The Mental as Fundamental: New Perspectives on Panpsychism 9783110319859, 9783110319330

Citation preview

Michael Blamauer (Ed.) The Mental as Fundamental New Perspectives on Panpsychism

Michael Blamauer (Ed.)

The Mental as Fundamental New Perspectives on Panpsychism

Bibliographic information published by Deutsche Nationalbibliothek The Deutsche Nastionalbibliothek lists this publication in the Deutsche Nationalbibliographie; detailed bibliographic data is available in the Internet at http://dnb.ddb.de

North and South America by Transaction Books Rutgers University Piscataway, NJ 08854-8042 [email protected] United Kingdom, Ire, Iceland, Turkey, Malta, Portugal by Gazelle Books Services Limited White Cross Mills Hightown LANCASTER, LA1 4XS [email protected]

Livraison pour la France et la Belgique: Librairie Philosophique J.Vrin 6, place de la Sorbonne ; F-75005 PARIS Tel. +33 (0)1 43 54 03 47 ; Fax +33 (0)1 43 54 48 18 www.vrin.fr

2011 ontos verlag P.O. Box 15 41, D-63133 Heusenstamm www.ontosverlag.com ISBN 978-3-86838-114-6 2011 No part of this book may be reproduced, stored in retrieval systems or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use of the purchaser of the work Printed on acid-free paper ISO-Norm 970-6 FSC-certified (Forest Stewardship Council) This hardcover binding meets the International Library standard Printed in Germany by CPI buch bücher.de

Table of Contents Michael Blamauer Introduction: The Mental as Fundamental

7

1. Godehard Brüntrup Panpsychism and Structural Realism

15

2. Pierfrancesco Basile The Last Man Standing Argument for panpsychism: A rejoinder 35 3. David Skrbina The Man Still Stands: Reply to Basile

53

4. Ludwig Jaskolla “Mind Matters...” – Towards a concept of proto-mental causation

57

5. Riccardo Manzotti The Spread Mind: Phenomenal Process-Oriented Vehicle Externalism

79

6. Michael Blamauer Taking the Hard Problem of Consciousness Seriously: Dualism, Panpsychism and the Origin of the Combination Problem 99 7. David Skrbina Mind Space: Toward a solution to the combination problem

117

8. Philip Goff There is no combination problem

131

9. Freya Mathews Panpsychism as Paradigm

141

10. Matthias Rugel / Benjamin Andrae Not bound to feel everything? A dialog on the scope of experience

157

Introduction: The Mental as Fundamental Michael Blamauer (Vienna) In recent years one could notice a revival of panpsychistic considerations of the mind’s place in nature.1 In his already classic essay, Panpsychism, Thomas Nagel defined this position as “the view that the basic physical constituents of the universe have mental properties, whether or not they are parts of living organisms” (Nagel 1979, 181). Thus, mental properties are considered to be ubiquitous. This of course stands in opposition to our common-sense view of the world, which strictly separates things that are deemed conscious from those deemed not conscious. Yet even in the field of philosophical reflection, panpsychism is often regarded with skepticism and for some philosophers, such as Colin McGinn, it is merely a “comforting piece of utter balderdash” (McGinn 2006, 93). However, despite, or perhaps precisely due to this skepticism, panpsychists supply rather strong arguments in defense of their position. The essays assembled in the present collection2 aim to contribute positively to this discussion by providing new perspectives on panpsychism, shedding new light on the arguments, problems, and impacts of this view. For the purpose of an introduction, I will attempt to offer some reflections on the title of this book: “The Mental as Fundamental – New Perspectives on Panpsychism.” Since it suggests a special kind of association between the concepts of mentality, fundamentality and panpsychism, it may be helpful to address this association by first saying something about each concept. In this context, mentality refers to consciousness considered as the dimension of subjective experience. This dimension of experience, manifest as phenomenal qualia, has been the subject of intense debate in philosophy of mind over the last fifteen years. Whereas reductive materialists have claimed that an exhaustive explanation of all-there-is (including conscious experience) can be provided in purely physical terms, and thus that the mental is really nothing 1

See e.g. the collections of Freeman 2006 and Skrbina 2009. The initial idea for this collection goes back to a workshop that took place in May 2010 at the Department of Philosophy of the University of Vienna. 2

8 more than a special form of the physical, dualists have argued for the opposite view. They hold that reality is not exhaustively accounted for through purely physical descriptions in that these cannot explain the qualitative way we experience this reality: Experience (with all its phenomenal features) is not entailed by the physical facts about the experienced reality. Thus, dualists commonly hold that there are two fundamental aspects of reality, the physical and the mental, by rejecting the possibility of reduction of either one to the other. The most striking argument3 provided by dualists for the ontological independence of mental properties from physical properties is their free variability over the physical, which means – very briefly – that a change of mental properties does not entail a change of physical properties and vice versa.4 Yet if phenomenal facts do not consist in physical facts, then phenomenal facts must be considered fundamental and materialism is false – or so they argue. Now, what is meant by fundamentality? David Chalmers provides a definition of fundamentality in The Conscious Mind: “Fundamental features cannot be explained in terms of more basic features, and fundamental laws cannot be explained in terms of more basic laws; they must simply be taken as primitive” (Chalmers 1996, 126). This definition at first suggests an epistemological reading of the concept of fundamentality. However, questions concerning the relationship of mind and body and the possibility of reduction are primarily questions of ontological dependence. Moreover, within the scope of this discussion between reductionists and non-reductionists, the paradigmatic interpretation of a fundamental feature strongly relates to the definition of fundamental physical features of elementary particles thus far discovered by physics (e.g. mass, spin, and charge). Yet these physical features are not merely epistemologically crude – they have ontological relevance in the first place by virtue of being constitutive in the make-up of other physical objects. Hence, whereas the epistemological concept of fundamentality (as I understand it) proceeds only from the impossibility of conceptual reduction, the ontological concept of fundamentality additionally postulates the ontological priority, and hence constitutional relevance, of certain properties of our world. Thus, if the mental is to 3

An excellent synopsis of several arguments for the irreducibility of consciousness can be found in Chalmers 1996. 4 For an outline, especially of Chalmers’s version of the argument, see Brüntrup, in this volume.

9 be considered fundamental in the same sense as the physical, then it must be understood in the stronger ontological sense of fundamentality. This last consideration actually brings us to the third and final question: Why panpsychism? William Seager states: “To show panpsychism we need to show that consciousness is both fundamental and ubiquitous” (Seager 2006, 137). I have just provided some preliminary cues concerning the fundamentality claim – which is explored in much more detail in the following contributions – so that we can ask: What reasons do we have to proceed from the assumption of the fundamentality of the mental to the assumption of its ubiquity? This question has received various answers. I present one here in a very abbreviated form. Fundamental mental properties are considered fundamental along the same lines as fundamental physical properties. Paradigmatic fundamental physical properties are – as I see it – ontologically fundamental because they constitute physical objects. And, in that every physical object supervenes on these fundamental properties, they are ubiquitous as well. The conclusion to be drawn from this definition of fundamentality of fundamental physical features with respect to the fundamentality of the mental as argued by nonreductive theorists of consciousness is obvious: If the mental is to be understood as fundamental in this sense, then it is not ontologically dependent on something more fundamental. Fundamental mental features are on par with fundamental physical features. As Amy Kind rightly put it: “Once you claim that the world contains fundamental features that are non-physical it is hard to find a principle way of limiting exactly where those fundamental features are found” (Kind 2006, 2). Thus, the arguments for the fundamentality of consciousness push in the direction of panpsychism.5 Following this line of reasoning, mental properties are best considered as ubiquitous too. However, given the fundamentality of consciousness, a range of arguments exists in favor of panpsychism, some of which are discussed in the present volume. In the opening article, “Panpsychism and Structural Realism,” Godehard Brüntrup explores a classical argument for panpsychism: the argument from intrinsic natures. With regard to Chalmers’s 5

See also my contribution to this book.

10 formulation of the hard problem of consciousness, Brüntrup argues for panpsychism as a viable option of accounting for consciousness as a fundamental feature of reality. By rejecting structural realism as an adequate scientific framework with which to describe reality, he points to the necessity of intrinsic properties. He then argues that phenomenal qualities would serve perfectly as the intrinsic carriers of the relational properties of objects, as well as the criteria of their individuation. In the next article, Pierfrancesco Basile critically examines The Last Man Standing Argument for Panpsychism, which was originally put forward by David R. Griffin and reconsidered by David Skrbina in his Panpsychism in the West (Skrbina 2005). The Last Man Standing Argument is a negative argument that derives its force from the rejection of alternative positions in order to cope with a certain problem. Basile challenges it by asking: “Could it be that, contrary to what is held by Griffin and Skrbina, panpsychism is not ‘the most viable alternative’ [to the problem of consciousness, MB] after all?” He thereby trades off the positive aspects of panpsychism against its problematic ones, especially the so-called “combination problem,” and asks for alternatives. He concludes that even if panpsychism must overcome significant theoretical challenges, it is still a position worthy of exploration with respect to the mind-body problem. Basile’s contribution is followed by a reply by David Skrbina. Another argument for panpsychism could be distilled through examining the problems of mental causation. In “Mind Matters...” Ludwig Jaskolla provides an idea of how proto-mental causation could be positively treated within an overall pan-experientialistic ontology. Starting with an evaluation of the classical problems of mental causation in order to gain some basic conceptual distinctions, he supplies reasons for the causal relevance of the mental in that “it provides the formal constraints necessary for every act of causation.” He then argues that a panexperientialistic metaphysics would serve as the best framework to positively accommodate the causal relevance of the mental. In “The Spread Mind,” Riccardo Manzotti focuses on a critical discussion of the internalism of experience and puts forward an externalist process-orientated position, which he refers to as “the spread mind.” Whereas internalism claims that phenomenal experience occurs within our nervous systems, externalism holds that experience is “identical to the occurrence of a process

11 instantiated by causal relations in the environment.” The processorientated externalism put forward by Manzotti is a kind of panpsychistic approach in that it proceeds beyond the borders of the human body as the place wherein phenomenal experience occurs in order to relate experience to universal physical processes in the environment. In my own paper, “Taking the Hard Problem of Consciousness Seriously,” I attempt to show that the most effective arguments against physicalism, which at first sight tend to lead to a kind of dualistic world view, in their last consequence actually lead to panpsychism. But I further argue that even if panpsychsim is a sound way of treating the mind-body problem, the challenge of the combination problem is a significant hurdle for panpsychism, which must overcome if the latter is to made intelligible. The combination problem of panpsychism is that of intelligibly explaining consciousness in terms of adding single perspectives or experiences to gain a composite conscious whole, i.e. a full-blown, human, experiential perspective. And this may be one of the major problems of formulating a theory of consciousness. I attempt to show that the origin of the combination problem can be traced back to the dualistic concessions of panpsychism, which emerged from its arguments against physicalism in the first place. I conclude that to overcome the combination problem, panpsychism must abandon these dualistic concessions. The next two papers offer definite solutions to the combination problem. David Skrbina presents an answer in “Mind Space.” He argues that, parallel to physical space (in which physical objects exist), there exists an experiential space that he calls “mind space,” which is constituted by experiential aspects of the ultimate particles that are likewise resembled in physical space. The idea is that this larger experiential space is not “composed of” lower experiential units, but rather that “they combine to create a space of experiential possibilities for the whole, allowing the existence of complex, higher-order mind.” Yet whereas most panpsychists find themselves facing a formidable theoretical obstacle in the combination problem, Philip Goff tries to untangle this Gordian Knot by simply cutting through it with a bold stroke. In his paper of the same name, he argues that “There is no Combination Problem” as such if the panpsychist is willing to accept the idea of emergence. The problem of emergence of macroscopic

12 subjects from their microscopic parts is not – as he states – the combination problem, but the problem of defining a precise scientific law that governs this emergence. Yet, if the definition of the laws of emergence is scientifically questionable in that they would be highly arbitrary, the default assumption “ought to be that all combinations of fundamental particles form an emergent macroscopic subject.” In “Panpsychism as Paradigm,” Freya Mathews primarily deals not with particular questions of the mind-body problem, but rather with panpsychism as a viable metaphysical theory to explain the world as such and our place within it. She develops a holistic view of the universe as a self-actualizing “non-contingent unity” and explores the question of how “local, individual minds come to differentiate themselves within the matrix of a global mind.” The book closes with an article by Matthias Rugel and Benjamin Andrae. They set up a kind of platonic dialog on panpsychism between two characters, M and B. Rugel and Andrae introduce panpsychism by examining the question of the interrelationship between things and the concept of experience as the paradigmatic model of connection. B then vindicates the position “that every individual is connected to all other individuals,” thereby following Aristotle, Leibniz and Whitehead. In opposition, M argues that “every individual is only connected to some and definitely not all other individuals,” following Gregg Rosenberg’s theory of natural individuals. All of the contributions to this book – even those discussing theoretic hurdles still to be overcome – point in one direction: Panpsychism is a promising (and perhaps the most promising) position from which to treat the mind-body problem, not only with respect to the human species, but with respect to the evolution of the universe as such. Only time will tell. References Chalmers, D. J. (1996): The Conscious Mind: In Search of a Fundamental Theory, Oxford et al. Freeman, A. (Ed.) (2006): Consciousness and ist Place in Nature. Does Physicalism entail Panpsychism?, Exeter u. a. Kind, A. (2006): “Panexperientialism, Cognition and the Nature of Experience.” In: Psyche 12 (5): 1-15.

13 McGinn, C. 2006. Hard Questions – Comments on Galen Strawson. In: Freeman, A. (Ed.), Consciousness and ist Place in Nature. Does Physicalism entail Panpsychism?, Exeter u. a. Nagel, T. (1979): Panpsychism. In: Idem, Mortal Questions, Cambridge. Seager, W. (2006): The 'Intrinsic Nature' Argument for Panpsychism. In: Freeman, A. (Ed.), Consciousness and ist Place in Nature. Does Physicalism entail Panpsychism?, Exeter u. a. Skrbina, D. (2005): Panpsychism in the West. Cambridge/Mass.: MIT Press. Skrbina, D. (Ed.) (2009): Mind that Abides. Panpsychism in the new millenium. Amsterdam/Philadelphia: John Benjamins.

Panpsychism and Structural Realism1 Godehard Brüntrup (Munich) The Problem Structural realism is a popular view among philosophers of science. Definitely with the late David Lewis' paper on “Ramseyan Humility” (Lewis 2009), if not earlier, it has become widely discussed among analytic metaphysicians as well. It promises to avoid the pitfalls of both classical scientific realism and scientific anti-realism by restricting realism to the structural features of the world only. In the tradition of Whiteheadian process philosophy, any form of structural realism, however, commits the fallacy of misplaced concreteness. Structure alone will never suffice to ground the existence of a concrete entity. Intrinsic properties that carry the relational structure are needed. This line of argument lends support to a kind of panpsychism if the grounding realizers of the structural and relational properties are conceived to be intrinisic properties analogous to the properties of the phenomenal mind. To the structural realist, it is by no means obvious that he is committing a fallacy of misplaced concreteness. On the contrary, the need for unobservable realizers of the mathematical structures described by science is often flatly denied. If – by assumption – all there is to matter is its relational or structural properties, then the impetus to seek an intrinsic ‘background’ that underpins them obviously evaporates completely. For the panpsychist in the tradition of process philosophy, it is thus a pressing task to carefully and critically scrutinize the prospects for structural realist metaphysics. Things are complicated by the fact that there are many forms of structural realism. In what follows some general features of the most important subdistinctions of structural realism will be discussed. What is Structural Realism? Classical realism affirms that the nature of unobservable objects is at least in approximation correctly described by our best scientific theories. Antirealism holds that truth is to be understood in epistemic terms either as 1 This paper was initially presented at the “Toward A Science Of Consciousness” conference, Tucson, Arizona, April 2010, and again, slightly expanded, at the “The Mental as Fundamental” conference, Vienna, May 2010. I wish to thank David Chalmers, William Seager, Philip Goff, Sam Coleman and David Skrbina for critical questions and commentaries.

