection. Waters (2007), for example, sought to show that, for any given variable whose value varies in a given population, there is an objective fact about which causal variables are ‘actual difference makers’ and which are merely ‘potential difference makers’. If true, such an account vindicates the practice of causal selection. This paper accomplishes three goals. First, I provide evidence that causal explanations often posit a continuous ordering causes according to their ‘significance’. Thus, a simple binary ranking of variables, such as that between actual and potential difference-makers, cannot capture the full range of ordinary causal selection. For example, we currently believe that mosquitos are the most significant cause of Zika virus in humans, while male-to-partner sexual transmission is a less significant cause. But this is not the same distinction as that between actual and potential difference makers. A more nuanced account is required. Second, I argue that such significance judgements reflect the judged effect size of causes on a given effect within a population. If there is an objective fact of the matter about the relative effect size of some variable, then, contrary to the causal parity thesis, causal selection can be vindicated on objective grounds. Finally, I argue that there is indeed an objective fact of the matter about which causes are more significant than others within a population. In short, I claim that actual difference makers whose variation explains more of the variation in the effect are more significant. Defense of this position relies on finding an account of variance explained that is independent the set of variables entered into a model. I provide some initial support for this.
CDC. (2016). Retrieved July 01, 2016, from http://www.cdc.gov/zika/transmission/index.html
Lewis, D. (1973). Causation. The Journal of Philosophy.
Mill, J. S. (1874). A System of Logic (Eighth ed.). Harper & Brothers Publishers.
Waters, C. K. (2007). Causes that make a difference. The Journal of Philosophy, 104(11), 551-579.
Washington University in St. Louis