Rosenberg on Causation

This paper is an explication and critique of a new theory of causation found in part II of Gregg Rosenberg's A Place for Consciousness. According to Rosenberg's Theory of Causal significance, causation constrains indeterminate possibilities, and according to his Carrier Theory, physical properties are dispositions which have phenomenal properties as their causal bases. This author finds Rosenberg's metaphysics excessively speculative, with disappointing implications for the place of consciousness in the natural world. Rosenberg’s theory of causation is novel, complex and challenging. I can’t say that I have mastered its intricacies, and I admit that much of my puzzlement may be failure of understanding. I also cannot discuss all parts of the view here and will inevitably neglect some important aspects. For example, I will not discuss his arguments against Humean theories, or his theories of space and time. However, I will explain what I take to be the major elements his account of causation as I understand them, highlight some features and raise some questions. Overall, while I appreciate the effort to give consciousness a place in the natural world, I don’t think that Rosenberg succeeds in providing a metaphysical foundation for a philosophy of mind that we want to have. In Part II of A Place for Consciousness, Rosenberg considers the following argument for Epiphenomenalism: 1. The physical facts alone do not entail the facts about conscious experience. PSYCHE: http://psyche.cs.monash.edu.au/ PSYCHE 2006: VOLUME 12 ISSUE 5 2 2. We can conclude from (1) that 2' Experience is a nonphysical aspect of the world. 3. A completed physical theory is, in principle, a descriptively adequate characterization of the dynamical evolution of the physical world. 4. We can conclude from (3) that 4' Our physical explanations are complete explanations of the causation involved in producing bodily movements. 5. We can conclude from (2') and (4') that 5' Consciousness lies outside the causal structure of the world; it is an irrelevant epiphenomenon (130). If we don’t want to be epiphenomalists, we best dispense with one of these claims. Rosenberg explores the possibility of denying premise (4). Doing so amounts to saying that the fact that a complete physical theory would explain causation in the physical world does not entail it would completely explain the causes of human action. But that is to say that some of the causes of human action are not part of the physical world, and yet the laws of a complete physical theory are never violated. How could that be? Rosenberg attempts to develop a new theory of causation that will provide a way. It has two major components, the Theory of Causal Significance and the Carrier Theory of Causation. I will consider each of these in turn. 1. The Theory of Causal Significance 1.1. The Determination Problem Traditional theories of causation focus on causal responsibility: How is it that some entity (event, state, or object) is responsible for the production of some effect? Why do changes occur? These questions presuppose a world of determinate entities and explore the relations between them. Rosenberg’s theory of causal significance starts with different questions: Out of the many ways the world could be, why is it this way? What constrains the way things are? For Rosenberg, “causation is about constraint on a space of possibilities” (158). Causal production, which most accounts have focused on, is but one kind of causal significance. According to Rosenberg, the world’s individuals have many potential states. To be actualized and not merely potential, an individual must take on one and only one potential state. But how does a determinate world emerge from indeterminate potentials? This is the determination problem, which can only be solved by a theory of causation, according to Rosenberg (158). 1.2. Effective and receptive properties Rosenberg begins his attempt to solve the determination problem by distinguishing two types of properties, effective and receptive. Effective properties of an individual give it capacities to constrain the states of other individuals. By experimental investigation, science can reveal the patterns of constraints some individuals place on others and thus discover effective properties. However, a property of an individual may be effective only if some individual is receptive to the property’s presence, hence the need for receptive properties (154). PSYCHE: http://psyche.cs.monash.edu.au/ MCKITRICK: ROSENBERG ON CAUSATION 3 Receptive properties are akin to the passive powers discussed in Aristotle (Metaphysics 9.1, 1046a11-13), Locke (1959: 116) and Harre (1970: 87). Rosenberg characterizes them as powers to be effected. However, on Rosenberg’s view, a receptive property is a connection between individual properties, enabling them to become members of a causal nexus and to be sensitive to constraints (159). Rosenberg likens receptive properties to the plastic rings on a six-pack, where the cans (effective properties) plug into the available slots. So, even though the rationale for receptive properties was that they must be had by individuals in order for other properties to have effects on them, on Rosenberg’s view, primitive individuals don’t have receptive properties directly. Instead, they are related to other individuals via a receptive connection, whereby they gain the ability to be constrained by other properties. The complex individual constituted by the primitive individuals instantiates the receptive property (165). Rosenberg’s primitive (level zero) individuals are the most primitive effective and receptive properties. He offers mass, charge, and spin as an illustration of possible primitive effective properties. A receptive connection binds together primitive effective properties. Since they are individuals, primitive properties must be something like property instances rather than universals. Furthermore, Rosenberg says “Individuals themselves are pure property complexes (i.e., there are no enduring substances)” (170). So, he seems to hold a bundle theory of particulars. Effective properties have incomplete natures which seek to bind with individuals possessing a complementary kind of property to be complete (165). Effective properties, being indeterminate, are determinables, and their completion, through binding in a receptive connection, makes them (more) determinate: “an incomplete effective property ... is an abstract entity that contains a propensity to become one of its determinates” (167). So, a primitive individual, say charge, is an indeterminate determinable. It could be either of its determinates, positive, negative, or neutral. The matter of which specific charge it will be is settled through being bound to some other receptive property via a receptive connection: “Effective properties are potential unless actually receptively bound” (246).