16 ideal rational acceptability or as maximal coherence; thereby any claim about the correspondence between our theories and the mind-independent world is avoided. Structural realism takes a middle position by maintaining that we should commit ourselves to the structural and mathematical content of our theories only. The standard example is the transition from Fresnel’s ether to Maxwell’s electromagnetic theory of optics. What was carried over in this theory shift was not the full theoretical content; there was rather a continuity of form or structure. (Worall 1989, 117) The best way to characterize structural realism is the standard Ramsey method of eliminating theoretical terms. Each referring term in a given theory is replaced by an existentially bound variable. The result is a claim of the following general structure: There exist certain entities (represented by variables only) that are configured structurally in a certain way. It is still maintained that the theoretical entities exist; however, the theory does not refer to them directly but rather by means of bound variables and the relational structure that is (supposedly) known by observation. In one reading of structural realism, the theoretical entities “in themselves” remain unknown; only their structural interdependence is knowable. Structural realism thus understood combines a Kantian anti-realist view about the nature of things in themselves with a realist view about the relational structure. More austere forms of structural realism embrace radical eliminativism about anything non-relational. Before these different positions will be discussed in greater detail, the most powerful and general challenge to structural realism by the panpsychist will be presented: the hard problem of consciousness. This problem serves as a backdrop for the entire discussion presented here. Only then will the analysis of the various strands of structural realism highlight the problems that the respective structural realists have to face in a more fine-grained way. In conclusion I will argue that the challenge of structural realism to panpsychism is real but that it poses no serious threat to the viability of a panpsychist view. The Incompleteness of Structure: A “Chalmersian” Presentation of the “Hard Problem of Consciousness” The key argument against any form of structural realism is that within it the hard problem of consciousness arises in its toughest form: there is a deep epistemological and ontological gap between relational structure and phenomenal consciousness. Following upon recent research in the philosophy of mind, the discussion of the “hard problem of consciousness” has resulted in a widespread

17 acceptance of the thesis that the qualitative mental properties of phenomenal experience cannot be fully reduced to relational or structural properties. It is the intrinsic qualitative aspect of the mental that cannot be captured by the causal and functional concepts of the physical sciences. Because of their intrinsic nature, qualitative mental states cannot be captured by scientific analysis. The properties that science finds are all dispositional. Dispositions require a categorical (non-dispositional) basis on which they supervene. For present purposes, “intrinsic properties” will refer to those properties that a thing has in itself, independently of its relations to other things: the properties it could have even if it were the only thing in the universe. The mental properties of the Cartesian thinking thing are intrinsic in this sense because it can be conceived that all of its mental states could exist even without a material external world. The richness of its intrinsic properties is thus in principle independent of external relations (maybe with the exception of God). In contemporary philosophy of mind, functionalism has been criticized as capturing only the relational aspects of mind, and as missing the intrinsic, qualitative mental properties. Can physical properties be intrinsic in this sense? That is a difficult question. Mass is considered by some to be an intrinsic property. But then, having a mass of m is a property such that something that has that property will play a certain functional role defined by a relation of force and acceleration: m=F/ a. What about rest mass? It is a consequence of general relativity that only an isolated system would have a coordinateindependent mass. Since a non-isolated system is constantly exchanging energy-momentum with its environment, the mass at a certain point in time would depend on the simultaneity determinations of the observer. In quantum mechanics, mass is ultimately explained by the Higgs mechanism, which is clearly a functional concept. It seems as though, if only we dig deep enough, even physical concepts that looked prima facie like non-relational intrinsic properties, turn out to be defined relationally. It is thus reasonable to ask: What is the intrinsic categorical nature of those entities that are relationally defined by physics? Many arguments have been advanced to show that facts about qualia are not implied by physical facts. I will focus here on David Chalmers' famous “zombie argument” because it helps to clearly indicate where in the logical structure of the debate the physicalist is possibly forced to draw panpsychist conclusions. The general form of the argument is this (Chalmers 2002, 249):

18 Let P be the conjunction of all micro-physical truths about the universe, and let Q be an arbitrary phenomenal truth about the universe. (1) P&~Q is conceivable. (2) If P&~Q is conceivable, P&~Q is metaphysically possible. (3) If P&~Q is metaphysically possible, materialism is false. (4) Materialism is false. But, says the standard scientific essentialist (type-B materialist), P&~Q is only conceivable but not metaphysically possible. The mistake, according to the type-B materialist, is that we are working with primary intensions when dealing with epistemic possibility (conceivability), and with secondary intensions when dealing with metaphysical possibility. In the first case, we consider a world as actual; in the latter case we consider a world as counterfactual (how things might have been but are not). If we consider Putnam's XYZ-world as actual, then “water is not H2O” is true. This is an instance of an epistemic possibility. If we rigidly hold the meaning of “water”, as fixed by the actual world, and counterfactually consider the XYZ-world, then “water is not H2O” turns out to be false. Then “water is H2O” expresses a Kripkean necessary truth. Let us call possibility associated with primary intensions “l-possibility,” and possibility associated with secondary intensions “2-possibility.” In order for Chalmers' argument to meet the challenge of type B-materialism, it should rather look like this (Chalmers 2010): (1) P&~Q is conceivable. (2) If P&~Q is conceivable, P&~Q is 1-possible. (3) If P&~Q is 1-possible, P&~Q is 2-possible. (4) If P&~Q is 2-possible, materialism is false. (5) Materialism is false. Here, the truth of (3) requires that both P and Q have primary and secondary intensions that coincide. In the case of Q, this seems unproblematic. If something feels like pain, it is pain. If something feels like consciousness, it is consciousness. In the case of P, however, the issue becomes much more problematic. Physical properties are functionally defined. We can say that the primary intension of “mass” picks out whatever plays the mass role in a given world. We can also say that the secondary intension of “mass” is tied to the property playing that role in

19 our world in such a way that in a world where something else plays the mass role, this role filler is not mass. Premise (3) can be rejected on these grounds. But what does that mean? In that case there would be possible worlds that verify the structural description of our world in physical terms without being an exact duplicate of our world. The physical structure of those other worlds would be indistinguishable from our world, but the intrinsic natures carrying those relations would be different. The most interesting case would be worlds verifying P&~Q. This leads to an interesting metaphysical picture: the structural properties of physics in our world do not necessitate the Q-properties (phenomenal properties); the Q-properties do not supervene logically on the structural properties. However, the structural properties of physics together with additional intrinsic properties necessitate the emergence of phenomenal consciousness. This metaphysical picture has been eloquently expressed by Astronomer Sir Arthur Eddington in his work “Space, Time, and Gravitation”: “Physics is the knowledge of structural form, and not knowledge of content. All through the physical world runs that unknown content, which must surely be the stuff of our consciousness” (Eddington 1920, 200). Russell's “neutral monism” was based on similar intuitions: “As regards the world in general, both physical and mental, everything we know of its intrinsic character is derived from the mental side, and almost everything we know of its causal laws is derived from the physical side” (Russell 1927, 402). The structure of Chalmers' argument thus comes finally down to this (Chalmers 2002): (1) P&~Q is conceivable. (2) If P&~Q is conceivable, then P&~Q is 1-possible. (3) If P&~Q is 1-possible, then P&~Q is 2-possible or Russellian monism is true. (4) If P&~Q is 2-possible, materialism is false. (5) Materialism is false or Russellian monism is true. Russell's intuition was that we lack information about the intrinsic nature of the physical world, in virtue of which (plus the relevant laws) the emergence of conscious mind can be explained. This is, of course, just a conjecture, but it is certainly a possibility that knowledge of the intrinsic properties of matter would help to overcome the puzzle of emergence. In Russell's words: “The physical world is only known as regards certain abstract features of its space-time structure – features which, because of

20 their abstractness, do not suffice to show whether the world is, or is not, different in intrinsic character from the world of mind” (Russell 1948, 240). To use Whitehead's term, the modern notion of matter presents us with “vacuous” entities whose intrinsic nature is unknown (Whitehead 1929, 29). Inspired by Humean arguments, Peter Unger has recently presented a visually compelling picture of this problem. Let us define two worlds in purely structural terms, not assuming any intrinsic qualitative properties. The first world is a classical Newtonian world of particles moving about in empty space according to the laws of physics. Call this the “particulate world.” In the second world there is a continuous material plenum (a continuous field of matter) in which there are little, perfectly empty spaces, or absolute vacua, or simply “bubbles.” Call this the “plenumate world.” Now let us assume that the two worlds stand in an isomorphic relation in such a way that for each particle in the particulate world there is a corresponding bubble in the plenumate world (in the same location, governed by the same laws). It is Unger's contention that these two worlds are functionally equivalent (Unger 2006, 21-31). A structural description would be unable to capture the differences between these two worlds. The structural realist might regard it as a theoretical advantage that his/her account abstracts away from the underlying differences and concentrates on the isomorphic structure only. But, as this simple example makes quite clear, the intuition that the structural description misses something of great importance is powerful. An alternative metaphysics would thus have to assume more than just structure. One way of spelling this out could be a dual aspect theory: The relational properties account for the structural form, but the absolutely intrinsic properties account for the ultimate realizers of the relational structure. One might even think of some kind of “hylomorphism” of the relational and the intrinsic; both aspects together constitute a concrete entity. Thus the relational and the intrinsic aspects of reality have basic ontological status, without one having clear priority over the other. Alternatively, a neutral monism can also serve as the metaphysical framework. In this case, the basic properties of the world are neither physical nor phenomenal; rather, the phenomenal and the physical are constructed out of them. “From their intrinsic natures in combination, the phenomenal is constructed; and from their extrinsic relations, the physical is constructed” (Chalmers 1996, 156). In any case, this ontology implies that something mental or proto-mental is a fundamental feature of the world.

21 Now, if the analysis sketched above is correct, then any form of structural realism is misguided. It cannot be a complete realistic view of the world. The panpsychist could stop right here and consider her case closed. But this would seem to be rushing things. After all, there is no widely accepted solution to the hard problem of consciousness. In not providing a satisfactory solution to the hard problem of consciousness, structural realism stands by no means alone. One can thus reasonably ask: Can the panpsychist offer more than just this argument from the hard problem of consciousness? Are there internal problems of structural realism that the panpsychist might exploit to bolster her case? The panpsychist is thus well advised to attempt a critical analysis of the different versions of structural realism. For this purpose, I will distinguish epistemological from ontological structural realism. Within ontological structural realism there is the further sub-distinction of moderate and eliminative ontological structural realism. (Cf. Ladyman 1998) Epistemic Structural Realism Epistemic structural realism remains agnostic about any properties of the unoberservable realm that are not structural. The Russell of “The Analysis of Matter” (1927) is a good example of this view. Russell argued for an agnosticism concerning the physical world with the exception of its purely formal and mathematical properties. The higher order properties of physical theories can only be expressed in mathematical terms. Epistemological structural realism limits the scope of scientific realism to exactly these properties. But according to the epistemic structural realist there is an objective world out there that contains unobservable objects. We can, however, only know the relational properties of these objects. Thus we only know the structure, ultimately only the formal structure, of the world. It is immediately obvious that the panpsychist is not threatened by this argument because she makes exactly this point, that the intrinsic nature of objects “in themselves” is not known to us, except in the case of our own consciousness. The panpsychist is thus not an agnostic about the ultimate intrinsic nature of things. That puts her at a distinct advantage over the epistemic structural realist in explaining the emergence of consciousness. But the panpsychist's view does not contradict the account of the epistemological structural realist who is simply silent on the intrinsic nature of unobservable entities.

22 Newman's Argument The classic critique of epistemological structural realism is “Newman's Argument” (Newman 1928), which was initially directed against Russell. Newman's Argument is best understood as a reductio aimed at epistemological structural realism, showing that it ultimately collapses into anti-realism. Two claims by Russel are relevant here; he argued: 1. “ … it would seem that wherever we infer from perceptions it is only structure that we can validly infer; and structure is what can be expressed by mathematical logic “(Russell 1927, 254) 2. “The only legitimate attitude about the physical world seems to be one of complete agnosticism as regards all but its mathematical properties” (Russel 1927, 270) Newman argues that the existence of a mathematical structure is trivially true of a set of objects. He gives the following example (Newman, 1928, 139f.): What does it mean to say that “two systems of relations have the same structure”? Assume, set A is a random collection of people, and R the twotermed relation of “being acquainted”. A map of A can be made by making a dot on a piece of paper to represent each person and then those pairs of dots which represent acquainted persons are joined by a line. Such a map is itself a structured system. This new system B has the same structure as A. The structure-generating relation, S, is in this case "joined by a line". The important aspect of this example is that it is not at all necessary for the objects composing A and B, nor the relations R and S, to be qualitatively similar. “In fact to discuss the structure of the system A it is only necessary to know the incidence of R; its intrinsic qualities are irrelevant. The existence of a structure is trivially true of a set of objects.” According to Newman, a statement describing a certain structure with regard to a number of objects is a trivial statement. Why is the existence of a structure trivially true of a set of objects? Because, for Newman, a structure is purely formal and mathematical and furthermore it is independent of the intrinsic qualities of the objects. If only the structure is known, then besides what is logically deducible from the properties of the structure, it is only the number of constituting objects that can be known. But if all we know about the objects is their cardinality, if we do not know any properties of the objects that ground certain relations and exclude others, then – mathematically – any system of relations over these objects is as good as any other; all of them are instantiated. Relations are simply

23 sets of ordered sequences of entities. Given the entities, all of those ordered sequences will exist, as a matter of pure mathematics. There is only the structural reality consisting of relations, and individual objects are points defined by their place in the overall relational system. Another illustration for this idea is mathematical graph theory. One could say that nature is like a mathematical graph, in which relations obtain between point-like nodes (Cf. Dipert 1997). There are no non-relational properties. Unless there is something, as Newman says, “qualitative” (involving intrinsic properties) about the relata that determines the nature of the relations, the relations, of which the structure is supposed to be a formal abstraction, are in no way determined. Without knowing what exactly is related, one does not know what the nature of the relation is, except for its purely formal characteristics. The very idea of realism, describing nature as it exists independent of the mind, is thus undermined. Epistemological structural realism is meant to vindicate and not to revise the ontological commitments of scientific realism. On this view the objective world is composed of unobservable objects between which certain properties and relations obtain. But we can only know the properties and relations of these properties and relations, that is, the structure of the objective world. The problem gets even worse: If Newman's argument is correct, we do not even know the objective structure of the world in any realist sense of “objectivity”. Scientific realism collapses into anti-realism. The same point can be made by referring to Putnam's famous modeltheoretic argument (Putnam 1980). As argued by Newman, given a number of objects, any relational structure configuring them is already given. If we picture objects as mere nodes in a relational graph with no intrinsic nature, then for each structure there are many different relations between the objects that make the propositions describing the structure true. What is the intended model of the structure? Which one is the real relational structure of the mind-independent word? There are too many ontological interpretations (models) of our theories. Our scientific descriptions of the world are unable to single out the intended model, i.e., the real world. Since science deals only with mathematical structure, and not with real relations between objects (which are determined by the qualitative intrinsic natures of the objects), we can never know the one true story about the world in a metaphysical-realist way. We have too many “truths.” The panpsychist can thus argue that epistemological structural realism fails in the attempt to provide a genuine alternative between classical scientific

24 realism and anti-realism. It is rather a form of anti-realism. As such it remains agnostic about the intrinsic nature of “things in themselves”, and is thus no threat to panpsychism. A nice illustration for this is provided by Kant. Even though Kant defended a version of epistemological structural realism, he was still on a metaphysical level sympathetic to panpsychism because “thinking” or something “analogous to thinking” was for him the only reasonable candidate for the intrinsic nature of substances in space. For Kant it was obvious that the only absolutely intrinsic properties we can conceive of are taken from the mental realm. It is worthwhile to quote him at length: It is quite otherwise with a substantia phenomenon in space; its inner determinations are nothing but relations, and it itself is entirely made up of mere relations. We are acquainted with substance in space only through forces which are active in this and that space, either bringing other objects to it (attraction), or preventing them penetrating into it (repulsion and impenetrability). We are not acquainted with any other properties constituting the concept of the substance which appears in space and which we call matter. As object of pure understanding, on the other hand, every substance must have inner determinations and powers which pertain to its inner reality. But what inner accidents can I entertain in thought, save only those which my inner sense presents to me? They must be something which is either itself a thinking or analogous to thinking. (CPR B321, transl. Norman Kemp Smith)

Kantian empirical agnosticism in regards to the intrinsic nature of unobservable objects (as opposed the mere phenomena) is compatible with metaphysical leanings towards panpsychism. The only alternative for the epistemic structural realist might be to expand her notion of the observable object, giving it more than just structural content. Thus it might be possible to escape Newman's and Putnam's argument. But then epistemic structural realism collapses into classical realism, which naively assumes that the nature of things is directly discoverable by science. Ontological Structural Realism It should thus not come as a surprise that the “real” adversary of panpsychism is ontological structural realism. An initial, and a still somewhat superficial, reading of ontological structural realism asserts simply that there is nothing more to the entities than their place within the relational system. Below we will call this version “moderate ontological structural realism”: According to this view, the Ramseyfication has to be

25 interpreted ontologically. The bound variables in the Ramsey sentence refer to objects that have no intrinsic nature. In other terms, it is not only our description of the world that becomes like a mathematical graph, but furthermore the nature of mind-independent entities is reduced to pointlike places within the structure. In the formalism of contemporary particle physics, the ultimate units of matter are described as point-like. In quantum mechanics individual particles have no clear identity conditions, no intrinsic natures. Ontological structural realists argue that the nature of physical reality as presented by contemporary physics is incoherent with standard metaphysical views about the ontological relationship between relations, individual objects, and intrinsic properties. Instead of reformulating classical substance theory in a dynamic, process-oriented way as Whitehead proposed, ontological structural realists argue that the best interpretation of contemporary physics is that there is ultimately only structure and all particular objects, if they are to be considered at all, are point-like and void of any intrinsic properties. The Newman/Putnam argument seems ineffective against ontological structural realism. The objective mind-independent world is, according to ontological structural realism, a complex structure. The epistemological question, whether our theories correctly represent this structure, is irrelevant. The relational world-structure is, so it seems, something concrete, not an abstract mathematical object to which Newman's argument would apply. It cannot be said that every possible relation between objects is instantiated, only one particular set of relationals is concretely realized in this world. But how is this possible within ontological structuralism? Ontological structural realism might initially look like functionalism, but there is supposedly a big difference. Functionalism requires that the system of relations it specifies be implemented or realized by some appropriately organized system of entities whose own properties permit them to ground the functional specification of the system. Ontological structural realism dispenses with the realization requirement: the system of relations all by itself is enough to underpin the reality of the entities at issue. The Carrier Problem It is precisely this idea that the panpsychist needs to question. In his paper “Pattern and Being”, John Haugeland assumes the traditional view that a substance needs properties, which it has regardless of anything else. He then considers the ontological status of the pieces in a chess game – say a rook or a pawn – and claims that their very nature is determined entirely by

26 how they move about in the chess game in relation to other pieces. A chess game is a perfect example for a small structural world. Haugeland writes: “No rook is a substance. ... Nothing about a rook is determinate, not even its 'rookness,' apart from its participation in a chess game” (Haugeland 1993, 63). The situation is even more complicated. The formal definition of a type in a chess game is circular. The nature of each type is completely determined by the set of allowable moves it makes within the game as a whole. The chess game as a whole, however, is defined by the interdependent set of types that play functional roles in it. Each part of the game presupposes the existence of the whole game, and the game presupposes the existence of its parts. Why isn't this circularity of chess categories vicious (Cf. Rosenberg 2004, 234)? How can chess games actually and concretely exist? Rosenberg claims that there must be something distinct from the formal structure that actually grounds the game in concrete reality. In the case of a chess game we have physically distinct objects that serve as stand-ins, or realizers, of the relevant types, thus allowing for the existence of concrete tokens of those types. Of course, there is much more to consider here, such as the concrete chessboard or the physical position of the players in space. Without such “carriers” of the formal structure, the game would remain too incomplete and abstract to exist concretely. Rosenberg extends this thought to other, more complex, conceptual systems such as those constructed by scientific theories which are also merely abstract and circularly defined. Consider cellular automata in computer science. Each cell is defined by its role in the entire system, and the entire system is defined by the cells. Cellular automata may exist as computational systems because there is something external to the formal system that realizes or carries the computational system. The physical states of the hardware are the carriers of the cellular automata. Biology, as an abstract conceptual system, is carried by the mechanics of molecular biochemistry, psychology by the dynamical properties of the neural system, and economics by the needs and desires of individuals. The crucial question is, however: What carries the most basic physical level? Physics presents us a world of interdependently defined functional roles. Are there any properties that can give this circularly defined conceptual system a foothold in concrete reality? This is a puzzling question. Let us call it the “ultimate carrier problem.” The ontological structural realist will have to argue that there is in fact no carrier problem. However, Leibniz argued that all extrinsic, relational properties have to be grounded in intrinsic properties. Relations have to relate to something. This idea has strong

27 intuitive force. How could it possibly be that all relational properties of material objects be grounded in intrinsic properties? If grounding means that the intrinsic properties are the constituting base of the relational extrinsic properties, then we have a metaphysical system in which certain fundamental entities with absolutely intrinsic properties constitute the entire remainder of reality – a view that Leibniz famously argued for in his Monadology. The key was to replace “mutual causal influence” with “mutual information” (Seager 2006, 4). The whole of physical space was constructed by giving each monad a spatial viewpoint, from which space was constituted as experienced space. The idea of ultimate carriers can lead directly to panpsychism. It is for this reason that the denial of the need for ultimate carriers in ontological structural realism is a direct threat to panpsychism. But the structural realist encounters a serious problem here. If the ultimate relational system is not carried in any way, then it is a purely mathematical abstract object. If this is the case, then Newman's argument applies to this relational ontology after all: If there are no intrinsic properties of the objects that ground certain relations and exclude others, then – mathematically speaking– any system of relations over these objects is as good as any other. All of them are instantiated. Relations are simply sets of ordered sequences of entities. Given the entities, all of those ordered sequences will exist, as a matter of pure mathematics. Thus Newman's argument gets ontologized. It is no longer the epistemological question, whether we know how a formal structure is realized in the mindindependent world. The question at this point becomes, how does a mindindependent world, which consists entirely of mathematical point-like objects and where all formal relations between objects are trivially given, singles out the appropriate relations, so as to constitute the concrete world in which we live? This seems to be a case of the utterly unintelligible emergence of the concrete from the abstract. So Newman's argument reemerges in an ontological reading. The only way to avoid this situation would be to eliminate individual objects altogether. At this point an important distinction needs to be made. The brand of ontological structural realism which we have discussed up until this point, was mostly moderate ontological structural realism. It has to be distinguished from its more radical cousin, eliminative ontological structural realism. The difference can easily be stated: Moderate ontological structural realism defends a two-category ontology: objects, relation. However, objects are bare of intrinsic properties.

