Do Biopsychosocial Causal Models Rule Out Physicalism?

Donald Price and James Barrell, two eminent pain researchers, argue that every time an experiment demonstrates that a biological variable is causally relevant to a psychological outcome, we are entitled to further rule out a possible mind-brain identity or supervenience relationship. As experimental evidence for two-way causal links between biological and psychological variables accumulates, more and more identity and supervenience relationships are ruled out, suggesting that psychophysicalism is empirically better supported than physicalism. I raise an objection to this line of argumentation by pointing out that, in the studies in question, causation is established within an operationalized framework; that is, irrespective of whether one knows what exactly is being manipulated and measured and how the intervention and measurement techniques work. By itself, evidence for causal relevance doesn't demonstrate that to each manipulated variable corresponds an ontologically distinct cause. This opens the possibility that some variables share common referents, as postulated by physicalist accounts. Moreover, even if it is not clear how to test for identity or supervenience relationships, it is still possible to test for causal mediation, which can generate empirical evidence discriminating between reductive physicalism and non-reductive alternatives.