Agency and Asymmetry
Research Proposal, Sydney University Philosophy PhD Program, 2003
Brad Weslake
In any case, since there was no seeing which way the other man faced, right and left meant nothing.
(Italo Calvino, 1963, p. 62)
We take the orientation of causation, like the orientation of time, for granted. Whether we conceive of time running past us like a river, or of us flying through time like an arrow, it is standardly taken that time has a direction. Similarly for causation; whatever mysterious powers causation consists in, it is standardly taken that these powers push in one way, and one way only. However there is nothing in the fundamental laws of physics that provides this orientation; nothing that tells us that one end (timewise) of the universe is objectively the beginning, and the other the end (Price, 1995). Likewise for the direction of causation; nothing in physics tells us the objective difference between cause and effect (Price, 1996b, pp. 132-134). The symmetry of physics with respect to causation is reflected in the symmetry of our evidence for causation in our everyday experience—Gasking (1980) illustrates this using the example of a metal which simultaneously reaches a particular temperature and changes state between a solid and a liquid. Each change in the properties of the metal necessitates the other, and yet they are simultaneous, and so no temporal asymmetry can provide the asymmetry of causation we intuitively require.
The research I propose to undertake is the development of the agency account of causation advocated by Price (1992a; 1996b, esp. Chap. 6-7)[1]. In particular, I propose to develop the idea that our experienced causal asymmetry can be explained by the fact that we, as agents, depend on the thermodynamic asymmetry of the universe (a strategy suggested by Price, 1994, n. 24; Price, 1992b). A high-level overview of the dependence of human life on thermodynamic asymmetry is given in Schrödinger (1944, Chapter 6, "Order, Disorder and Entropy", pp. 67-75) and a more detailed account is provided, for example, by the interactivist model of Hooker and others (see Collier and Hooker, 1999, for the conceptual foundations and references to additional papers). Two specific advantages of the agency account over the classical conventionalist account of Hume are that the difference between cause and effect becomes more than "merely a linguistic matter" (Price, 1996b, p. 137), and that it doesn’t logically exclude backwards causation. One of the flow-on benefits of the account, then, is that it opens the way for an interpretation of quantum mechanics that exploits backward causation to remain both realist and consistent with the causal localism of special relativity (Price, 1994; 1996a)[2].
An agency account of causation needs to establish itself on two fronts. Firstly, it must be demonstrably superior to the prevailing philosophical views on the orientation of causation, which are predominantly based on the fork asymmetry[3]. The fork asymmetry is, roughly, the feature of the universe whereby non-causally-related correlated events can nearly always be explained in terms of an earlier common cause rather than a later common effect; visualised as a directed graph showing causal relationships, the structure is a tree; and the orientation of this tree is said to ground the orientation of causation. Problems for this account of the direction of causation appear at two levels. Firstly, the fork asymmetry only occurs at the macroscopic level, and so it appears to be unable to account for causation in microphysics (Price, 1992a). Secondly, the fork asymmetry is unable to distinguish between cause and effect in the macroscopic world in all cases: for example, in cases of simultaneous causation such as in the case of the melting (or solidifying) metal already discussed (Gasking, 1980, p. 143); and in a range of hypothetical cases where individual causes and effects are isolated (Price, 1996b, pp. 140-142). The problems here stem from the fact that the fork asymmetry manifests itself macroscopically and statistically, being itself a corollary of thermodynamic asymmetry—and so it does not cover every imaginable case that we would intuitively wish to describe in causal terms. The agency account turns this disadvantage into an advantage, by explicating our (macroscopic, thermodynamic-dependent) agency in a way that draws on the advantages of the fork asymmetry, while avoiding the disadvantages by making the causal links a product of the projection of our own asymmetry rather than an objective feature of the world.
In addition to criticising competing accounts, an agency account of causation will need to defend itself against the type of general objections that are typically made against perspectival accounts (that is, accounts that take causation to be a product of our own situation as humans rather than an objective feature of the universe). Against the more common objections the account must (Menzies and Price, 1993; Woodward, 2001):
- Defend against the accusation that the account confuses epistemology with metaphysics.
- Provide an account of agency that is non-circular (that is, does not presuppose causation).
- Provide an account of the manner in which the projection occurs to events which are outside the (possible) control of any agent.
- Defend the anthropocentrism of the model.
With regard to 2. and 3., one promising line of investigation appears to be examining how the concept of causation actually emerges in the developmental stages of humans. The potential for a psychological approach to causation was explored for example by Piaget (Wildgen, 2000)—a more recent example is found in Lakoff and Johnson (1999), who propose a cognitive model whereby causation is a concept that emerges from more basic concepts rooted in human action. Lakoff and Johnson propose that the basic understanding of causation, where an agent controls an object, is projected metaphorically to abstract domains[4]. These types of model have the potential to inform a defence of an agency account against the charge of circularity, and also inform how the projection to remote events can emerge from the experience of immediate agency.