28 Eliminative ontological structural realism adopts a parsimonious, onecategory ontology: there are only relations. Both versions of ontological structural realism will now have to be interpreted separately. Moderate Ontological Relationalism This position tries to occupy space between two extremes. The first extreme is the view that all relations have to be grounded in the intrinsic properties of objects, thus granting ontological primacy to the objects. The other extreme is the outright denial of the existence of anything other than structure. The moderate ontological realist, however, places relations and objects on equal footing. Both ontological categories are interdependent. The relations unite the relata. The relata are the meeting points of the relations. Relations require relata. One presupposes the other; they are circularly defined (Cf. Eddington 1928). Objects are bundles of relations, and relations are connections between objects (Cf. Esfeld 2008). The challenge is to make sense of this initially appealing idea. If relations and objects are circularly defined, why is this circularity not vicious or even “deadly”? As we have seen, according to Rosenberg, such circularlydefined differences are harmless because their existence is grounded in something that carries those contrasts. Properties that are functionally defined within a system are relative to this system and cannot carry the relational structure of the system. Something external to the system is needed (Cf. Rosenberg 2004, 237). In order to avoid an infinite regress of ever more fine-grained systems, where each lower structure is serving as the carrier of the next higher one, a stopper is required. There are good scientific reasons to assume that nature has a lower size limit (Planck size scale). Only a property that is intrinsic tout court and not relative to a system could bring this about and serve as an ultimate carrier. Are there properties that are not intrinsic to any system, but at least partly intrinsic to themselves? The only candidates we know of are phenomenal qualities. One cannot understand the nature of phenomenal qualities by knowledge of their contextual relations alone. This radical intrinsicness is the very nature of phenomenal qualia. Whatever grounds the structural-relational properties of the world must have this radical intrinsicness. It might well be that our own consciousness is the closest analogue we get to this underlying reality. Now if objects and relations form such an interdependent system, the carrier must be external to this system. Rather than the intrinsic properties of the objects - whose existence the moderate

29 ontological structural realist denies - there must be something else, out of which objects and relations arise. And this ultimate reality had better have the radical intrinsicness that stops the regress. This line of reasoning opens up an argument for proto-mental properties in this underlying reality. Another problem with moderate ontological realism is the idea of objects that lack intrinsic properties. How can this notion of an object be spelled out? A thing, that has no nature that makes it what it is, is indeed an utterly mysterious thing. Even an omnipotent creator could not create a world in which this thing existed just by itself. With regard to numbers, Russell made this often quoted point in the “Principles of Mathematics”, which can be extended to other individual objects as well: It is impossible that the ordinals should be, as Dedekind suggests, nothing but the terms of such relations as constitute progressions. If they are to be anything at all, they must be intrinsically something; they must differ from other entities as points from instants, or colors from sounds... Dedekind does not show us what it is that all progressions have in common, nor give any reason for supposing it to be the ordinal numbers, except that all progressions obey the same laws as ordinals do, which would prove equally that any assigned progression is what all progressions have in common. … His demonstrations nowhere – not even when he comes to cardinals – involve any property distinguishing numbers from other progressions. (Russell 1903, 249)

To be an individual object implies having intrinsic properties. Esfeld's contention that objects are bundles of relations gives rise to similar worries. How is such a bundle unified as an individual? Esfeld inherits all the problems of the bundle theory of concrete particulars. There are far too many bundles of relations co-present at any given point in time. What distinguishes the true individual object from a random bundle of relations? Esfeld cannot refer to a “haecceitas” or “thisness” as the individuating property since this would obviously be an intrinsic property of the object. It is not sufficient to attribute invariance under certain symmetric transformations in the mathematical formalism of physics to individuate things. Transformation invariance is an entirely structural property, which is utterly silent on and has nothing to say about what is invariant during transformation. What constitutes the nature of the entity that endures during these transformations? Moderate ontological structuralism does not provide an answer to this question.

30 Eliminative Ontological Structural Realism The most promising strategy for the structural realist might then be to eliminate individual objects altogether. There is nothing but structure. Newman's original argument against epistemological structural realism is clearly not effective here. It is an epistemological argument and it builds on the assumption that the number of objects is given. Also, an ontological reading of Newman's argument is without force. If there are no individuals, then arguments based on the requirement of intrinsic natures have no target at which they can aim. It is futile to argue about the necessary properties of individual objects if the theory denies their very existence. However, there are other difficult challenges for the eliminative ontological structural realist. The first one is grounded in a simple conceptual analysis. Does the very notion of a relation make sense if there are no relata? Formally speaking, relations are construed as n-place predicates. It seems to be a conceptual truth that relations relate something. On the metaphysical side even deeper problems arise. The eliminative ontological structural realist has to deny the need for carriers of structures; otherwise the carrier problem and the question of intrinsic properties carrying the relational structure re-emerges. But if there are no carriers, then the formal, mathematical structure is the ultimate reality. The eliminativist ontological structural realist cannot give an account why the world is physical rather than mathematical. (Cf. Ladyman/Ross 2007.) The original motivation for structural realism was to abstract away from the unknown physical reality and work only with the second-order structural description given by mathematical physics. Thus there was a clear distinction between structure on the one hand and what is being structured on the other. But according to the eliminative version of structural realism, there is nothing but structure. The contrast between structure and non-structure has been lost (Cf. van Fraassen 2006). The distinction between classical scientific realism and structural realism has collapsed. Eliminative ontological structural realism is a classical metaphysical realism, which claims that the entire universe is a complex mathematical structure of relations. The resulting picture is Pythagorean in spirit. The entire universe emerges from formal structures. Very few eliminativist ontological structural realists embrace this Pythagorean view wholeheartedly (for example: Tegmark 2007), but this consequence seems all but inevitable if one thinks the position through. How the concrete physical world as we know it emerges from this mathematical structure is a case of mysterious inter-attribute emergence, in this case the emergence of concrete physical objects from abstract

31 mathematical structures. We began this paper with a Chalmersian argument for the incompleteness of physics: The functionally defined physical facts alone cannot account for the emergence of qualitative phenomenal facts. This problem gets aggravated if the physical facts are ultimately nothing but mathematical facts. Some structural realists have clearly acknowledged this problem, Dipert even hinted at panpsychism as a solution: There might at first seem to be no place in these cold graphs for minds, consciousness, and other mental phenomena – unless, that is, everything is mental. Although within the dialectic of this essay it is wild and possibly irresponsible speculation, we should perhaps consider seriously the possibility that something like the pan-psychism of Spinoza, Leibniz, or Peirce is true, and that vertices are pure feelings (Peircean 'firstnesses'), constituting a distinct thought or object only when connected to other such entities. (Dipert 1997, 358)

The question then is whether eliminative ontological structural realism is capable of providing a realistic metaphysics of the nature. Galen Strawson has in recent years mounted a critique of classical physicalism that can a fortiori be raised against the Pythagorean metaphysics, to which the eliminative ontological structural realist is committed (Strawson 2006). It is based on a venerable argument for panpsychism, the “genetic argument.” It rests on a claim concerning the intelligibility of radical inter-attribute emergence, as opposed to weaker intra-attribute emergence. In his paper, “Panpsychism” Thomas Nagel had argued that uniform psycho-physical correlations could not account for the emergence of mental properties from the physical components of a system: Instead, intrinsic properties of the components must be discovered from which the mental properties of the system follow necessarily. This may be unattainable, but if mental phenomena have a causal explanation such properties must exist, and they will not be physical. (Nagel 1979, 187)

Emergence cannot be brute in the sense of there being nothing in the emergence base in virtue of which the emergent phenomenon emerges: ex nihilo nihil fit. Nothing can give what it does not possess. In order to make sense of the emergence of consciousness, the evolution of the phenomenal mind must be smooth. Consciousness in some form must be present “at the very origin of things” (James 1890, 149). Inter-attribute emergence is to be

32 avoided. Otherwise even a Cartesian soul could mysteriously emerge from a Cartesian physical body (or, as in the case of eliminative ontological structural realism, the emergence of a concrete physical object from a configuration of abstract mathematical objects). If such inter-attribute emergence is considered intelligible, then one will inevitably end up with a radical Humean view of causal powers in which “any thing may produce any thing” (Treatise III, xv), and any reasonable prospect for a substantive metaphysical account of the mind-body relation will have vanished. If emergence is construed as an intra-attribute relation, however, then one has to give up the standard physicalist principle that there are no (proto)mental properties at the ontological base levels. Strawson calls this dogma of contemporary physicalism the “NE principle”: “physical stuff is, in itself, in its fundamental nature, something wholly and utterly nonexperiential” (Strawson 2006, 11). The intuition, that a system that is only structurally defined cannot give rise to qualitative experience, is more than just an epistemic problem of cognitive upward opacity that might be overcome by more empirical research; it is a deep metaphysical puzzle. This problem becomes even more serious if the relational structure that is supposedly generating consciousness is conceived of in Pythagorean terms. Conclusion We have come full circle. We started with the hard problem of consciousness and returned to it at the end. As it turns out, the hard problem of consciousness is the major stumbling block for any ontological structural realism. The panpsychist claims to have a better answer to the hard problem of consciousness because no miraculously strong emergence is required. Epistemological structural realism is another matter, however. Since it is silent on the intrinsic properties of matter, it is compatible with a panpsychist metaphysics. Indeed, someone who endorses structural realism for epistemological reasons only, can have “secret” panpsychistic inclinations at a metaphysical level. As was shown, Kant is certainly a case in point. We have argued that moderate ontological structural realism provides no conceptual means to account for the individuation of objects without intrinsic properties. Also, if relations and objects are interdependent and circularly defined, a grounding problem for the circular structures arises. For those reasons, moderate ontological structural realism seems to be a conceptually unstable theory. Eliminative ontological structural realism is conceptually more coherent.

33 But it has problems of its own, the most important one being the collapse of the distinction between the physical and the mathematical. The Pythagorean metaphysics implied in this account is the source of a plethora of metaphysical problems, not the least of them being the emergence of consciousness. Thus, the panpsychist can endorse epistemological structural realism as an epistemological interpretation of the scientific endeavor. For the reasons given above, the panpsychist will reject structuralist metaphysics in both forms: moderate and eliminative. References Chalmers, D. (1996): The Conscious Mind. In Search of a Fundamental Theory, Oxford: Oxford University Press. Chalmers, D. (2002): 'Consciousness and its Place in Nature' in: Chalmers, D. (ed.), Philosophy of Mind, New York: Oxford University Press, 247272. Chalmers, D. (2010): 'The Two-Dimensional Argument Against Materialism' in: Chalmers, D. (ed.), The Character of Consciousness, Oxford: Oxford University Press, 141-191. Dipert, R. (1997): 'The mathematical structure of the world: the world as graph' in: Journal of Philosophy 94 (1997), 329-358. Eddington, A.S. (1920): Space, Time and Gravitation, Cambridge: Cambridge University Press. Eddington, A. S. (1928): The nature of the physical world. New York: MacMillan. Esfeld, M. (2008): Naturphilosophie als Metaphysik der Natur. Frankfurt: Suhrkamp. Haugeland, J. (1993): 'Pattern and Being' in: Dahlboom, B. (ed.), Dennett and His Critics, Cambridge: Blackwell, 53-69. Hume, D. (1739): A Treatise of Human Nature. New Edition. London: 1886. James, W (1890): The Principles of Psychology, vol. 1, New York: Henry Holt. Reprinted New York 1950: Dover Publications. Ladyman, J. (1998): 'What is structural realism?' in: Studies in History and Philosophy of Science, 29 (1998), 409-424. Ladyman, J. (2009): 'Structural Realism”, in: Stanford Encyclopedia of Philosophy (http://plato.stanford.edu/entries/structural-realism/). Ladyman, J. and Ross, D. (2007): Every Thing Must Go: Metaphysics Naturalised, Oxford 2007: Oxford University Press.

34 Lewis, D. (2009): 'Ramseyan humility' in: Braddon-Mitchell, D. and Nola, R. (eds.), The Canberra Programme, Oxford: Oxford University Press, 203-222. Nagel, T. (1986): The View from Nowhere, Oxford: Oxford University Press. Newman, M.H.A. (1928): 'Mr. Russell's causal theory of perception' in: Mind, 37 (1928), 137-148. Putnam, H. (1980): 'Models and Reality', in: The Journal of Symbolic Logic 1980, 45, 464-482. Rosenberg, G. (2004): A Place for Consciousness. Probing the Deep Structure of the Natural World. Oxford: Oxford University Press. Russell, B. (1903): The Principles of Mathematics. London: Allen and Unwin. Russell, B. (1927): The Analysis of Matter. London: Routledge. Russell, B. (1948): Human Knowledge. Its Scope and its Limits. London: Routledge. Seager, W (2006): 'Rosenberg, Reducibility and Consciousness', in: Psyche 12, (2006) 1-15. Strawson, G. (2006): 'Realistic Monism' in: Strawson, G. (ed.). Consciousness and its Place in Nature. Does Physicalism entail Panpsychism?, Exeter: Imprint Academic, 3-31. Tegmark, M. (2007): 'The Mathematical Universe', in: Foundations of Physics, 38 (2007), 101-150. Unger, P. (2006): All the Power in the World. Oxford: Oxford University Press. van Fraassen, B.C. (2006) 'Structure: Its shadow and substance' in: The British Journal for the Philosophy of Science, 57 (2006), 275-307. Whitehead, A. N. (1929) Process and Reality: An Essay in Cosmology. (1978, rev. ed., D. R. Griffin & D. W. Sherburne). New York 1929: Free Press. Worrall, J. (1989) 'Structural realism: The best of both worlds?' in: Dialectica 43 (1989), 99–124.

The Last Man Standing Argument for panpsychism: A rejoinder Pierfrancesco Basile (Bern) 1. Introduction In Panpsychism in the West, David Skrbina mentions, approvingly, an argument by process philosopher David Ray Griffin: In light of “the ‘terminal’ failure of the approaches built on the Cartesian intuition about matter,” panpsychism stands as the most viable alternative. This is an important point, and one that has been neglected in the past. If intensive critical inquiry of dualism and materialism over the past, say, few hundred years has failed to produce a consensus theory of mind, then it stands to reason that a third alternative like panpsychism, in some positive formulation, should gain its viability. This “negative” argument for panpsychism may be called, for want of a better name, the Last Man Standing Argument. (Skrbina 2005, 252; see also Griffin 1997, 91)

This is an intriguing argument. But it is surprising that Skrbina should endorse it. One of the main purposes of his book is precisely to show that panpsychism has been an underlying theme in Western philosophy. Since this is a task the book achieves remarkably well, one is left wondering why panpsychism should be regarded as having been “neglected in the past.” It is true that panpsychism has been ignored in the last fifty years or so within analytic philosophy of mind. Nevertheless, half a century amounts to very little if, as Skrbina contends, panpsychist motives are a constant theme from the Presocratics onwards. This preliminary observation suggests a question. William James speaks in his A Pluralistic Universe of “the great empirical movement towards a pluralistic panpsychic view of the universe, into which our generation has been drawn.” (James 1909, 270) This is probably a correct estimate of the intellectual situation of his time. Philosophers such as Hermann Lotze, Gustav Fechner, James Ward, Charles Sanders Peirce, Charles A. Strong, Josiah Royce, Gabriel Tarde and Henri Bergson, just to mention a few, endorsed a version of the theory. But if about one century ago panpsychism was such a widespread position, why was it forgotten in the later decades of the twentieth century? This is a question for historians of philosophy to answer. Apart from obvious reasons such as the rise of analytic linguistic philosophy, it seems possible to speculate that

36 panpsychism might have suffered exhaustion in the way most philosophical theories do. Such theories are seldom conclusively refuted; rather, they suffer deposition when they appear unable to answer fundamental questions and spark fruitful lines of novel inquiry. After a first period in which they ignite romantic enthusiasms, philosophical theories either grow or cease to look interesting. Skrbina does not ask in his ample survey of the history of panpsychism whether the theory might have come to be perceived, at one point or another in the earlier part of the twentieth century, as one more “terminal failure.” This raises the question for this paper: could it be that, contrary to what is held by Griffin and Skrbina, panpsychism is not “the most viable alternative” after all? 2. What’s so good about panpsychism? In order to clarify this point, let’s briefly rehearse the case for panpsychism. One main argument is the so-called “intrinsic nature” argument. This can be reconstructed as consisting of two parts, the first one of which is as follows: (1) natural science provides a description of the relational structure of reality; (2) structure must be the structure of something—i.e. relations require terms between which to hold; therefore, (3) such terms, if real at all, must possess an intrinsic nature of their own. This part of the argument is meant to secure the conclusion that there is a meaningful question, namely, the one concerning the inner nature of the fundamental constituents of reality that is left unanswered by natural science. Once this is recognized, there is room for drawing the panpsychist conclusion: (4) since the only reality whose intrinsic nature we are acquainted with is our own subjectivity, the only way to account for the inner being of reality’s basic constituents is in experiential terms—that is, by conceiving of them by way of analogy with our own conscious, phenomenal selves. Why should the self, as this is actually lived through, be assigned such a fundamental metaphysical role? One main reason adduced by panpsychists is that the concretely experienced self is precisely what it seems to be; here, there is no distinction between the phenomenon and the noumenon. As the late Timothy Sprigge put it: “Surely it is good sense to take the one example of a thing we know concretely as thing in itself as our paradigm for conceiving the nature of the other concrete things in themselves which we know are there as the noumenal backing or basis of our phenomenal physical world.” (Sprigge 1983, 105) This line of reasoning is questionable on several accounts. First, it is not obviously true that science provides only a formal description of reality.

