Causation in Self-Management

In his thoughtful commentary, De Bruin invites us to say more about the notion of causation in our two-dimensional model of self-management in (mental) health care. In particular, he thinks there is a tension between 1) self-management-as-facilitation (e.g., long-term intentions) being causally efficacious and 2) “surgical” self-management interventions on specific variables being practically impossible in psychiatric conditions due to their complex dynamic nature. In particular, he asks us: “How can we establish the causal efficacy of these intentions if we are dealing with complex dynamic systems that do not allow for precise surgical interventions?” (p. 372). The assumption seems to be that one can only legitimately speak of causal efficacy when there is story to be told about how to establish the causal relation between the relevant variables. In response to de Bruin’s challenge, let us highlight two distinctions pertaining to causality that help to bring out our stance towards the role of the notion of causality in our model of self-management.

First, as foreseen by De Bruin, our immediate response would be to stress the difference between the epistemological question how to establish a causal relation in a particular case of self-management, and the ontological question how certain self-management interventions can be causally efficacious. De Bruin dismisses this rebuttal, stating that causation is a thoroughly practical notion. Embracing an interventionist concept of causation, he thinks that one cannot give a satisfying answer to the ontological question without giving an answer to the epistemological question: an account of causation entails a story about how causal relations are established in practice.

In response, we want to make a distinction between the practically and in principle discernibility of causal relations. Both can be fleshed out in broadly pragmatist, counterfactual terms (interventionism being one possible way). We would agree with De Bruin insofar as the notion of causal efficacy entails the in-principle discernibility of causal relations. But this does not imply that the legitimacy of using the concept of causation in clarifying the notion of self-management depends on the possibility of discerning the causal relations between variables in actual practice.

Despite De Bruin’s worries, we think that it is actually one of the strengths of our analysis that the conceptual distinction between management-as-control and management-as-facilitation can remain neutral about the notion of causation. As explained in section four of our article, the distinction hinges on 1) the degree of unpredictability of the behavior of the system one is trying to manage and 2) the context-(in)dependency or (in)directness of the impact of one’s interventions on the target outcome. Both unpredictability and context-dependency are notions that can be practically loaded in specific cases, without needing a substantive account of causation. It is precisely this feature, we believe, that makes it suitable to have real import in mental health care practice and to help service users gain a realistic understanding of the self-management possibilities in their own situation.

Second, we would like to invoke Fred Dretske’s distinction between structuring and triggering causes in order to highlight the kind of causal efficacy of long-term intentions as we envisage this in our model (Dretske, 1988). A structuring cause S of effect E is that which has put in place the right conditions under which trigger T produces E. For example, the electrician’s wiring the lightbulb to the light switch and the power source is the structuring cause for the fact that my flicking the switch causes the light to go on. Without the proper structuring cause, a given trigger will not yield the required result. We have depicted the causal role of longer term intentions as structuring causes for desired behavior. In many circumstances, these intentions should be not directly aimed at desired behavior, but at manipulating one’s environment in such a way that salient triggers will bring about desired behavior—quite possibly without much further conscious thought or effort. This is what the idea of facilitation entails. Thus, there are two ways in...