DUALISM VS. MATERIALISM I

Transcription

1 DUALISM VS. MATERIALISM I The Ontology of E. J. Lowe's Substance Dualism Alex Carruth, Philosophy, Durham Emergence Project, Durham, UNITED KINGDOM Sophie Gibb, Durham University, Durham, UNITED KINGDOM In the contemporary mental causation debate there is a general assumption that interactive substance dualism must be false in virtue of the supposed causal completeness of the physical domain, according to which every physical event has a sufficient physical cause. Interactive substance dualists typically respond to this problem by rejecting the causal completeness principle. However, E. J. Lowe s model of psychophysical causation instead offers a way of reconciling interactive substance dualism with the causal completeness principle. It does this by denying the homogeneity of the causal relata more specifically, by invoking a distinction between 'fact causation' and 'event causation'. According to Lowe, purely physical causation is event causation, the bringing about of particular effects. (For example, moving your arm in just this fashion as you take a drink). In contrast, psychophysical causation involves fact causation, the making it the case that an effect of a certain kind occurred. (For example, the fact that an event of the sort 'drinking' happened at all). Given this model of psychophysical causation, the dualist can therefore accept that every physical event has a sufficient physical cause as, according to it, mental events do not cause physical events. Rather, the mental is causally efficacious in virtue of causing certain facts to obtain, about which the causal completeness principle is silent. But Lowe s dualist model is only as plausible as the distinction between fact and event causation upon which it rests. In this paper it is argued that a suitable distinction between fact and event causation can only be maintained by someone who is also willing to accept something along the lines of the four-category ontology which Lowe defends a theoretical link which is neither obvious prima facie nor highlighted in Lowe's own work on either subject. The fact that accepting Lowe s model of psychophysical causation compels one to accept his four-category ontology may not have troubled Lowe. However, it makes his dualist account unattractive to those who wish to accept no more than one or two fundamental ontological categories in their ontology. Zombie Worlds and the Conditional Analysis Strategy Jussi Haukioja, Department of Philosophy and Religious Studies, NTNU Trondheim, Trondheim, NORWAY One strategy for responding to the zombie argument against materialism (ZA) is the Conditional Analysis Strategy (CAS) suggested, independently, by Braddon-Mitchell and Hawthorne. CAS claims that phenomenal concepts have a conditional structure, such that the following is a priori: (CA1) If the actual world contains spooky states, then our phenomenal concepts refer (rigidly) to those states. (CA2) If the world is merely physical, then our phenomenal concepts refer (rigidly) to physical

2 states that play the right sort of functional role. CAS explains why zombies are conceivable: we cannot know a priori whether the actual world is merely physical or not, and if it isn't, then there is no incoherence in the supposition that there are zombies. But it doesn't follow that zombies are possible: if the actual world is merely physical, zombies are not possible. Since we cannot know a priori which kind of a world we inhabit, we cannot infer the possibility of zombies from their conceivability, and ZA fails. In this paper I look at two objections against CAS. The first of these (Chalmers, Alter, Yetter- Chappell, Han) is that CAS delivers the wrong zombie intuition. The right zombie intuition, according to the critics, is that full zombie worlds are directly conceivable. But directly conceiving of a full zombie world would involve conceiving of myself as lacking phenomenal consciousness. The critics claim that zombie worlds are directly conceivable, because my knowledge of my own phenomenal states is a posteriori. I respond as follows. A posteriori evidence is needed to acquire phenomenal concepts. But once one has acquired them, one can know a priori that one has (had) phenomenal states: it is in the nature of phenomenal concepts that they are epistemically guaranteed to refer, though we do not know, merely on the basis of possessing the concepts, what kinds of states they refer to. Once I've acquired a phenomenal concept, I cannot coherently entertain the possibility that there are no phenomenal states. The second objection is by Yetter-Chappell: she claims that once the analysis is fleshed out in sufficient detail, it will turn out to be either circular, or committed to analytic functionalism. I will argue that even if she is right, this does not make CAS void of interest: it will merely suggest rethinking the dialectics around CAS. Yetter- Chappell points out herself that, for example, Papineau and Balog's constitutional theory of phenomenal concepts will also entail a conditional structure like (CA1) and (CA2), but at the metasemantic rather than the semantic level. The crucial thing to notice is that it doesn't matter for the CAS, as an answer to ZA, whether the conditional structure follows from the semantics or the meta-semantics. What matters is that (CA1) and (CA2) are a priori. If one thinks of meta-semantics as a priori, then one can use the CAS to give a quick and elegant response to the zombie argument, while denying that our phenomenal concepts strictly speaking have a conditional analysis. One-Dimensional versus Two-Dimensional Zombies Robert Michels, Department of philosophy, University of Geneva, Genève, SWITZERLAND One of the most important philosophical argument against materialism is Chalmers's Zombie Argument. (See Chalmers 2009.) The Zombie Argument is a conceivability argument, which means that it involves an inference from conceivability to metaphysical possibility. In classical conceivability arguments, this inference involves two premises (See e.g. Gendler Szabó & Hawthorne 2002): 1. p is conceivable. 2. If p is conceivable, p is metaphysically possible. The Zombie Argument instead relies on the framework of two-dimensional semantics and in particular on the distinction between 1- and 2-possibility, where 1-possibility is an epistemic and 2- possibility a metaphysical modality. The crucial inference from conceivability hence requires three instead of two premises:

3 1'. p is conceivable. 2'. If p is conceivable, p is 1-possible. 3'. If p is 1-possible, p is 2-possible. Classical conceivability arguments are widely rejected, since most philosophers adhere to the Kripkean view that conceivability does not entail metaphysical possibility. (See Kripke 1989.) In contrast, two-dimensional conceivability arguments such as the Zombie argument are taken seriously and there is an ongoing discussion about their soundness. This suggests that in the context of conceivability arguments, "going two-dimensional" amounts to a real improvement. In this talk, I will argue that this is not the case. One of two arguments for the same conclusion clearly improves on the other, if it is sound whereas the other argument is not. Assuming that both arguments are logically valid, this means that the premises of the improved argument have to be more easily made true than those of the argument it improves on. Drawing on the fact that premise 3' of the Zombie Argument can only be true if all concepts involved in p have identical 1- and 2-intensions (see Chalmers 2009, Walde 2005), I will argue that the inference from conceivability to 2- possibility in a two-dimensional conceivability argument is a sound sub-argument if, and only if, the corresponding inference in a classical conceivability argument for the same conclusion is, too. I draw the conclusion that twodimensional conceivability arguments and, in particular, the two-dimensional argument against materialism do not improve on their classical counterparts. After arguing for my main thesis, I will discuss the effects on Chalmers's modified Zombie Argument, which aims to establish the disjunctive conclusion that materialism is false or Russellian Monism is true, rather than just that materialism is false. Bibliography: Chalmers, David J. (2009). The Two-Dimensional Argument Against Materialism. In McLaughlin, B. P. & Walter, S. (eds.), Oxford Handbook to the Philosophy of Mind. Oxford University Press. Gendler Szabó, Tamar & Hawthorne, John (2002). Introduction: Conceivability and possibility. In Szabó Gendler, T. & Hawthorne, J. (eds.), Conceivability and Possibility. Oxford University Press. Kripke, Saul A. (1980). Naming and Necessity. Harvard University Press. Walde, Bettina (2005). On epistemic and ontological aspects of consciousness: Modal arguments and their possible implications. Mind and Matter 3 (2): Fundamentality Physicalism and the Intelligibility Constraint Gabriel Rabin, Philosophy, ANU; New York University in Abu Dhabi, Canberra, AUSTRALIA Physicalism says that the fundamental features of our world are purely physical. At rock bottom, it s just bosons, quarks, gravity, and the like. A prominent line of argument due to David Chalmers uses the famous explanatory gap to challenge physicalism. If physical fundamentalia generate consciousness, there must be an intelligible epistemic route from the physical to consciousness. There is no such route. Therefore, there is more to the fundamental than the physical. The intelligibility constraint says that if the fundamental level is composed purely of Fs, then it must be intelligible how the Fs could generate the rest of reality: the cars, coffees, koalas, and

4 consciousness. In my talk, I discuss this constraint in detail. But for now let s focus on Chalmers argument for the impossibility of an intelligible route from the physical to consciousness. (P1) Intelligibility is a priori scrutability: If there is an intelligible route from the Fs to the Ms, then the conditional If F then M is knowable a priori. (Where F and M are complete descriptions of the Fs and Ms). (P2) No a priori route: The conditional If P then C is not knowable a priori, where P and C completely describe the physical and consciousness facts. (C) Therefore: There is no intelligible route from the physical to consciousness. Chalmers master argument against physicalism, of which the conceivability argument is one version, depends crucially on this reasoning. I deny both premises. Against (P1). Intelligibility should not be cashed out via a priori scrutability. In your armchair, you have been told that the Fs generate the Xs. You a priori reason long and hard, but you just don t see it. How could the Xs arise from simple interactions between the Fs? You are then released from the armchair. You empirically interact with the Xs and the Fs. You watch them, touch them, manipulate them, and observe their behavior. With your increased understanding, you finally get it. That s how it goes! How the Fs yield the Xs is now intelligible to you. But you could not have seen this from the armchair. Sometimes, intelligibility is not a priori. Against (P2). Our inability to a priori derive consciousness from physical truths does not entail (P2). The a priori in (P2) is highly idealized. If a futuristic scientist with enhanced brain capacity, space alien, or god, can do the a priori reasoning, that suffices. (P2) requires a lack of a priori route, and an explanatory gap, at the limit of inquiry, not just today. Chalmers structure and function argument takes up the challenge. Physical truths are structuralfunctional truths. Consciousness truths are not. Structural-functional truths a priori entail only more structural-functional truths. Therefore, physical truths can never a priori entail consciousness truths. I argue that there is no precisification of structural-functional truth such that it is simultaneously plausible that (a) physical truths are structural-functional truths and (b) structural-functional truths are incapable, even in principle, of a priori entailing consciousness truths. But I ll save that argument for Helsinki. Increased coincidence detection for quantum versus pseudo random generated numbers Lieze Boshoff, Psychology, University of Groningen, Groningen, NETHERLANDS Jacob Jolij, Psychology, University of Groningen, Groningen, NETHERLANDS People often see meaning in stimuli that are typically considered meaningless. According to Von Lucadou's idea of Generalized Quantum Theory (GQT), such perceived coincidences, or examples of synchronicity, may be the result of entanglement between a conscious observer and the physical world. Here we test this idea by means of a 'coincidence detection task'. Participants were shown a series of 150 randomly generated stimuli (here, a set of three numbers between 0 and 255), and had to indicate per stimulus whether that set of numbers held any meaning to them. Critically, the random source could either be a quantum event generator, or a pseudo random number generator. If 'coincidences' are indeed the result of quantum entanglement, or a process

5 like it, we would expect an increased number of reported coincidences for quantum versus pseudo-random generated events. A preliminary data analysis shows that this indeed is the case: over 150 random stimuli, participants (N=49) reported an average of 9% being meaningful (eg., birthday, telephone number) for pseudo-randomly generated numbers; this increased significantly to 12% for quantum generated random numbers. This result seems to suggest that synchronicities may indeed be interpreted as non-local correlations between an observer and the physical world.