37 (Rosenthal 2006, 123; McHenry 2010, 17-18) Secondly, a doubt might be raised as to whether reality, as disclosed by physics, really requires positing substantive (i.e. non-relational) terms or whether reality could be “relations” all the way down; pan-relationalism is a counterintuitive view, but it has its advocates (cf. Seager 2006, esp. 140-144, for an extended discussion of this issue). But even granting these two controversial points, what precisely should we project onto the noumenal backing? Should it be “will,” as Schopenhauer held, or something else, such as a sheer capacity for feeling? (This is an important point which, in different argumentative settings, will keep claiming our attention later on in this paper, especially in sections 3 and 4). Lastly, the argument is conjectural at best: why should we conclude that the unknown natures of the ultimate constituents of reality must be experiential? There is nothing in the argument’s premises that obliges us to take such a momentous step; we can concede the point that in self-experience we have direct contact with things in themselves, but what reason do we have for assuming that all things in themselves are of the same kind? With respect to non-human (or non-animal) beings, Du Bois-Reymond’s sceptical dictum, ignoramus et ignorabimus, would seem to express an equally appropriate response. Consider the way David Chalmers phrases the argument in The Conscious Mind: “After all, we really have no idea about the intrinsic properties of the physical. Their nature is up for grabs, and phenomenal properties seem as likely a candidate as any other.” (Chalmers 1996, 155) This is a candid recognition of the limitations of the intrinsic nature argument; the ascription of phenomenal properties to the ultimate constituents of nature, while not incoherent, is not a compulsory step to take. The object of this paper is to assess what Skrbina terms the Last Man Standing Argument for panpsychism. Since this is an argument by exclusion, we need to consider a line of reasoning that capitalizes on the weaknesses of panpsychism’s main alternative, materialist physicalism. The non-emergence argument is one such argument. The core idea is simple enough and was stated long ago by William James. On a Cartesian understanding of matter as wholly insentient, so the argument goes, the emergence of mentality would seem to amount to an unfathomable creatio ex nihilo: “...with the dawn of consciousness [in the course of evolution] an entirely new nature seems to slip in, something whereof the potency was not given in the mere outward atoms of the original chaos.” (James 1890, 146) Here we have panpsychism at its strongest, but there are ways of responding to this argument. (1) A first observation is that there is no proof

38 that creation out of nothing is impossible; hence, brute emergence might occur, while being an ultimate irrationality, that is, something that would remain incomprehensible even to God, i.e. to a mind satisfying the requirements of an ideal, perfected rationality; analogously, the idea of creatio ex nihilo does not seem to be logically incoherent; stated in Leibnizian terminology, it violates the principle of sufficient reason, rather than the principle of non-contradiction. But it could also be argued, following a suggestion by Colin McGinn (1989, 359), that (2) brute emergence is intrinsically rational but cognitively closed to finite minds such as ours. Alternatively, and in a more optimistic fashion, one could hold with Thomas Nagel (1986, 52) that (3) we lack at present the conceptual categories to understand mind’s place in nature; however, we should not give up the hope to develop such categories in the future. The fact that these options are available shows that the non-emergence argument is less than conclusive. Admittedly, the first option is not particularly attractive. But the case is different with the second and the third. If panpsychism should turn out to be a “terminal failure” in the same way in which materialism and Cartesian substance dualism are taken to be, then radical conceptual revision is precisely what would be needed to show that defeatist sceptics such as McGinn are wrong. For the time being, the conclusion to be drawn is that the non-emergence argument is less than a proof. Nevertheless, it is as strong as any of the best arguments in philosophy. It therefore succeeds in vindicating panpsychism as a position that deserves to be taken seriously in contemporary debates on consciousness and its place in nature. 3. Mental unity But does that mean that panpsychism is the most viable alternative? Serious difficulties arise as soon as one asks how the mind/body relation should be modelled. The panpsychist can opt either for an identity or for a dualist theory of the mind-brain relationship. We will deal with the former option in this section, while discussing the latter in the next. Before doing this, however, another question needs to be addressed: how are the substances (i.e. fundamental beings) that populate a panpsychist universe to be conceived? Specifically, should they (1) be endowed with both physical (i.e. non-experiential) and mental (i.e. experiential) features or should they (2) be conceived as being purely mental (i.e. experiential)? (In what follows, I will assume a pluralistic ontology, sidestepping the question as to the viability of Spinozistic monism for the sake of

39 argumentative convenience; but Spinoza might have been right after all and we could all be aspects of a single, unified Reality; as a matter of fact, it is not even clear that pluralism and monism exhaust all possible options; as John Leslie (2001) has argued, there could be a plurality of universes, each of which of a Spinozistic type.) Note that the question is not what is more expedient to hold, but what is consistent with the panpsychist’s appeal to the Last Man Standing Argument. Skrbina recognizes that Cartesian dualism rather than materialism or panpsychism expresses the “common intuitive feeling.” (Skrbina 2005, 13) Nevertheless, he rejects dualism on the conventional ground that two radically different substances could not interact; and since no dualist has yet been capable of explaining how this is possible, Skrbina concludes that “interactionist dualism is... currently held more as a matter of faith than of philosophical reasoning.” (Skrbina 2005, 13) At the same time, Skrbina is not willing to endorse the idealist view that all there is to the basic constituents of reality is experience. Hence, he stresses the point that metaphysical idealism (or, as it is perhaps better called, “mentalism,” since it is the view that all ultimate realities have a mind-like nature rather than the view that they are all ideas) is not to be straightforwardly identified with panpsychism; the latter is the genus of which the former is the species. (Skrbina 2005, 10) But is the panpsychist in a position to reject idealism? The first of the two options identified above, the conception of the ultimately real as a psycho-physical unity, faces the same difficulty of Cartesian dualism. This now reappears in a mutated form: how could a single thing possess both a physical and a mental side? What would hold these two heterogeneous aspects or properties within the unity of the thing? If the aim is to achieve a view that could enable us to think of experience as a wholly natural phenomenon, then it seems contra-productive to posit duality in the ultimate constituents of nature. Hence, a consistent form of panpsychism will have to reject a dual-aspect theory and settle for the second option identified above, which is a form of metaphysical idealism. This view stands in need of further clarification. One possibility is to hold that the ultimate mental items are independent existing, individual experiences of the like of Hume’s substantial impressions or Clifford’s bits of “mind-dust.” Nagel envisions a theory of this kind when he speculates that “the components out of which a point of view is constructed would not themselves have to have points of view.” (Nagel 1979, 194) As it has been argued by William James and more recently by Galen Strawson, however, it is unintelligible that an experience could exist independently of a self.

40 According to William James, it is a phenomenological truth that our mind, i.e. the total psychical whole that constitutes our conscious self at any one moment, has no genuine parts; rather, “pulses of consciousness are unitary and integral affairs from the outset,” and it would be absurd to “suppose feelings which exist separately and ‘fuse together’ by themselves.” (1890, 181) If we think otherwise, this is because we do not pay sufficient attention to the concrete nature of our mental life and are mislead by spurious analogies. James makes the point quite vividly in a note to his chapter on “The Mind-Stuff Theory”: I find in my students an almost invincible tendency to think that we can immediately perceive that feelings do combine. “What!” they say, “is not the taste of lemonade composed of that of lemon plus that of sugar?” This is taking the combining of objects for that of feelings. The physical lemonade contains both the lemon and the sugar, but its taste does not contain their tastes, for if there are any two things which are certainly not present in the taste of lemonade, those are the lemon-sour on the one hand and the sugar-sweet on the other. (James 1890, 158)

In other words, when we begin to scrutinize our experience and discern the taste of lemon from the taste of sugar, we have not provided an analysis of our mental state; rather, when we perceive the two flavours as distinct we have moved into a novel mental state. A person’s total mental state at any one moment (his momentary self) is the concrete reality; the isolated, independent perception one encounters in philosophical books is a fiction. Strawson’s route to the conclusion that experience is impossible without subjectivity is more straightforward. In his lights, it is a conceptual truth that an experience entails a corresponding experiencing subject and cannot therefore exist without a feeler of some sort or another: “An experience is impossible without an experiencer. Many... have made this point, and many have taken it to be too obvious to mention... there can’t be experience without a subject of experience, because (this is the whole argument) experience is necessarily experience-for: it is necessarily experience for someone or something.” (Strawson 2009, 271) Taking these points by James and Strawson together, we obtain a “logicophenomenological” argument that is difficult to resist. If so, the conclusion to be drawn is that all the basic constituents of reality must possess the status of subjects. This means that the most viable form of panpsychism is a form of Leibnizian monadism. (Another question I sidestep in what follows concerns the ontology of the self; the contention that selves are

41 real, however, does not need to be interpreted as a commitment to a preHumean understanding of our selves as substrata ontologically distinct from our experiences; on these issues see Strawson 2009, Sprigge 1993 and Basile 2003.) With the basic metaphysical framework more or less in place, it is now possible to address the problem of modelling the relation between the mind and the brain. The identity theory gives rise to the well-known problem of mental composition. How do the neuronal selves amalgamate so as to yield a moment of human mentality? This is a real question, but it has little force if it is taken to provide a conclusive refutation of panpsychism. As I have argued more fully elsewhere (Basile 2009b and 2010), attempts to construe the composition problem as a reductio ad absurdum implicitly ascribe to the panpsychist the view that the human mind must be constituted by smaller experiential unities in the same way in which a mosaic is constituted by its stones. I see no reason why the panpsychist should be committed to such a crude view of consciousness. Nagel gets the point exactly right when he says that the real challenge is that of working out adequate mereological categories for phenomenal parts and wholes. (Nagel 1986, 50) No argument against panpsychism can presuppose that we already have such an understanding. Furthermore, it should be noted that attempted refutations based on the composition problem prove too much. The problem of explaining how one gets a single unified moment of human mentality out of a plurality of neuronal minds is an instance of a larger ontological problem. How does one get one substance out of many other substances? The problem of combination arises with regard to non-mental entities as well; even such a simple object as a chair is more than the mere aggregate of its physical particles. If all combination had to be by way of aggregation (i.e. in the same way in which a heap of sand is made of grains or a wall is made of bricks), then not solely a unified self but none of the objects in our every-day environment could ever be real. Faced with this reply, the objector could try another strategy. Since no panpsychist has yet produced an explanation of how the human mind could emerge out of the lesser neuronal minds, no one is in a position to claim that the theory has any explanatory advantages over competitor approaches to the mind/body problem. (Goff 2009, 134) Is there really such a dialectical stalemate? As it is usually put, this reply fails to fully grasp the significance of the argument from non-emergence. To the best of my knowledge, no panpsychist has yet been capable of providing an explanation of how human mentality is generated from the several

42 mentalities in the neurons. Nevertheless, this problem is of a different order than the problem of explaining the emergence of the mental out of the non-mental. Prima facie, the generation of a human mind out of myriad lesser minds does not involve a violation of the principle ex nihilo, nihil fit. It is true that to speculate that experience might be a fundamental feature of the ultimate constituents of reality is not to have explained human consciousness; however, a context or metaphysical framework has been established in which such an explanation ceases to appear impossible. Still, the objection that panpsychism has to face its own peculiar version of the problem of emergence is probably correct. But it needs to be reformulated in a more effective way. A. C. Ewing puts it as follows: Can it be consistent dogmatically to deny the possibility of the conscious having developed out of the unconscious (in the sense of the totally unfeeling), and yet to assert the development of the humanly intelligent out of what is quite incapable of reasoning? If we are to reject the former supposition on the ground of unintelligibility, it seems that we ought to reject the latter too. (Ewing 1934, 412)

This passage captures what is sound in the objection under consideration. Why should the transition from the non-cognitive (raw feels, sensations, emotions) to the cognitive (grasping the meaning of a sentence, identifying a perceived object as an object of a given kind, solving complex mathematical problems) be less difficult to understand than the transition from the non-experiential to the experiential? There is clearly something that stands in need of explanation here. Is the gulf between sentience and cognition a less dramatic one to bridge than the gulf between sentience and non-sentience? On the face of it, it is not clear that this is so. In the case of the non-emergence argument the burden of proofs lies on the materialist; now the situation is reversed and the panpsychist has an obligation to produce a convincing explanation. The same point can be made by asking a question already touched upon when discussing the internal nature argument, namely the question as to how monadic experiences should be conceived. Should the panpsychist endow his monads with a mere capacity for sentience and then try to work out a theory capable of explaining how they could generate cognition? Should we follow the opposite path? Or should we posit both sentience and cognition as original features of a monad’s mentality? Apart from one notable exception to be discussed below, Whitehead’s panexperientialism, panpsychists are usually shy on this question. And when it is raised, they

43 fail to achieve a consensus. Gabriel Tarde criticizes Hermann Lotze for having endowed his monads with an emotional appreciation of the world, whereas he conceives of them as endowed with cognitive abilities. (Tarde 1893/2009, 19n) The book by Ewing from which the above passage was taken, Idealism: a Critical Survey, was published in 1934. I leave it as a suggestion for further research: could this be one of the points at which panpsychism appeared to early twentieth century philosophers to have reached a dead end? 4. Leibnizian interactionism If one rejects the identity theory, then the alternative model is a (noncartesian) dualist theory in which the mind and the brain are numerically distinct, yet both intrinsically experiential. This is to revive Leibniz’s idea of a dominant monad. In a forgotten but once very influential book entitled Microcosmus, Hermann Lotze rejects the identity theory on the ground that it leads to the composition problem. This he regards as posing a problem as serious as that of explaining the emergence of mind out of insentient matter: No more than it formerly seemed to us possible to explain the peculiar elements of mental life by the crossing of physical actions of the nerves, do we now find the spiritualized nature of the parts adequate to render more comprehensible the rise within us of the one consciousness. (Lotze 1885, 365)

Then, he offers his alternative: The image which we have now to form of the living form and its mental life is that of an association of many beings. The governing soul, placed at a favoured point of the organism, collects the numberless impressions conveyed to it by a host of comrades essentially similar but lower in rank from the inferior significance of their nature. (Lotze 1885, 367)

This theory resurfaces in the philosophy of Whitehead, who probably derives it from the Cambridge philosopher James Ward. (Like many of his contemporaries, Ward had studied with Lotze in Germany; see Basile 2009a for a more detailed account of the relationship between the two philosophers.) With some minor modifications, it was held by other thinkers as well: hence, since the theory is based on an analogy between a king and his subjects, Gabriel Tarde asked why we should say that the dominant monad is intrinsically “higher” than the subservient monads. And

44 indeed, how many of us truly think that our rulers are better than ourselves? (Tarde 1893/2009, 61) The naiveté of this doctrine is compensated by its advantages. First, on this doctrine the composition problem does not arise; secondly, one great obstacle to understanding mind-body interaction, the heterogeneity of the interacting substances, is removed; lastly, because it assumes a numerical distinction between the mind and the body, the theory does justice to our intuitive feeling that we are not our body yet we are somehow blended with it. This is the feeling so well expressed by Descartes in his Sixth Meditation: “Nature also teaches me, by these sensations of pain, hunger, thirst and so on, that I am not merely present in my body as a sailor is present in a ship, but that I am very closely joined and, as it were, intermingled with it, so that I and the body form a unit.” (Descartes 1641/1986, 56) The theory of the dominant monad as distinct from the monads in the body yet collecting their experiences does full justice to this feeling that “I and the body form a unit,” since the life of the mind is now literally nourished by the neuronal experiences that flow into it. However, the theory reopens the Leibnizian problem of monadic interaction. How does the dominant monad “collect” the “numberless impressions conveyed to it” by his comrades? It doesn’t seem obvious that panpsychist mind/mind interaction is any bit more intelligible than Cartesian mind/body interaction. Do we have any understanding of how two selves could causally interact? One philosopher who has tried to answer this question is Alfred North Whitehead. There are good reasons for focusing on his philosophy, for his is probably the most articulated version of panpsychism ever produced. Whitehead’s metaphysics is a relational monadology, one in which the basic constituents of reality are endowed with a capacity for causally affecting one another. Whitehead breaks with the classical tradition in philosophy and conceives of his physical ultimates as momentary occasions of experience (actual occasions) instead of as permanent, enduring substances. Such units are organized into series. Within each series, each occasion inherits (prehends) some of the ingredients of previous occasions. Through their prehensive activity, occasions come to be linked in such a way as to form diachronically unified streams of experience. Such streams interact in several ways and form complexes (societies) of different kinds. The large variety of natural beings that exist at different levels of reality—i.e. such things as electrons, molecules, cells, crystals, plants, human beings—are all explained in terms of the way

45 streams of actual occasions have come to interact in the course of cosmic evolution. They are all in one way or another “societies” made up of the more basic experiential occasions. (See Hosinsky 1993 and Kraus 1998 for accurate yet accessible introductions to Whitehead’s philosophy; my own interpretation is in Basile 2009.) The question now is whether Whitehead’s panexperientialism, as Griffin has named it, provides a satisfactory account of monadic interaction, which in that system is explained in terms of the notion of prehension. As Whitehead’s discussion in such works as Adventures of Ideas and Symbolism makes clear, the concept of prehension (or, more precisely, physical prehension) has its immediate phenomenological counterpart in what Edmund Husserl referred to as “retention” and William James as the “echoing” of the past in his analysis of the specious present. The hearing of a melody exemplifies the phenomenon. The notes just heard resonate long after they are gone and color our present appreciation of the musical piece. Now, there is little doubt that such experiences of continuation of the past into the present are real. But Whitehead goes well beyond both Husserl and James in that he provides a metaphysical interpretation of what is originally a phenomenological concept. One first source of concern here is that retention is an intramonadic relation; if it can be interpreted as an instance of causation at all, then it will be an instance of immanent causation. What we are in search of, however, is a relation of causation between different monads, that is, transeunt causation. That Whitehead takes immanent causation to provide the key for an understanding of transeunt causation suffices to cast his entire theory into doubt. For it now appears that the theory is committed to the surprising claim that we inherit from our previous moments of mentality in very same way in which we inherit from the previous moments of mentality of other actualities, i.e. the monads constituting our bodies. There is also a worry as to whether Whitehead’s analysis of the causal nexus is meaningful. In Whitehead’s account, the phenomenological “echoing” of the past is turned into a real “influx” of past experiences into the present moment of mentality: “The present moment is constituted by the influx of the other into that self-identity which is the continued life of the immediate past within the immediacy of the present.” (Whitehead 1933, 181) Whitehead understands causation as a kind of transference or contagion: something must be passed over from the cause to its effect. However, in the present moment of mentality past experiences are felt as “just gone” rather than as “immediately enjoyed.” This means that, when a

