Causation, further themes

2. Conceptual versus empirical approaches

Most philosophical discussions of causation have tried to provide conceptual analyses of token causation. In other words, philosophers have sought to describe the conditions that a token causal statement must satisfy in order to count as true in any possible world. The methodology of these attempts is familiar, if not entirely free of controversy. Philosophers test the adequacy of suggested truth conditions against their intuitions about the applicability of causal statements in actual and possible situations. In this way, a conceptual analysis is arrived at by a priori means.

In contrast to this approach, some philosophers have advanced a rival – sometimes called the empirical approach – that seeks to understand token causation as it is constituted in the actual world, rather than in all possible worlds, and to arrive at this understanding by an a posteriori investigation of what physics tells us about causation. David Fair (1979) and Wesley Salmon (1984) were pioneers of this approach, both emphasizing the importance of the physical conservation laws in understanding the nature of causal relations in the actual world. However, perhaps the most developed theory along these lines is Phil Dowe’s (2000) conserved quantity theory of causation. Dowe’s theory essentially consists of two simple postulates: a causal process is the world-line of an object that possesses a conserved quantity; and a causal interaction is an intersection of world-lines that involves an exchange of a conserved quantity. A conserved quantity is any quantity that is governed by a conservation law: for example mass, energy, linear momentum, etc. These postulates imply the following empirical analysis of causation: two events are connected in a causal relation if and only if there is a continuous line of causal processes and interactions between them (see Conservation principles).

However, many questions can be raised about this account. It is, for example, doubtful whether the account provides sufficient conditions for the existence of a causal relation in the actual world, as there are many quantity-conserving processes between events that are causally independent: for instance, my tapping on my computer keyboard sends sound waves that impinge on the moving hands of the wall clock, but there is no causal interaction between my tapping and the movement of the clock hands. More generally, doubts can be raised about whether the empirical approach can be pursued independently of the conceptual approach. It does not make sense to ask whether some process found in the actual world is causal or not without having first settled on an answer to the question: ‘What conditions does something have to satisfy in order to count as causal within our conceptual framework?’.