A New Approach to the Logical Theory of Actions and Causality

Abstract

In this article I try to show that the notions of game-tree and game in extensive form, as explained in Luce and Raiffa (1957), Chapter 3, provide an excellent starting point for the development of a theory of actions and causality that is formally rigorous to a high degree and, moreover, promises to be significant to various fields of application. For instance, the idea of an agent’s causing harm by his performance of a given act — important in Tort Law and Criminal Law as a condition for liability or responsibility — turns out to be interestingly analyzable and logically reconstructible within a theoretical framework built on the Luce and Raiffa conception, as I argue (Åqvist, 1973; see also the well known Hart and Honoré (1959) in this context). Now, let me list a few conditions of adequacy to be met by any acceptable theory in the present area, which seems still to be in a state of regrettable conceptual flux:

• (i) The theory should do justice to the important insight of von Wright (1963, 1966, and 1968), to the effect that the notion of action involves fundamentally that of change, transformations of states, or transitions among such.

• (ii) A clear distinction between individual acts and generic ones should be provided by the theory.

• (iii) I should lay down the conditions under which an agent can be said to perform, to omit, and ‘merely’ not to perform a given act (whether the individual or generic) in a given situation at a given time; as well as those under which an act is performable for an agent, or is an alternative open to him.

• (iv) Justice should be done by the theory to the important insight of a large number of writers who claim that a causal notion of an agent’s bringing something about, or making something happen, is a central ingredient in the concept of action. Among philosophers trying to develop and apply this idea systematically, we just mention Kanger (1957 and 1972), Chellas (1969), Pürn (1970), and Åqvist (1972), others not to be forgotten.

• (v) Justice should however also be done to the insight (we claim it to be one) that the above causal notion derives from a more fundamental one, which is of a ‘binary’ nature: that of an agent’s bringing something about by performing such and such an act. This binary conception has been dealt with predominantly by jurists (who have made impressive efforts to explicate it especially in the areas of Tort and Criminal Law), but seemingly hardly at all by philosophers (those working in this field of performatives and speech-acts certainly constitute an exception here, see Åqvist, 1972; another one may be Danto, 1965). So, our desideratum with regard to the notion under discussion is that the theory should enable us to analyze it in a satisfactory manner.