46 past experience is retained, what changes is not the experience itself but that experience’s experiential mode. And this involves a distinction between the experience per se (the item transferred from one occasion to the next) and its ways of appearing (the experience immediately enjoyed as against the experience as a vanishing echo from the past). Is this notion of an experience remaining (numerically) self-identical while changing its experiential mode an intelligible one? This is doubtful. As James points out in a number of places, the being of an experience is wholly exhausted by its appearance (see for example James 1909, 198-199). This seems entirely correct: taken naively, in his primary, pre-philosophical meaning, a word such as “pain” surely refers to our felt experience, and not to any modification in our bodies, which we would rather view as the cause or source of the pain. But if this notion of experience as being nothing over and above its immediate felt quality is correct, then the experience immediately enjoyed and the experience retained must be numerically different; surely, the retained experiences come with a different qualitative feel than those which are enjoyed for the very first time. These strike me as powerful objections against Whitehead’s account of monadic interaction. What about the ultimate experiential nature of his monads, the actual occasion? What other features do they have, other than an ability to relate to past moments of mentality? A Whiteheadian philosopher such as Griffin acknowledges that it would be a mistake to posit in all occasions the complexity we find in a human mind. As a matter of fact, one reason why Griffin terms his position “panexperientialism” instead of “panpsychism” (or “psychicalism,” as Hartshorne has it) is precisely that the term “psyche” inevitably suggests “a higher form of experience than would be appropriate for the most elementary units of nature.” (Griffin 1997, 78) This is reasonable, but problematic. Like most panpsychists, Griffin emphasizes that the only “aspect of the world we know from within, by identity” is our own conscious experience (1997, 90). If this is so, then the question is, once again, how to decide which features of our conscious experience can be generalized to all occasions in nature. In Whitehead’s theory, each actual occasion is endowed with a “physical pole,” i.e. a capacity for grasping aspects of other actualities, as well as with a “mental pole,” i.e. with a capacity for grasping universal forms or abstract concepts. But the ascription to all actual occasions of a mental pole seems at odd with the requirement that experiences occurring at the lower levels of reality—“micro-experiences” if one wants—be significantly different from the experiences of human beings. Once they

47 are endowed with a capacity for grasping abstract forms—hence, in one way or another, with an intellect—such things as electronic occasions become more human-like that one originally wanted them to be. 5. Conclusion In section 3 I have reformulated the standard objection based upon the composition problem, so as to strengthen the point that panpsychism has its own version of the problem of emergence. In section 4, I have examined what has a fair claim to be regarded as the more articulated version of panpsychism ever produced, i.e. Whitehead’s panpexperientialism, and explained my scepticism concerning the capacity of that philosophy to account for mind-body interactions. Thus, it might seem at this point that panpsychism is not in any better shape than materialist physicalism. The following question could therefore be raised: on a monistic view of reality as being made up of a single kind of ultimate stuff, the only alternatives are materialist physicalism-cum-brute emergence and panpsychism; but since both face serious difficulties, why not retreat to old fashioned Cartesian dualism? After all, even a committed pansychist such as Skrbina admits that this is the view that better captures our ordinary, pre-philosophical grasp of the mind-body relation. One of the few contemporary dualists, Geoffrey Madell, has recently argued that dualism rather than panpsychism should be regarded as the Last Man Standing: ...I am often tempted to echo Churchill, when he claimed that democracy is the worst imaginable form of government apart from all the others, and say the same of substance dualism. Reflecting on it, I am sometimes inclined to think that it, like democracy, is the worst possible solution to a problem, in this case the mind-body problem, but reflection on other suggested solutions to this problem, including that of panpsychism, leads me to think that the problems with it may after all be less daunting than those which bedevil all other positions. (Madell 2007, 51)

There is something importantly right in this response. One reason usually advanced for rejecting dualism is the alleged impossibility of making sense of causal interaction between a mental and a physical substance. In the same way in which we lack a proof that the notion of brute emergence is logically incoherent, however, we lack a proof that such interaction is impossible in principle. (See Madell 2010 for a defence of this claim.) Not surprisingly, however, dualism faces the greatest difficulty precisely there where panpsychism is at its strongest, namely with respect to the notion of

48 emergence. How could substance dualism be reconciled with the view that human mentality is an evolutionary product? Apparently, the dualist must here choose either the irrationality that comes with the notion of brute emergence or postulate some kind of divine intervention. Here is Madell’s solution: The point I wish to emphasize... is that if we accept that the emergence of consciousness... betokens some deep necessity or purpose, then the suggestion... that such emergence must seem a sheer, brute inexplicable fact is one we no longer have to accept. (Madell 2007, 50)

Madell does not elaborate on this issue, so it is unclear what his solution eventually amounts to. But the general idea is clear enough: the emergence of mind in the course of evolution would not appear irrational if we could come to see it as part of a larger teleological, purposive scheme. This response exploits the fact already noticed in section 2 that the irrationality of the notion of brute emergence stems from a violation of the principle of sufficient reason, and not of the principle of non-contradiction. From a purely dialectical standpoint, Madell’s response is therefore wholly adequate. Nonetheless, this solution is not very attractive: do we have any independent reason for believing in such “deep necessity or purpose” apart from our incapacity to understand how the sentient could be generated from the non-sentient? If not, then such an attempt to rationalize emergence might begin to look like trying to distil knowledge out of ignorance. Trying to develop Madell’s suggestion would lead us once again well beyond the relatively safe precincts of professionalized philosophy of mind into the more adventurous regions of general speculative metaphysics. However, these reflections also lead us back to the question with which this paper began. Is panpsychism the Last Man Standing? As it has been argued, the doctrine needs to overcome some significant hurdles before this claim can be vindicated, and we are yet far from seeing how that could actually be done. Hence, and although I do think that panpsychism deserves a fair hearing, I am led to conclude that its advocates cannot afford to take slightly Nagel’s remark that “panpsychism should be added to the current list of mutually incompatible and hopelessly unacceptable

49 solutions to the mind-body problem.” (Nagel 1979, 193) Radical revisionism, not panpsychism, might be what is needed.1 References Basile, P. (2011): Materialism vs. Panexperientialist Physicalism: Where Do We Stand?, in: Process Studies, forthcoming. Basile, P. (2010): It Must Be True – But How Can It Be? Some Remarks on Panpsychism and Mental Composition, in: Basile, P./ Kiverstein, J./Phemister, P. (eds.), The Metaphysics of Consciousness, Cambridge: Cambridge University Press, 92-113. Basile, P. (2009a): Leibniz, Whitehead and the Metaphysics of Causation, Basingstoke-New York: Palgrave Macmillan. Basile, P. (2009b): Back to Whitehead? Galen Strawson and the Rediscovery of Panpsychism, in: Skrbina, D. (ed.), Mind that Abides, Amsterdam: Benjamins Publishing, 179-199. Basile, P. (2003): Self and World. The Radical Empiricism of Hume, Bradley and James, in: Bradley Studies 9.2, 93-100. Chalmers, D. (1996): The Conscious Mind. Oxford and New York: Oxford University Press. Descartes, R. (1841/1986): Meditations on First Philosophy, translated by J. Cottingham, Cambridge: Cambridge University Press. Ewing, A. C. (1934): Idealism: A Critical Survey, London: Methuen & Co. Goff, P. (2009b): Can the Panpsychist get Around the Combination Problem?, in: Skrbina, D. (ed.), Mind that Abides, Amsterdam: Benjamins Publishing, 129-36. Griffin, D. R. (1997): Unsnarling the World Knot: Consciousness, Freedom and the Mind-Body Problem, Berkeley: University of California Press. Hosinsky, T. E. (1993): Stubborn Fact and Creative Advance: an Introduction to the Metaphysics of A. N. Whitehead, New York: Rowman & Littlefield. James, W. (1890/1950): The Principles of Psychology, Vol. I, New York: Dover.

1

I wish to thank Michael Blamauer for the invitation to attend the Vienna workshop on panpsychism. I am indebted to David Skrbina, Riccardo Manzotti and all other participants for helpful comments and discussions. In this paper, I have made occasional use of ideas already presented in Basile 2009a, 2009b, 2010 and 2011.

50 James, W. (1909): A Pluralistic Universe: Hibbert Lectures at Manchester College on the Present Situation in Philosophy, London-Bombay-Calcutta: Longmans, Green and Co. Kraus, E. (1998): The Metaphysics of Experience: a Companion to Whitehead’s ‘Process and Reality’, New York: Fordham University Press. Leslie, J. (2001): Infinite Minds. A Philosophical Cosmology, New York: Oxford University Press. Lotze, H. (1885): Microcosmus: An Essay Concerning Man and his Relation to the World, Edinburgh: T. & T. Clark. Madell, G. (2010): The Road to Substance Dualism, in: Basile, P./ Kiverstein, J./Phemister, P. (eds.), The Metaphysics of Consciousness, Cambridge: Cambridge University Press, 45-60. Madell, G. (2007): Timothy Sprigge and Panpsychism, in: Basile, P./ McHenry, L. B. (eds.), Consciousness, Reality and Value: Essays in Honour of T. L. S. Sprigge, Frankfurt: Ontos, 41-52. McGinn, C. (1989): Can We Solve The Mind-Body Problem?, in: Mind 98, 349-66. McHenry, L. B. (2010): Sprigge’s Ontology of Consciousness, in: Basile, P./ Kiverstein, J./Phemister, P. (eds.), The Metaphysics of Consciousness, Cambridge: Cambridge University Press, 5-20. Nagel, T. (1979): Panpsychism, in: Mortal Questions, Cambridge: Cambridge University Press, 181-195. Nagel, T. (1986): The View from Nowhere, New York: Oxford University Press. Rosenthal, D. M. (2006): Experience and the Physical, in: Freeman, A. (ed.), Consciousness and its Place in Nature: does Physicalism entail Panpsychism?, Exeter: Imprint Academic, 117-128. Seager, W. (2006): The ‘Intrinsic Nature’ Argument for Panpsychim, in: Freeman, A. (ed.), Consciousness and its Place in Nature: does Physicalism entail Panpsychism?, Exeter: Imprint Academic, 129-145. Simons, P. (2006): The Seeds of Experience, in: Freeman, A. (ed.), Consciousness and its Place in Nature: does Physicalism entail Panpsychism?, Exeter: Imprint Academic, 146-150. Simons, P. (1998): Metaphysical Systematics: A Lesson from Whitehead, in: Erkenntnis 47, 377-93. Skrbina, D. (2005): Panpsychism in the West, Cambridge, Mass: The MIT Press. Sprigge, T. L. S (1983): The Vindication of Absolute Idealism, Edinburgh: Edinburgh University Press.

51 Strawson, G. (2009): Selves, Oxford: Clarendon Press. Tarde, G. (1893/2009): Monadologie und Soziologie, Frankfurt: Suhrkamp. Whitehead, A. N. (1920): The Concept of Nature, Cambridge: Cambridge University Press. Whitehead, A. N. (1927): Symbolism, New York: Macmillan. Whitehead, A. N. (1929/1978): Process and Reality. An Essay in Cosmology, corrected edition by D. R. Griffin and D. W. Sherburne, New York: Free Press. Whitehead, A. N. (1933/1967): Adventures of Ideas, New York: The Free Press.

The Man Still Stands: Reply to Basile David Skrbina (Dearborn/Michigan) Pierfrancesco Basile makes several excellent points in his rejoinder on the Last Man Standing argument for panpsychism. A few of them call for further comment, and I offer my observations as follows. To begin with, it should be made clear that the Last Man Standing argument (LMS, for short) is but one of 12 arguments for panpsychism that I cited in my book, Panpsychism in the West (2005). Some of these (e.g. Non-Emergence) are actively discussed and debated; others (e.g. Theological) less so. But notable figures throughout history have taken strikingly different paths in arguing for the viability of panpsychism. The LMS, in fact, was explicitly cited only recently, and only by David Ray Griffin (1997: 91)—though it was anticipated by Friedrich Paulsen in 1892. It bears mention due to its recent articulation, but in the larger sweep of things, LMS is perhaps not a central issue. I do endorse it, as Basile states, but I would not call this ‘surprising.’ But I was perhaps unclear in my explanation of panpsychism’s standing vis a vis other contenders. Doubtless, virtually all current philosophers accept some form of physicalism, as both a metaphysical stance and as a theory of mind. Other forms of monism, collectively, take a distant second, and (true, substance) dualism is so far back that to call it ‘third place’ is a gross misnomer. At the beginning of my book, I argue that panpsychism is best thought of as a meta-theory, as a higher-order concept addressing which entities in this universe are enminded. So conceived, it is clear that panpsychism can subsume nearly every conventional theory of mind. One can be a panpsychist materialist, a panpsychist dualist, a panpsychist idealist, a panpsychist functionalist, and so on. Panpsychism does not, per se, offer a theory of mind; it only claims that, however conceived, mind is a characteristic of all extant, concrete individuals. Thus, I was imprecise in my articulation of the LMS argument. I should have stated that of the two historically-significant theories of mind—(1) conventional physicalism, a materialist monism that ascribes mind only to humans and (perhaps) the ‘higher animals’ (whatever they may be), and (2) classical dualism, that posits two fundamental substances of mind and matter, and that (as above) restricts the former only to limited beings—both have major, unresolved problems. Consequently, some form of panpsychism is more plausible than either of these two

54 alternatives. Thus my three contenders are ‘non-panpsychist (we need a new term for this!) physicalism’, ‘non-panpsychist substance dualism’, and ‘panpsychism (of some indeterminate form)’; of these three, the latter is, in my opinion, the LMS. But I don’t think this would satisfy Basile! And rightly so: as he goes on to offer a more detailed critique of panpsychism, and to suggest, in closing, that dualism is not such a bad option after all; more on this in a moment. In his short critique of the ‘intrinsic nature’ argument (a variation on what I have called the Continuity Argument), Basile raises four valid criticisms. I would offer a simplified, three-part critique: (1) why must we postulate an intrinsic nature at all?, (2) if we do, why must it be experiential, or mind-like?, and (3) why must all things share a similar intrinsic nature? In brief response: To deny (1) is arguably a form of eliminativism, and I think we can, with Strawson, plausibly ‘eliminate’ this alternative. Point (2) is subject, itself, to a kind of ‘last man standing’ argument: if the intrinsic nature of things is not mental, then what is it? Mind, experientiality, will—these fall into a category of which we have no better alternative. In the case of that object we know best—ourselves—it is intuitively obvious. Denying point (3) demands virtual solipsism: that we (humans) alone have a mental intrinsic nature, and no, or only few, other things do. Such a standpoint, according to Schopenhauer, is near insanity: “as a serious conviction, it could only be found in a madhouse, and as such it would need not so much a refutation as a cure” (World as Will and Idea, II, 19). In short, the intrinsic/continuity argument has its problems, but they are not insurmountable, nor are they fatal. Of course, it is not an analytic argument—but that is another matter. The Non-Emergence argument for panpsychism, which began all the way back with Epicurus (see my 2005: 52, 53), is, perhaps, “less than conclusive,” but only theoretically. In practice, at present, it is conclusive—because, for whatever reason, we cannot conceive how mind can plausibly emerge from no-mind. To say that ‘someday we may (may!) be able to comprehend this fact’ is to make an arbitrary and irrefutable claim. Strawson’s recent and influential essay “Realistic monism”—reprinted in my 2009 anthology Mind That Abides—offers powerful reasons why emergence is, and will remain, an incomprehensible concept.

55 In his third section Basile comes down hard on the dual-aspect variant of monism, arguing that even here a kind interaction problem reigns, one hardly better than the Cartesian case. But the problem is not so severe with dual-aspectism precisely because it is a form of monism. The interaction problem is at its ugliest with substances; aspects, or dimensions, or perspectives on a unitary reality simply do not interact in the way that substances must. And I would further argue that it is more intuitive that we have dual aspects within ourselves than that we have dual substances. The substances are one possible interpretation of our firsthand experience of the aspects—but only one, and a very problematic one at that. I continue to believe that dual-aspect monism is a most promising way forward. Two final points bear mention: Near the end of section three, Basile states that “no panpsychist has yet been capable of providing an explanation of how human mentality is generated from the several mentalities in the neurons”—in other words, no one has yet solved the combination problem. True enough. This is an important area of research for all those who wish to defend panpsychism. I myself have put forth some tentative ideas along this line, in my chapter “Minds, objects, and relations”, in Mind That Abides. I think that a ‘dynamic systems’ approach is very promising, and that we should view the lesser minds in our bodies as operating in a sub-space of the larger mental space that corresponds with our bodies. Theoretical physics works very well with such a concept: parts of objects change dynamically in a ‘state space’ that is a portion of, but of lesser dimension, than the state space of the whole object. In a similar way, the ‘mind space’ of our atoms, cells, or organs can coexist quite nicely within the larger mental space of our brain or (better still) entire body. But this is just the beginnings of a possible solution. Finally, Basile concludes that panpsychism’s standing is scarcely better than even Cartesian dualism. Citing both our pre-philosophical intuition and a recent statement by Madell, he suggests that it is not unreasonable to argue that dualism, not panpsychism, is the LMS. Of the notorious interaction problem, Basile and Madell opine that substance interaction is, like brute emergence, not theoretically impossible. But of course the Cartesian form of dualism has both problems: interaction, and emergence in humans alone. It has both the substance dualism and a kind of phylogenic dualism between humans and all other beings. (It is a kind of ‘double dualism,’ as I tell my students.) But there is an

56 obvious, if partial, solution here: eliminate the emergence, and allow all beings a dual substance of mind and matter—in other words, a panpsychist dualism. Now there’s an interesting position! I think that could make for some very interesting speculation. So is panpsychism ‘terminally unacceptable,’ to take Basile’s pessimistic inference? I think not, if only because of its meta-theoretic status. Panpsychism is simply a universal application of a given theory of mind. The argument for universalism is quite strong, given the grave weaknesses in mental emergence. Apart from this, any ‘panpsychist’ theory of mind is only as strong or as weak as the underlying theory itself, whether monist or dualist, idealist or materialist, and so on. Critics of panpsychism: bear this in mind!