The agency approach to causation may also have implications for cosmology—for instance, it might turn out that an agency account is uniquely placed to give a satisfactory answer to the first-cause argument and/or related cosmological arguments. Whether or not this turns out to be the case, the agency account has not been adequately defended in the literature on causation, and so it is to this cause that I intend to direct my research.
Footnotes
[1] Early accounts along these lines are given in, for example, Collingwood (1948), Gasking (1955) and Ramsey (1978).
[2] Price (1991), suggests two further benefits may be firstly (the logical possibility of) instantaneous action at a distance, and secondly the possibility of a causal foundation of time—though the position argued in Price (1996b) is that the orientation of both time and causation are perspectival in nature, and therefore it would be nonsensical to attempt to ground one in the other (ie. it would need to be a substantially revised position that took up the latter possibility).
[3] The most influential early account is that of Reichenbach (1956).
[4] The parallels here between the account of Lakoff and Johnson and the agency theory are perhaps brought out by the fact that the agency theory is also sometimes referred to in the literature as the manipulability theory (Price, 1996b, p. 157; Woodward, 2001).
Bibliography
Calvino, Italo. 1963. Marcovaldo, William Weaver (Trans.), Minerva, London, 1993.
Collier, J. D. and Hooker, C. A. 1999. "Complexly Organised Dynamical Systems", in Open Systems and Information Dynamics, Vol. 6, Iss. 3, 1999, pp. 241-302. Online: http://www.newcastle.edu.au/centre/casrg/publications/Cods.pdf (Accessed 18 October 2002).
Collingwood, R. G. 1940. An Essay in Metaphysics, Oxford University Press, Oxford, 1940.
Gasking, Douglas. 1955. "Causation and Recipes", in Mind, Vol. 64, pp. 479-487.
Gasking, Douglas. 1980. "Causation and Action", in I. T. Oakley and L. J. O’Neill (Eds.), Language, Logic and Causation: Philosophical Writings of Douglas Gasking, Melbourne University Press, Melbourne, 1996, pp. 132-143.
Lakoff, George and Johnson, Mark. 1999. Philosophy in the Flesh: The Embodied Mind and Its Challenge to Western Thought, Basic Books, New York.
Menzies, Peter and Price, Huw. 1993. "Causation as a Secondary Quality", in British Journal for the Philosophy of Science, Vol. 44., pp. 187-203.
Price, Huw. 1991. "Agency and Probabilistic Causality", in British Journal for the Philosophy of Science Vol. 42, pp. 157-176.
Price, Huw. 1992a. "Agency and Causal Asymmetry", in Mind, Vol. 101, pp. 501-520.
Price, Huw. 1992b. "The Direction of Causation: Ramsey’s Ultimate Contingency", in David Hull, Micky Forbes and Kathleen Okruhlik (Eds.), PSA 1992: Volume 2, Philosophy of Science Association, East Lansing MI, pp. 253-67.
Price, Huw. 1994. "A Neglected Route to Realism About Quantum Mechanics", in Mind, Vol. 103, Num. 411, July 1994, pp. 303-336.
Price, Huw. 1995. "Cosmology, Time’s Arrow and That Old Double Standard", in Savitt, S. (Ed.), Time’s Arrows Today, Cambridge University Press, 1995, pp. 66-94.
Price, Huw. 1996a. "Backward Causation and the Direction of Causal Processes: reply to Dowe", in Mind, Vol. 105, Num. 419, July 1996, pp. 467-474.
Price, Huw. 1996b. Time’s Arrow and Archimedes’ Point, Oxford University Press, Oxford, 1996.
Ramsey, F. P. 1978. "General Propositions and Causality", in Foundations: Essays in Philosophy, Logic, Mathematics and Economics, D. H. Mellor (Ed.), Routledge and Kegan Paul, London, 1978, pp. 133-151.
Reichenbach, H. 1956. The Direction of Time, University of California Press, Berkeley CA, 1956.
Schrödinger, Erwin. 1944. "What is Life?: The Physical Aspect of the Living Cell", in his What is Life? with Mind and Matter and Autobiographical Sketches, Cambridge University Press, Cambridge, 1967, pp. ix-90.
Wildgen, Wolfgang. 2000. "Causation, Control and Evolutionary Semiotics", paper presented at Structures of Causal Meaning, Center for Semiotics, University of Aarhus, January 27-29 2000. Online: http://www.hum.au.dk/semiotics/docs/epub/WS2k/wild/Caus&Evolution.html (Accessed 22 October 2002).
Woodward, James. 2001. "Causation and Manipulability", in The Stanford Encyclopedia of Philosophy, Edward N. Zalta (Ed.), Fall 2002 Edition. Online: http://plato.stanford.edu/archives/fall2002/entries/causation-mani/ (Accessed 23 October 2002).