“Mind Matters...” – Towards a concept of proto-mental causation Ludwig Jaskolla (Munich) “How wonderful! That even The passions, prejudices and interests, That sway the meanest thing, the weak touch That moves the finest nerve, And in one human brain Causes the faintest thought, becomes a link In the great chain of nature.” Percy Bysshe Shelley Queen Mab, Canto II, 102-108 Chapter 1: Basic Distinctions Causation is among the central, conceptual minefields of modern, analytic philosophy of mind. The essential characteristics of this concept are notoriously obscure in many debates. And this again is especially true for the philosophical discussion of mental causation. The long-term objective of my paper (cf. Chapter 4) is going to be a systematic sketch of a positive theory of protomental causation within the framework of an overall panexperientialistic ontology. Nevertheless, in a first step (cf. Chapter 1), I am going to provide some basic conceptual definitions that are important for the main goal of my paper. Due to the complexity of the problem at hand (cf. Chapter 2), I will only line out two basic, conceptual problems of classical approaches to mental causation – and these only as far as they contribute my overall goal. These basic distinctions and problems will be used (cf. Chapter 3) in order to define the conceptual environment in which mental causation is possible. But first things first: The concepts “Causation” and “Physical Worlds” are often used uncritically within analytic philosophy of mind. This is why, I am going to present my understanding of these concepts. Causation: My understanding of causation is oriented towards our ordinary use of this concept as well as towards some intuitions, I consider to be basic concepts mapping the world reasonably adequate. Within ordinary language, we form sentences of the following structure.

58 A The baseball destroyed the window. B If you step onto the frozen surface of the lake, it will crack. C This key does not fit into the keyhole, because it is not sculpted in the right way. D This decision is going to hunt me for the rest of my life. These sentences share a structural common element: They contextually refer to the fact that a certain final state x2 is brought about by an initial state x1. I consider the paradigm of 'bringing about' to be the smallest common denominator of all the different ordinary language talk about causation. It seems to me that our intuitive approach towards causation is of just this kind: We want to understand under what circumstances a certain final state is brought about by an initial state. I am going to argue for the following first stipulative definition: Definition 1.1 (Causal Relevance) A state x1 is considered to be causally relevant to a state x2, iff. x2 is brought about by x1. Three short annotations: (a) If one was to name the “mental father” of this approach, it would have to be Aristotle: In De Generatione Animalium he puts forth a concept of causation that is structurally oriented to the paradigm of artisanal production (cf. Aristotle, 715a4-7): This is why, it is not surprising that the aristotelian causes (i.e. efficient and final cause, as well as material and formal cause) depict a rather liberal concept of causation. At this moment, I want to emphasize that there are strong similarities between my definition 1.1 and the aristotelian concept of causation: Sentence A denotes an example of an efficient cause – the kind of causation we are familiar with from physics. B denotes a material cause, because the frozen surface resists weight up to a certain critical amount. C denotes a formal cause, because certain effects are only brought about under specific structural circumstances. At last, D denotes a final cause. I would like to point out that the preceding considerations exemplify the overall spectrum of a liberal account of causation – I do not argue that they are exhaustive and/or minimal. (b) One central concern of this liberal account of causation is to map as many of our ordinary intuitions about causation as possible. Both the

59 concept of physical causation as well as a specific account of human causation (cf. mental causation) can be mapped within the framework of this liberal approach. Nevertheless, one needs to concede that this liberal account is at this point not yet determinate enough to pose a full-blown philosophical theory of causation. This is why, it will be necessary to specify this account during the next considerations in order to give a detailed account of how the different causal contributions within the causal framework of the world can be mapped adequately. (c) A last annotation: One critical concept of the modern philosophy of mental causation is the claim that complete causal explanations are exhaustive – e.g. that they tell the whole causal story of a respective causal process. Following Jaegwon Kim (Kim 2002, 174), I call this the principle of the exclusiveness of complete causal explanations (PEC). And I emphasize that a causally relevant state (in terms of 1.1) does not entail that the initial state is a complete causal explanation in the sense of PEC. Physical World: Contrary to the notion of causation, the concept of the physical world seems to be at least reasonably clear within the systemic framework of modern philosophy. Following the extraordinary success of the quantitative explanations of physics since mid-19th-century, most philosophers consider the physical world to be the part of reality which is described by physics: This means that an entity will be part of the physical world, if and only if this entity is described by the differential equations of modern theoretical physics (or is at least reducible to the entities described by these equations). Again, I am going to propose a liberal account of the physical world: Definition 1.2 (Physical World) The physical world consists of the set of non-supernatural experiences arranged within the ordinary spectrum of human experiences. Further, two short annotations: (d) Unfortunately, 1.2 is a rather bad example of a definition, because it poses an iridescent concept of the natural world. At the end of the day, I am arguing that most people have a rather clear concept of what it means to be a natural phenomenon (as well as what not). Among the experiences of natural phenomena count artifacts, animals, human persons or societies, but also the causal relevance human consciousness. It is important to note

60 that the set of entities described by physics (or the world of physics) is a subset of the physical world as defined above. Wether the world of physics is a proper subset of the physical world is left open at this point. (e) I want to mention at least two philosophers that have influenced my account of the physical world: David Hume argues in his Treatise on Human Nature that our theories about the nature of the world are essentially dependent upon the structure of experience. He mentions that to first approximation different experiences are on par – this would imply that there is no reason to consider the existence of bodies as better substantiated than the existence of the mind (cf. Hume Treatise, 5). A similar account has been put forth by Bertrand Russell in his Our Knowledge of the External World. Russell considers experiences to be the ontologically most basic entities, and argues that external to the framework of experiences no substantiated knowledge about the world can be stated. Both approaches are structurally extremely similar to the account given above in 1.2. Chapter 2: A Classification of the Mind-Body-Problem in Terms of Causation Godehard Brüntrup within the scope of his dissertation Mental Causation developed a trilemma in order to give a systematic mapping of the modern, analytic philosophy of mind. His central conceptual concern comes down to the following question: What modes are there to map the causal relevance of the mental with respect to the physical? A philosophical trilemma consists of three different theses. Within the framework of a trilemma only two of the thesis can hold true simultaneously. This entails that a trilemma can be resolved in three distinct ways – by negating one of the respective thesis – rendering the remaining two thesis a consistent system of philosophical ideas. Here are our three “causal thesis” about the nature of the mental. Thesis [1]: The physical world is causally closed. Corollary Thesis [1*]: The world of physics is causally closed. Thesis [2]: IF The physical world is causally closed. THEN The mental is causally irrelevant. Thesis [3]: The mental is causally relevant. Negating thesis [3] does not seem to be a very promising road to go, if one wants to ensure the causal relevance of the mental. This means that thesis [2] and [1] or [1*] remain. In the following considerations, I am going to

61 sketch these approaches and I am going to show that neither nonreductive physicalism (non-[2]) nor dualism (non-[1]) are able to give intelligible accounts of mental causation Negating [2]: Nonreductive physicalists embrace to two substantial proposition: 'The physical world is causally closed' and 'The mental is causally relevant'. This entails that nonreductive physicalists need to negate the conditional thesis [2] telling that by logical necessity [1*] implies non[3]. They make the interesting move of trying to persuade us that we are mistaken assuming that the relation between the two substantial thesis [1*] and [3] is a logical subjunction. Brüntrup argued 1995 that the following characteristics hold true for every version of nonreductive physicalism: (a) On the one hand, nonreductive physicalists are (or rather want to be) metaphysical realists concerning mental causation. This is due to the following argument: If you want to secure the causal relevance of the mental and you interpret physical causation realistically – you'd better interpret mental causation also realistically (otherwise you will fall prey to epiphenomenalism). (b) On the other hand, nonreductive physicalists need to construe some kind of ontological dependence rendering the physical more fundamental than the mental, because at the end of the day nonreductive physicalists are still physicalists. Nonreductive physicalism is served in three different flavors: Emergence Theory: Like causation, emergence counts among the most opaque concepts in modern philosophy. Basically, every theory of emergence puts forth a two-layered theory of reality – I am going to argue for the following simple definition: Definition 2.1 (Emergence) 'A' emerges from 'B', iff. (i) A and B belong to distinct ontological categories, (ii) no bridging laws hold between A and B and (iii) the occurrence of A cannot be deduced from B in advance. Timothy O'Connor argued 2002 that emergence should be understood as causally relevant, non-supervenient relation between to levels of reality. At first hand, the causal relevance seems to be unproblematic, because the occurrence of level A is dependent upon the existence of B. This is why, O'Connor argues: “the emergent state {is} a 'causal consequence' of the

62 object's complex configuration” (O'Connor 2000, 8). His approach to mental causation is dependent on his theory of the physical: Not all microphysical states are completely determinate. This why emergent, mental states – via downward causation – are able to act as determinants of physical states. This is one emergentists way of 'securing' the causal relevance of the mental. Nevertheless, O'Connors theory bears the foul stench of the metaphysical laboratory. Is it really possible to combine this kind of downward causation with a causally closed world of physics? Causal closure of the world physics means that there is no causal interaction between the domain described by physics and the nonphysical (cf. Montero, 2, Definition TCC). To me it seems obvious that there is a contradiction within the framework of emergence, because the postulation of independent downward causation conflicts with [1*]. If taken seriously, the notion of downward causation drives the theory of emergence towards substance dualism - rendering it inconsistent in respect of being some kind of nonreductive physicalism. Supervenience Theory: Supervenience is another way to describe the dependence between the mental and the physical. Jaegwon Kim defined supervenience as a specific version of mathematical covariance. This is because (i) supervenience is a modal version of covariance and (ii) because supervenience makes a statement about which level of reality is more fundamental – to put this simple: 'supervenience is covariance + ontology'. Here is my definition: Definition 2.2 (Supervenience) Properties of kind A supervene strongly on properties of the kind B, iff. ∀w1, w2 ∀x∈w1, y∈w2 : (IF x is A-indiscernible from y THEN x is Bindiscernible from y), where w1, w2 are possible worlds and x,y are individuals. This implies that a certain configuration of A-properties entails necessarily a specific configuration of B-properties, because these configurations occur together through all possible worlds. Nevertheless, supervenience is asymmetric: Sure, certain A-configurations are entailed by certain Bconfiguration, but in different possible worlds the same A-configuration could be realized (multiply) by different B-configurations.

63 Proponents of supervenience argue that the causal relevance of the mental is guaranteed by the physical realization, but complete reduction is circumvented by multiple realization. Is this approach convincing? I say: not at all. Supervenience implies that complete translation of the causal role of the mental in terms of the causal role of the physical basis is possible, because the truth-values of every causal proposition is fixed by the physical basis alone (cf. Brüntrup 1995, 211f). At the end of the day, supervenience seems to lead to causal epiphenomenalism. Tokenidentity Theory: The last approach, I want to discuss in this paper is Donald Davidson's proposal to model the causal mind-body-nexus as a threefold structure based on the distinction between token- and typeidentity. In a nut-shell his argument runs like this: A mental event is type-identical with functional state – this means that every property of the mental state has to be a property of the functional state (and vice versa). The functional state poses a complete description of the mental state in terms of the causal role the mental state fulfills within our ordinary approach to reality. The functional state itself is tokenidentical with a physical event: Implying that every occurrence of a functional state is identical with exactly one physical event. Nevertheless, this does not entail that a certain occurrence of a functional state has to be identical with the same kind of physical event through all possible worlds. Again, this can be interpreted in terms of multiple realization of the mental state. Proponents of token-identity argue that the causal relevance of the mental is secured, because every mental event is identical with a physical event, showing the intellectual proximity of token-identity to reductive physicalism. Still, it puts forth that the distinction between types and tokens entails nonreductionism and therefore secures causal significance of the mental. To me this seems like a bad magic trick: I argue with Honderich (1982) that within the framework of token-identity the causality of a mental event can be adequately described as the disjunction of the causal roles of all physical realization of this event. And this seems to imply that the causal relevance of the mental is at best epiphenomenal – this means also proponents of token-identiy have a hard time securing the causal independence of the mental. Negating [1]: The second, classical way of trying to secure the causal relevance of the mental, is to insist that the physical world is not causally

64 closed. Albeit, I will discuss dualistic ontologies in detail, it is interesting to remark that dualists negate thesis [1] – not thesis [1*]. Interactionistic dualists argue that there are equally real domains of reality which are completely categorically separate. In particular this means: IF Q is an essential property of the first domain, THEN Q is not a property of the second domain – yet it is put forth, that these domains are able to interact causally. The ontological independence of the two domains is guaranteed via so called 'Detachability-Arguments' deducing the ontological independence from the conceivability that the one domain could exists without the other. For the actual world, these two domains are the mental and the physical, which are intimately tied together via causal interaction. Now, the interesting question seems to be: How do interactionistic dualists make sense of this causal interaction? It is argued that mental causation is structurally very similar to physical causation. One mental event can surely bring about another mental event, but one mental event can also bring about a physical event, says the dualist. Nevertheless, how this mechanism exactly works is a contingent fact of the respective actual world – for example, Descartes proposed that the epiphysis acts as the causal nexus between the mental and the physical. Though, one pressing question remains: how on earth should the causal interaction bridge over the categorical chasm between the mental and the physical? I argue that the dualist is not able to provide a systematic criterion in order to transfer causal powers from one domain to the other (lack of a tertium comparationis). At the end of the day, this results from the fact that the dualist takes the 'the negation of [1]' in a literal sense: If the physical is completely described by physics and the mental is real as well as causally relevant, we need to assume that there are two ontologically complete distinct domains of reality (cf. Zimmerman 2006, 113-122). In the course of this argument, I want to stop my analysis with this intuitve problem of interactionistic dualism. Synopsis: My goal during the last considerations was to 'show' – as far as this is possible within the range of this paper – that most modern approaches trying secure the causal relevance of the mental fail. If you take a close look at the debate all theories sketched above have one common structural element: Nonreductive physicalists as well as dualists try to model mental causation within the paradigm of physical causation – structurally physical causation can be described via the 'Billiard Ball Universe': A physical event – the trajectory of one billiard ball – causes

65 another physical event – the changed trajectory of another – via some kind of efficient causation. Now, I argued that this approach leads either to causal epiphenomenalism or to causal elimination. And again this fact is intimately tied the the principle PEC (sketched above): If one complete causal explanation of a certain kind (like in physics) is at hand, then providing another causal explanation of the same kind seems to be senseless. To me two different approaches to this problem seem to be promising: On the one hand, you could just accept that the causal significance of the mental is to be reduced to the physical. On the other hand, you could try to find a 'natural place' for the mental within the causal framework of the world that is not conflicting with the causal talk of the physical. I am going to opt for the second approach – putting forth the following inquiry of philosophical theory formation: Is the concept 'causal role' univocal in terms of the concepts mental and physical? In the course of the following considerations, I am going to sketch a theory of protomental causation dealing explicitly with this inquiry. Chapter 3: Preliminaries of a Panexperiential Theory of Mental Causation Causal Closure of Physics? A philosophical theory of mental causation needs to find an independent, systematic place for the causal relevance of the mental in the world. It seems too limited to just search for gaps within the theoretical framework of physics. Nevertheless, let's take another look at thesis [1*]: The world of physics is causally closed; i.e. there is no causal contribution of entities that are not described by physics. But, is there reason to believe that this thesis is entailed by theoretical physics? I do not think that there is. Physics as theoretical science is occupied with mathematical models of reality. If we take for example a close look at physics best theory of reality – quantum mechanics and quantum field theory: The von Neumann formalism of quantum mechanics describes states of reality as observables. Observables are self-adjunct operators within a Hilbert space. The spatio-temporal development of

66 systems of observables can be adequately described by differential equations – for example the Schrödinger equation for non-relativistic quantum systems. These describe the states of a system in terms of spatiotemporal tensors. Now, the important lesson is that quantum mechanics only correlates spatio-temporal tensors and states – i.e. these equations assert nothing more than that some initial state x1 at t1 is followed by some state x2 at t2 with a certain probability. Similar considerations can be carried out for the Hamilton formalism describing the temporal development of physical fields within the framework of quantum field theory. At the end of the day, I argue that the mathematical formalism behind quantum mechanics and quantum field theory is complete indeterminate considering the question wether these correlations are interpreted to be strict causal relations or mere regularities (in the sense of Hume). Yet, what is the sequitur for our inquiry? I argue as cautious as possible for the following assertion: Theoretical physics does not necessarily entail strict causal laws. During the last years, philosophers as well as physicists have opted for similar interpretations. Mentioning only two eminent examples: Henry Stapp in his 2009 'Quantum Interactive Dualism' as well as Gregg Rosenberg in his 2004 'A Place for Consciousness' chapter 9.3. I want to carry on this train of thought: 1. Theoretical Physics does not necessarily entail strict causal laws. 2. IF the world of physics is causally closed, THEN it is necessarily causally closed. 3. IF the world of physics is necessarily causally closed, THEN there are necessarily strict causal laws. 4. **The world of physics is causally closed** 5. There are necessarily strict causal laws. One can easily see that there is a conceptual conflict between proposition 1 and 5. I argue that the claim for causal closure of the physical is indeed stronger than what can be deduced from theoretical physics, because I showed earlier that our most basic physical theories do not necessarily entail strict causal laws. I want to defend the thesis that there is good reason to believe one is not only able to negate [1*], but that [1*] does not even follow from physics.

67 Let's take a look at this analysis from a different angle: Are these results surprising for physicist? I don't think so. Most physicists are well aware that the set of differential equations they employ to describe the world is nothing more than a functional representation – yet a very precise one – describing the physical world incompletely. And this is also precisely the way, I understand the second intuitive definition from chapter 1: The domain of the non-supernatural experiences is greater than the domain that is completely described by physics. Mental Causation (Step 1): At this point, I want wrap up my preceding analysis in order to provide a first, positive concept of mental causation. I am going to argue for the following thesis: Besides the well-known causally relevant events of physics, there is the possibility of other forms of natural, nevertheless causally relevant events. I need to mention one important aspect of this thesis: Contrary to other for example interactionistic dualism, my approach does not imply that the 'other forms' of causality are not physical in the sense that they are not part of the natural world. My thesis postulates merely (i) that the complete causal framework of the world is not entirely described by physics and (ii) that other causal contributions do not conflict with the causal story told by physics. Within this context, I consider my approach to be naturalism at its best. In the course of the following considerations, I am going to present a first positive characterization: Definition 3.1 (Mental Causation [intuitive]) Physical causation can be understood (structurally) adequately by the paradigm of 'pushing'– whereas mental causation can be understood by the paradigm of 'deciding'. Physical causation is best described as the direct 'bringing-about' of one event by another event. Again, we can make use of the billiard-ball-model that has been developed above – the cause brings about the effect, like the first ball hits and pushes the second. Essential aspects of this kind of causation are: (1) Physical causation is always localized exactly in spatiotemporal terms. This means that cause and effect can be modeled (structurally) by some tensor consisting of three spatial coordinates and

68 one temporal coordinate. (2) Within the framework physical causation, cause and effect are always arranged temporally linear. Implying that physical causation is temporally asymmetric – the cause occurs always at least one instant before the effect. But how about mental causation? I will be relying mainly on two philosophical intuitions: At the beginning of the 20th century, Alfred North Whitehead realized that mental causation has to be of very different kind compared to physical causation. He argued that our ordinary conception of mental causation holds that the mental is causally relevant by restricting the possible development of events via formal limitations in a way that certain effects are favored while other are cut off. Whitehead summarized the setting of these limitations via the paradigm of 'evaluating' – which I will call 'deciding'. My intention will be to extend one of Dean Zimmerman's intuitions in his Immanent Causation: Physical causation is always dependent on certain formal limitations clarifying under which conditions which effects are brought about. At this point, I am going adopt Zimmerman's view – arguing that mental causation is merely a specific form of setting these limitations. Of course, the idea sketched above already a very technical concept mental causation – and this is why, I want to explore this notion within the context of a rather simple example: “I'd rather have coffee than tea with my pie” If you were to analyze this decision, probably the first thing you would remark is that this decision is already a decision within a very specific context: Now, this very decision provides a formal limitation of its environment – ideally causing that the person will get coffee with her pie. Besides this rather vague substantial criterion, some essential characteristics of mental causation are: (1) Mental causation is always localized exactly in spatio-temporal terms. This means that cause and effect can be modeled (structurally) by some tensor consisting of three spatial coordinates and one temporal coordinate. (2) Mental causation sets formal limitations – cause and effect are (strictly speaking) synchronous. This kind of synchronous 'top-down-causation' is the most important line of demarcation between physical and mental causation. Within the following considerations, I am going to provide a systematic sketch of one positive of proto-mental causation.

69 Chapter 4: A Positive Theory of Proto-Mental Causation Panexperientialism, Naturalism and the Causal Relevance of the Mental: During the preceding considerations, I argued that the negation of thesis [1*] poses the possibility of being able to provide a natural place for the mental within the causal framework of the world. One central claim was: You need to drop the causal closure of physics as well as the thesis that the physical describes the natural world completely in order to secure the causal relevance of the mental. Bottomline: I am aiming at a theory of the natural world synthesizing physicalistic as well as dualistic intuitions. Panexperientialism (one particular fashion of panpsychism) meets these criteria: It secures the causal relevance of the mental without tearing it out of the natural world completely, by arguing that some kind of mentality is an essential aspect of every state of reality – complex or simple – organic or inorganic. It has been argued that panexperientialism is in fact the only true naturalism, because it renders the mental to be just one phenomenon among the other basic characteristics of the natural world – nothing special, like in non-reductive physicalism or interactionistic dualism. A few paragraphs ago, I argued that panexperientialism matches beautifully the demands of a realistic account of mental causation: Within a concept of the world that renders the mental as fundamental, it seems to be implausible to assert that the one science not describing the mental as well as the causal relevance of the mental (i.e. physics) has among its essential characteristics the necessary causal closure of physics as well as the necessary completeness of physics. The foregoing argument traces its roots back to the definitions 1.1 and 1.2 giving a very liberal account of the natural world that adjusts to our ordinary, everyday concept of the world. I want to refer back to Gregg Rosenberg who argued in his 2004 A Place for Consciousness that panexperientialism should be called 'Liberal Naturalism' (cf. Rosenberg 2004: 76f). This seems to be the very essence of panexperientialism in conceptual terms. A Systematic Sketch of ONE Panexeperientialistc Ontology: Since the different, competing of accounts of panexperientialism are legion, I want to give a short introduction on what I consider to be the fundamental aspects of viable notion of panexperientialism. My version of panexperientialism traces back to Alfred North Whitehead's Process and Reality as well as Galen Strawson's Consciousness and its Place in Nature (and to some extent to Selves). I define panexperientialism to be the following thesis:

70 Definition 4.1 (Panexperientialism) Panexperientialism is the thesis that the mental-experiential and the physical are ontologically on equal footing within the basic aspects of every fundamental state of reality. The mental-experiential aspects are the intrinsic carriers of the states, whereas the physical aspects pose the objectified outsides of the states. In a nut-shell, I assume that every micro-event described by physics has some features that can be in loose way of speaking be described as 'experiential'. Within the context of this analysis, I will not be able to give positive arguments (for a profound introduction to these arguments, please refer to chapter 8 of Godehard Brüntrup's Das Leib-Seele-Problem) why it seems reasonable to make this 'panexperiential move' – I am just going to assume it as viable theory within the context of the modern, analytic philosophy of mind. Nevertheless, I need to give one additional remark: I will be making use of the prefix 'proto-X' a lot in the following considerations and mostly within the context of anthropomorphic concepts like 'experience'. The interesting thing about these concepts is that they are almost exclusively used to describe features of humans. Nevertheless, there is no intrinsic reason to believe that these concepts are solely applicable to humans. I am going to use the prefix 'proto-X' in order to describe some concept that (i) bears substantial similarities to the human concept 'X' but that nevertheless (ii) is not identical to 'X' in structural terms. To give one simple example: One could argue that a slide rule is a 'proto-computer', because the slide rule is very similar to the computer – both are used to solve problems relying on binary arithmetic operations - but the slide rule is because of its very limited possibilities no modern computer. So every time I am going to use some anthropomorphic concept in the following considerations, it is going to be an acronym for the respective proto-concept – for important exceptions; I am going to use the prefix. Here are some basic definitions: Proto-Perspectives: Definition 4.1 raises a first question – we need to inquire what the mental-experiential aspect is that poses the intrinsic carrier of every micro-event? I am going to argue that this mentalexperiential of a micro-event is the phenomenal perspective of the event upon itself as well as its environment: Every event bears some protoexperience of what it means and how it feels to be this particular event.

71 Rephrasing Thomas Nagel, one could argue that intrinsic natures of events are essentially 'views from somewhere'. Described in structural terms, this intrinsic nature is best understood as a quale. These qualia are individual – entailing that they are not multiple realizable. Nevertheless, I take qualia to be able to stand in various similarity-relations to other qualia: For example, the qualia of two different electrons Qe,1 and Qe,2 are much more similar to each other than to the quale of a proton Qp – this observation about greater or smaller similarity rests upon the fact that electrons (compared to protons) behave very similar in physical terms. To give a full review of definition 4.1, we need to clarify two subsequent concepts. Proto-Representation: The theory of perspectives as qualia told us something about the exterior characteristics of the mental-experiential – nevertheless, I need to give an account of the interior substantial structure of the mental-experiential in order to make my account of panexperientialism intelligible. (1) One of the central (inner) characteristics of a protoperspective is that it represents its environment. Implying that it is essential for a perspective to have some (maybe crude) idea of the circumstances it is situated in. These circumstances can be specified via two features: (1-a) The past of the state (complete or partial) is present within the state in terms of fixed, causal relations the state bears to its predecessors as well as the past environment of the predecessorstates. (1-b) Giving an account of the representation of the state's future is more complicate: We need to discern certain equivalence-relation. For the present, I am referring to the theory of abstraction by Bob Hale and Crispin Wright (an extensive summary can be found in Hale/Wright 2009). They argue that we are able to pick out and define abstract entities by providing equivalence-relations – for example Hume's famous definition: The quantum of elements A = The quantum of elements in B, iff. There is a f:A→B such that for all a A there exists some b B mit f(a)=b, and vice versa. In this special case, the equivalence relation in question is defined by the bijective function f which assigns to every element of the first set exactly on element of the second set – and again vice versa. Now, let's go back to our proto-perspective. Have a look at the set of propositions describing all the logically possible states of the perspective E at one point of space-time: We can give a structure of this set which discerns the different logically possible developments of E in its future. For example:

72 X is part of one possible future W1 of E, iff. X, all proposition of W1 and all propositions of E's past form a maximally-consistent system. Basically, this is nothing more than the abstract description of a possible world W1, which can be adequately characterized as one possible development of the state E originating from E's (fixed) past. Within this context, we are able to talk about a perspective representing its future: The set of the possible developments W1...n that are consistent with E's past, is used in order to define E's future (for some point of space-time). I want to propose a certain limitation of this principle: The complexity of the associated abstractions is directly, positively correlated with the mental complexity of E – implying that the representation of past and future for micro-events is supposed to be very simple. In his 2008 Structure and Being, Lorenz B. Puntel argued that the structure described above can be adequately stated as the ability of a perspective to integrate its environment as well as past and future within one intensionalcoextensional act of unity (cf. Puntel 2008, 375). Proto-Intentionality: (2) The second central (inner) characteristic of a perspective is that perspectives are of that kind that they evaluate their environment – i.e. representation of past and future, as well as the actual environment – from their particular point of view. Implying the following: A perspective is able to attribute qualitative characters QW,1...n to its possible developments W1...n – that means that the perspective is representing its environment as well as evaluating it via attributing some specific quale. Within the context of 'proto-intentional evaluations' the process can be pictured in a rather simplistic fashion: 'W1 is better than W2'. These comparisons provide an external description of what it means to assign one particular quale to a possible world W. A Panexperiential Theory of Mental Causation: Within the context of the proposed panexperiential ontology, we need re-evaluate our understanding of mental causation layed out in chapter 3. I argued that every kind of physical causation needs some kind of formal or structural element in order to bundle and route the causal power of efficient causation. Furthermore, I argued in chapter 3 that mental causation is some special occurrence of these formal constraints. Now, panexperientialism puts forth the thesis that strictly speaking every physical event bears mental characteristics. And this is why, I need to re-adjust my foregoing considerations:

73 Definition: 4.2 (Protomental Causation [Step 1]) The formal constraint within the causal net of micro-physical events is provided by the protomental aspect of the respective event. On the one hand, these results are not exceedingly surprising, because if you consider the mental as a fundamental aspect of every state of reality, it makes sense to consider these aspects to be causally relevant. But on the other hand, there are some rather severe misconceptions luring in the background of 4.2: Query (1) – Isn't it weird that protomental causation does not make any difference in the causal framework of physics? No, on the contrary within he context of a panexperiential ontology this is among the results you would expect. Among my central points was that the protomental is similar to the human mental, nevertheless I argued that it is much simpler in structural terms. One would expect that a protomental event has only a very limited scope of action in terms of its rather simple proto-representation and protointentionality – furthermore, this scope of action is so exceedingly small that it diminishes in the differential equations of physics. Query (2) – Why is there no 'mentality operator' within the differential equations to render the laws of physics a complete description of the world? Well, this is a hard question that is intimately tied to the concept of a law of nature. In my opinion, laws of nature are theoretical generalizations mapping our best systematizations of the empirical facts (basically, this is very similar to David Lewis' proposal from 1994). Nevertheless, I am going to extend Lewis' thesis by advancing Nancy Cartwright's claim that even at the very basis of physics there are no laws of nature in terms of universally valid links (cf. Cartwright 1983). At the end of day, I am going to opt for the following, substantial thesis: Structural features that are interpreted as laws of nature within the context of physics, are nothing more than the abstracted protomental-behaviour patterns of micro-events behaving extremely predicable due to lack of complexity. In order to describe a protomental causation adequately, I am going to define three processes partitioning the causal significance of the protomental apsect of a micro-experiential act of causation. Nevertheless,

74 one short preliminary remark: I am going to limit my approach to protomental causation because one would need to solve the combination problem in order to give a detailed account of mental causation (of human level). Since I do not have such a solution at hand, I need to make this restriction. Definition 4.3 (Protomental Causation [Process 1]) A micro-event Et (at point of time t) is located (i) within an environment of events P1 … Pm and (ii) within the context of its past Et-1 … Et-n. The event Et represents environmental influences as well the constraints of the past via phenomenal content, AND abstracts a future consisting of its possible strains of development {W1 … Wx}. This definition contains two assumptions: Firstly, it is implied that E's past (i.e. Et-1 … Et-n) remains existent in some substantial way, because they need to contribute to specific way the event E experiences its environment. Loosely, I am going to address this assumption as some kind of pasteternalism. Nevertheless, I want to point out that past-eternalism does not imply that the past events are equally existent compared to the actual now. Furthermore, I think that there is no reason to believe past-eternalism entails either an a-theory or a b-theory of time. Via the means that have been sketched above, the micro-event generates qualitative representations as well as the representation of a future. This representation of the future seems to imply that the future is itself is taken to be non-existent, otherwise E would be able to just perceive its future directly. This is why, I am calling this position future-presentism. To illustrate my approach, I am going to make use of rather anthropomorphic examples in the following considerations. Nevertheless, I take it as a matter of fact that some non-anthropomorphic reformulation within the framework of proto-concepts is possible. This is why, I am arguing that these examples should be understood in purely illustrative terms: … In order to reach a decision, I am representing the way back from university and marking this way as 'unpleasant', because a deep depression is just passing over Bavaria. I am than abstracting different possible ways of getting back home – for example walking or taking the tube …

75 This example leads over to the definition of the second process: Definition 4.4 Protomental Causation [Process 2]) The possible developments W1 … Wx of the event Et are, after process 1, evaluated by Et one by one and marked with specific phenomenal content. These contents form a hierarchy mapping the weighted structure of possible developments of the event. Evaluation as well as phenomenal marking are achieved via the protointentional aspect of the micro-event. The 'weighted hierarchy' itself is accounted for by means of the comparison of the qualitative markings of the different possible developments. This is done via propositions of the following kind: 'Wq poses a better possible future for Et than Wp.' Let's have look at a system of three different possible developments {W1 W2 W3} as well as following 'weighing' of this future – 'W2 is better that W1', 'W2 is better than W3'. This system is still incomplete, nevertheless one is able make the assertion that W2 is the possible development of the event that is mostly likely the one that is going to be realized. … I need to decide between going home by foot or taking the tube. According to the circumstances sketched above, I am going to evaluate the way home by tube as better than walking … One classical account of such hierarchies can be found withing the framework pf physics: Under ideal circumstances physical states strive for certain conditions that are maximally fitting in energetic terms. Implying that the realization of different (not as fitting) states, is going to require a very high amount of energy in order to achieve and maintain the (nonfitting) state. And this, among other reasons, shows how nicely the proposed account fits with the causal framework of physics: The theory of protomental causation does not alter the description of physics, but provides another ontological carrier. Definition 4.5 (Protomental Causation [Process 3]) Linear-structured physical causation is bundled by Et's developmenthierarchy in exactly such a way that a certain successor-event Et+1 is generated, where (i) Et+1's immediate past consists of Et. (ii) Et+1 realizes the hierarchical highest possible development of Et either (ii-

76 a) partial or (ii-b) complete. (iii) Et transmits its characteristic quale unto Et+1. The central feature of this process consists in the insight that we need to discard the concept of a purely formal description of the protomental aspect of causation: A purely formal constraint is not able to bring about anything – the constraints need activity. With this in mind, I am going to discard the concept of a complete internal description of protomental causation. The initial- and the successor-event are bound together in such a way that the initial-event is the first event of the past of the successorevent. The hierarchy establishes the overall structure of the successorevent, nevertheless the development of the events in Et's environment could alter the realization of the hierarchy. And this is why, I am going leave open wether a complete realization is achieved or only a partial one. … Well, I am going to decide to take the tube to get back home. To realize this goal, I am going to initiate different processes in order to bring about my desired sequitur. I am walking to the station, get on the train, and so on. Constraints from my environment could, for example, consist in a rail-malfunction that forces me to take a different route than fastest one … One concluding annotation concerning this process: Strictly speaking, the transmission of the characteristic quale from the initial-event unto the successor-event is not among the essential characteristics of protomental causation. Nevertheless, this transmission of this quale can be mapped ontologically very easy within the framework of a panexperiential ontology, and it makes sense, because it comes in very handy when trying to give an account of the persistence-conditions of micro-events. Definition 4.6 (Protomental Causation [Process 4]) Process 4 is identical in structural terms to Process 1; Within the framework of these three (or rather four) processes, one is able to describe protomental causation in following way: Definition 4.7 (Protomental Causation [Step 2]) A microphysical process of causation can be adequately described as:

77 … Process 3' → [ Process 1'' → Process 2'' → Process 3'' ] → Process 1''' → … where the bold words denote what we would typically call one, distinct act of causation – different times are denoted by simple quotation marks. During the course of the present analysis, I tried to provide a positive concept of protomental causation: The initial question was posed by the severe structural problems of classical 'solution' trying to solve the problem of mental causation. This critique lead over to a very general and intuitive notion of mental causation. Within the framework of a panexperiential ontology, I argued that the mental is causally relevant in the world because it provides the formal constraints necessary for every act of causation. I argued that definition 4.7 puts forth a concept of causation that (i) secures the causal relevance of the mental and (ii) fits nicely with the already developed framework of physical causes. I am pretty aware that there are a lot of loose ends in my approach – among the most pressing questions are: (i) What is human mental causation within a panexperiential ontology that is able to solve the combination problem? As well as: (ii) Does the assumption that 4.7 fits seamlessly with the causal framework of physics really hold true? Just one little meta-philosophical remark in order to sketch the significance of the proposed approach. A discussion with John Donaldson at the University of Edinburgh culminated into the following demand: Just give me one genuine causal rôle of the mental that is not already provided by the physical. An answer from my point of view could like this: What if the physical would loose its causal powers, if it were loose the protomental aspect first? Maybe all the time the question was ill-formulated … maybe there is no reason to tinker the mental into the causal framework of the physical … maybe … References Aristoteles: De Generatione Animalium, translated into English by Arthur Platt unter the editorship of J.A. Smith and W.D. Ross, Oxford: 1949. Brüntrup, G. (1994): Mentale Verursachung – eine Theorie aus der Perspektive des semantischen Anti-Realismus, Stuttgart. Brüntrup, G. (1995): Mentale Verursachung und metaphysischer Realismus. in: Theologie und Philosophie (70), p.203-223. Brüntrup, G. (2008): Das Leib-Seele-Problem – eine Einführung, Stuttgart.

78 Cartwright, N. (1983): How the Laws of Physics lie, Oxford. Honderich, Ted (1982): The Argument for Anomalous Monism, in: Analysis (42), p.59-64. Hume, D. (Treatise): A Treatise on Human Nature, edited by David F. Norton and Mary J. Norton, Oxford: 2008. Kim, J. (2002): The Many Problems of Mental Causation, in: Chalmers, D. (ed.): Philosophy of Mind - Classical and Temporal Readings, Oxford, p.170-179. Lewis, D. (1994): Humean Supervenience Debugged, in: Philosophical Topics (24), S.101-127. Montero, B.: Varieties of Causal Closure. Retrieved as well as cited January 25th 2010 at 3 p.m.: http://barbara.antinomies.org/papers/VarietiesofCausalClosure O'Connor, T. (2000): Causality, Mind and Free Will, in: Noûs – Philosophical Perspectives 14: Action and Freedom, p.105-117. Puntel, L. B. (2006): Struktur und Sein – Ein Theorierahmen für eine systematische Philosophie, Tübingen. Rosenberg, G. (2004): A Place for Consciousness – Probing the Deep Structure of the Natural World, Oxford. Rugel, M.: Panexperientialismus – Das Erleben kleinster Einheiten als Bauplan von Geist, Materie und Welt (Arbeitstitel), forthcoming. Russell, B. (KEW): Our Knowledge of the External World – as a Field for Scientific Method in Philosophy, Reprinted from the Original (1914) with a new Introduction by John G. Slater, New York, 1996. Stapp, H. (2000): Quantum Interactive Dualism, II – The Libet and EinsteinPodolsky-Rosen Causal Anomalies in: Erkenntnis (65), p.117-142. Strawson, G. (2006): Consciousness and its Place in Nature - Does Physicalism entail Panpsychism? Exeter. Strawson, G. (2009): Selves – An Essay in Revisionary Metaphysics, Oxford. Whitehead, A. N. (PR): Process and Reality, New York, 1985. Wright, C. / Hale, B. (2009): The Meta-Ontology of Abstraction, in: Chalmers, D. / Manley, D. / Wasserman, R. (eds.): Metametaphysics, Oxford. Zimmerman, D. (1997): Immanent Causation, in: Noûs - Philosophical Perspectives 11: Mind, Causation and World, p.433-471. Zimmerman, D. (2006) Dualism in the Philosophy of Mind, in: Encyclopedia of Philosophy 2nd Edition, London, p.113-122.

The Spread Mind: Phenomenal Process-Oriented Vehicle Externalism Riccardo Manzotti (Milan) 1. Does phenomenal experience really occur inside the nervous system? In neuroscience, most of the available literature assumes that the mind is the result of some yet-to-be-defined activity taking place inside the nervous system, possibly inside the brain. Many different subsystems have been proposed (ranging from microtubules up to the complete talamo-cortical system, Crick and Koch 1998; Hameroff 1998; Tononi and Edelman 1998; Edelman, Baars et al. 2005; Tononi and Koch 2008), but no conclusive evidence has been presented up to now. Neuroscientists are looking for the neural underpinnings of phenomenal experience (Dehaene, Changeux et al. 2006; Tononi and Koch 2008; Koch 2010; Revonsuo 2010). Consider Christof Koch’s certainty that “Scientists are now revealing the material basis of the conscious mind. In coming years they will gradually fill in the details, making much of the armchair philosophizing moot. […] Such theories will provide quantitative answers to questions that have long stumped us.” (Koch 2010, p. 76) Is such confidence justified? For one, there are not yet any available theories explaining how and why quantitative phenomenal should lead to the emergence of qualitative phenomenal qualities. Nor has it been explained in the least how to connect the normative aspect many mental phenomena to the physical domain. On this regard, Shaun Gallagher warns that “Neither the cognitive neurosciences nor phenomenological approaches to consciousness, however, should be satisfied with simple correlations that might be established between brain processes described from a third-person perspective and phenomenal experience described from a first-person perspective. Such correlations do not constitute explanations, and indeed, such correlations are in part what need to be explained.” (Gallagher 2005, p. 6) Notwithstanding the great success in many respects of neuroscience, many scholars do not share such faith in neuroscience as the forthcoming ‘mindscience’ (for a review see Manzotti and Moderato forthcoming). As a result of these and similar issues, a few authors raised some skepticism as

80 to the soundness of such neural sciovinism (Faux 2002; Bennett and Hacker 2003; Noë and Thompson 2004; Manzotti and Moderato forthcoming). In this paper I will consider a different approach that, albeit still in its infancy, has not yet been exploited adequately: could the neural activity be only a subset of a more extended bundle of physical processes either responsible or identical with consciousness? 2. The location question As to consciousness, it is often assumed that the most basic question is about the nature of consciousness. What is the phenomenal experience? What is the stuff phenomenal experience is made of? If you are a physicalist, as I assume you probably are, you should conceive consciousness as something physical (in this I agree with most available scientific and philosophical literature; see Dennett 1995; Kim 1998; Strawson 2005). This is a very broad statement that is not particularly commitment and yet it is sufficient to draw some preliminary basic suggestions. If phenomenal experience is a physical phenomenon, it has to be located somewhere in space and in time. This means that it should be possible to pinpoint a spatiotemporal region that corresponds to a certain mental occurrence. Let us dub the question as to where the mind is the location question. Looking for a spatiotemporal region is perhaps a convoluted way to look for what is commonly referred to as a process. This is not a trivial conceptual step, shifting from other widespread and less physically-sound yet philosophically popular notions such as state of affairs, states, events. For in physics what happens is usually closer to the occurrence of a process rather than to the instantiation of a state of affairs. Just to make a few examples, in physics it makes more sense to speak of the process of oxidation rather than of a state of affairs corresponding to oxidation. In physics things happens and that is another way to say that processes are the basic building blocks of a physical understanding of the world (Whitehead 1925; Eddington 1929; Heisenberg 1958; Pylkannen 2007). But even without committing too strictly to a process-ontology, it is rather uncontroversial that any physical phenomenon is located in time and space. If phenomenal experience is a physical phenomenon – and I do not see any viable alternative – it ought not be an exception. The location question lies at the very foundation of current scientific research on the nature of subject and so far there is neither consensus nor

81 hope of getting close to a solution anytime soon. Yet it is possible to envisage two rivaling alternative views whose respective value is still largely unresolved. On the one hand, many authors tried to locate the relevant physical phenomena inside the physical boundary of the subject’s body – often inside the physical limits of the nervous systems or some subset of it. This trend can be labeled internalism – namely, the idea that the mind depends on or is identical with properties or events taking place inside the boundary of the subject’s body or some subset of it. Internalism is the view that all the conditions that constitute a person’s thoughts and sensations are internal to the nervous system (Koch 2004; Adams and Aizawa 2008; Mendola 2008). Consider Koch’s claim that “If there is one thing that scientists are reasonably sure of, it is that brain activity is both necessary and sufficient for biological sentience.” (Koch 2004, p. 9) The crucial claim in this sentence is whether brain activity is sufficient or not. Internalism advocates that it is, while other views either weaken to a certain extent or straightforwardly reject the sufficiency of brain activity. As to the necessity, more or less all agree, although what kind of necessity is still an open issue. Coherently, internalism can be seen as the view that answers to the location question by pinpointing to some neural process. According to internalism, the spatiotemporal boundaries of mental phenomena are limited to a short span of a few hundreds of milliseconds and to the bursts of neural firings spreading through axons, dendrites and cellular bodies. An aesthetic experience would allegedly correspond to some neural process initiating and ending inside the nervous system although undoubtedly historically originated by external stimulation, learning, and development. On the other hand, many scholars are skeptical as to resources of the body alone. In particular, it doesn’t seem plausible that the internal properties of a body can cope with certain aspects of the mind whose properties seem unmatched by the properties shown by neural activity. Among such resistant features of the mind two broad categories can be outlined: semantic/intentional/relational properties and phenomenal/1st person properties. The first group expresses the fact that the mind seems to be projected outward the body towards events scattered in time and space accordingly to the ends and stimuli of the subject. The mind seems to have a not reducible relational and externally oriented attitude expressed by hard core philosophical issues such as intentionality and semantics. As a result, it has been considered whether the environment and the spatio-temporal

82 physical surrounding of the subject’s body could be literally constitutive of the mind. This other approach has been labeled externalism – the view that the mind depends on or is identical with properties or events taking place outside the boundary of the subject’s body. In its more radical form, as we will see, an externalist answers to the location question by suggesting that the spatiotemporal boundaries of the physical processes identical with the mind are, at least in principle, larger than those of the body and temporally as extended as they need to be to encompass what they refer to (in practice a lot more than a few hundreds of milliseconds). From the onset it is worth to stress that externalism is no less physicalist than internalism. This should not come as a surprise although it often is. There is a widespread and rather surprising misconception according to which only internalism is a genuine physicalist position. As a matter of fact, physicalism requires only that the mind is explained in physical terms. Physics is not made by the nervous system alone – being “neural” is a subset of being “physical”. Thus, in principle, there are many physicalist explanations which are not confined to the neural domain. Equating the mind with the brain is not the only possible option for physicalism. However, it is the only possible option for a physicalist who embraces internalism in some form. Yet, for the physicalist internalism is only one possibility among many and there is no conclusive evidence for it. In the following, after having outlined the present state of the internalism vs. externalism debate, the various forms of externalism will be listed. The variety of the available models of externalism highlights the value of externalism as a broad approach. In the near future, it is probable that some of these versions will be discarded – some because too daring and others because too conservative. It is something to be expected. 3. Journey to the center of the nervous system At the onset, I will start from a relatively simple case of direct veridical perception. There is an object in front of a human subject – for instance, a red apple in full sunlight. As every schoolboy knows, the sunlight hits the apple surface and other photons are reflected anywhere. Some of these photons go straight against the subject’s eyes. They pass through his cornea, are deflected by the lens, and eventually end their journey smashing against its photoreceptors (rods and cones). The energy of the photons is transferred to very fast chemical reactions whose result is to increase the density of rodopsin in localized portion of the retina. Such increase triggers the emission of action potentials in the surrounding

83 neurons. In turn the action potential travels down an axon headed toward the junction between this neuron and another called a synapse. The presynaptic ending contains synaptic vesicles that contain transmitter chemicals. When an action potential reaches the presynaptic ending it causes some of these vesicles to bond to the presynaptic membrane and to spew its transmitter chemical into the synaptic cleft. It migrates across the cleft and is received by the postsynaptic receptors. In this way, synapsis after synapsis, cell after cell, axon after axon, the sunlight triggered a chain of neural activities whose result is, among other things, the subject having an experience of a red apple. This is very well known. What it is usually forgotten is that we have a physical chain of events and that we keep considering only the last part. The situation can be represented as follow E1 → E2 → E3 → …→ EN-1 → EN When the process is complete, the subject reports a phenomenal experience of red. Now, what is the relevant part of the chain of events as to the occurrence of phenomenal experience? We can be a little more precise and assign to each step a more or less defined role - something like the following Esurface → Ephotons → Elens → Ereceptors → Erodopsin → EopticalNerve → Egeniculate → → EV1 → EV2-N → Einferotemporal Of course this is still a simplification of the actual complexity involved in such a process (many steps were omitted). And yet, for the sake of the argument, we can further simplify it as follows (Pextended) Eexternal word → Ereceptors → Enerves → Eearly cortical areas → Efinal cortical areas Of course leaving to the reader to define precisely what are the final and early cortical areas. An interesting fact is that, although it is possible that such chain, especially in the final steps, is made of feed-forward as well as feedback connections between neurons, from a temporal point of view is nevertheless a feed-forward chain. Regard the complete perceptual chain of events, neuroscientists usually focus their attention on the final part of it (Pinner). Enerves → Eearly cortical areas → Efinal cortical areas They tried to find some good reasons to justify the appearance of phenomenal experience of this physical chain of events. Famously and mistakenly, during XIX century, Johannes Muller tried to assign phenomenal quality to undetermined specific energies located in the afferent peripheral nerves (Muller 1840). The failure to find anything in

84 peripheral nerves led scientists to consider only the inner portion of the chain: Eearly cortical areas → Efinal cortical areas Afterwards, other authors tried to locate consciousness in the early cortical areas. Recently many authors discredited this idea. For instance, V1 is now believed to not contribute directly to conscious experience (Crick and Koch 1995). The ablation of many early areas does not always lead to the disappearance of the corresponding mental content. As a result, many authors are now trying to identify an even smaller and inner portion of the original chain (Zeki and Bartels 1999; Kreiman, Koch et al. 2000; Zeki 2001; Rees, Kreiman et al. 2002; Koch 2004). They focus on Pcortical Efinal cortical areas However this choice, which I have to stress once again it is based the undemonstrated assumption that consciousness has to be produced inside the nervous system – is plagued by the above mentioned faults. Apart from the fact that the properties of anything located in Efinal cortical areas are completely different from our conscious experience – a fact that I will consider at greater length below – there is one more dilemma to be solved. If our experience of the red comes out of a certain neural activity by itself, we are faced with two options: either certain neural patterns are by their very nature phenomenal (and there is no evidence for this so far) or certain neural patterns becomes phenomenal due to their relation with something else. The second option blatantly contradicts the assumed sufficiency of neural activity. Let me spend some more words on this last issue. Let us suppose that I want to defend a position like the following. A neural pattern N, by itself, is not phenomenal. Yet when it has some relation R with something else, it is phenomenal. Now, the relation could be anything you like. For instance, it could be the fact that N takes place inside a human skull, or it could be the fact that N was caused by an external red apple, or it could be the fact that N is the result of filogenesis/ontogenesis/epigenesis. These are very interesting facts about N, but facts that do not change what N is, when N takes place. So if N is the same in all these situations, it cannot have different properties (at least physical properties). Thus, either N is different or N is the same and then it cannot be the fact that N has some relation that assign to N different physical properties. To claim that N becomes different leads to absurdity since by definition it hasn’t changed by being in some kind of external relation.

85 In sum, neither we can say that neural activity is identical with phenomenal experience since nor we can say that a certain neural activity is phenomenal depending on its past history. It would violate a simple physical principle according to which two physically identical systems must have the same physical properties. 4. The Spread Mind After having outlined some of the problems that plague the search for consciousness inside the brain, we can reconsider the problem as a whole. What are the physical limits of the conscious mind? And what are the physical phenomena that can be suggested as being identical with the mind? Let’s go back to the whole perceptual process sketched above (Pextended): Eexternal word → Ereceptors → Enerves → Eearly cortical areas → Efinal cortical areas Is there any really strong reason why the physical process either responsible or identical with consciousness couldn’t be the whole process from Eexternal word up to Efinal cortical areas? I suggest that the physical foundation of phenomenal experience is larger than the neural portion of it. This is the core idea of this paper. I suggest that whenever a phenomenal experience occurs the relevant physical process is not confined to the boundaries of the nervous system but is physically extended to comprehend the external world as well. It may sound like a strange idea. But it is hardly stranger, I think, than the commonplace conception that our phenomenal experience of the world stems out of neural activity alone. To begin with, I will address the apparently simpler case of veridical direct perception (which is still unpredictable and unexplained for most perceptual theories, by the way). I suggest that the physical set of events either responsible or identical to conscious experience is the whole causal chain going from the external perceived thing up to the final relevant neural activity. I admit that the proposal could appear as rather counterintuitive and thus, if it ever has surfaced, it has been, as fast as possible, dismissed. Yet, I will argue, there are no substantial reasons to reject it. Although many authors suggested that the content of consciousness (as well as intentional beliefs and mental representation) could be external to the body (Putnam 1975; Burge 1979; Gibson 1979; Varela, Thompson et al. 1991/1993; Honderich 1998; O' Regan 2001; O' Regan and Nöe 2001; Nagasawa 2002; Noë 2004; Tonneau 2004; Rockwell 2005; Byrne and Tye 2006; Honderich 2006),

86 almost no one ventured to consider the possibility that the mind could be physically larger than the body, especially as to phenomenal experience. It seems that it is very hard to question the widespread consensus as to whether the core physical substratum of the conscious mind is internal to the body. For instance, Jaegwon Kim proclaims that “if you are a physicalist of any stripe, as most of us are, you would likely believe in the local supervenience of qualia” (Kim 1995, p. 159), that is the mind must somehow depend on what take place inside the body. He must believe that the notion is so self evident that does not need any explicit explanation. Yet, a physicalist could appeal to physical phenomena external to the body and thus be a physicalist and not accept the local supervenience of qualia. Yet, there have been many counterexamples to this kind of premise in other areas. For instance, take flight. Is flight only a biological phenomenon since birds and insects perform it? Yes and no. Of course, the biological machinery – made of wings, feathers and membrane – is necessary in order to take advantage of the atmosphere. Yet animals would not fly if there the atmosphere were not dense enough. It would not make sense to deny that flight is not physical because it extends in time and space beyond the confines of the muscles and the wings of the flying animal. Yet, no one would deny that muscles and feathers are very useful, either. Yet, muscles and feathers are neither necessary nor sufficient. For one, a helicopter flies without any feather or muscles. Flight is a physical phenomenon that extends beyond the boundaries of pure biological machinery Isn’t consciousness akin to flight in this respect? Here I would like to contradict explicitly the apparently widespread conviction that if consciousness is a real physical phenomenon, it has to reside inside the nervous system. This is a non sequitur both from a logical and an empirical perspective. First, because it could be a physical system not constrained into such confines. Secondly, because all the empirical evidence does not show that the brain is sufficient, but rather that the brain is necessary. Therefore, I suggest considering seriously whether there are any scientific and empirical reasons to reject phenomenal externalism. This hypothesis has several consequences in the way in which we conceive and single out the physical boundary of the subject. Usually we take the subject to be made of a body and, possibly, by a mind. The mind has been forcefully located either in the brain or inside the body by many recent authors. In the lack of any empirical confirmation, we consider here the possibility that the mind is actually physically larger than the body.

87 Thus, at every instant, for a given subject, we should consider two physical bundles. The first bundle is roughly defined by the skin (although such boundaries are much fuzzier than it could seem at first sight). To avoid any ambiguity, we could call this bundle the traditional body. At each instant there is another set of physical processes. This set is identical with the subject consciousness. These processes, not differently from any other physical process and not differently from the more orthodox neural processes, are extended in time and space. Therefore, they began earlier than the instant in which we consider the subject traditional body. How much time before? There is no common span. They have different temporal duration depending on their causal structure. All together they form a bundle of processes, larger than the traditional body, that we could call the extended body. According to the position presented in this paper, the extended body is identical with the subject consciousness. It comprehends all those events – either near or far, either in time or in space – constituting the phenomenal experience of the subject. In short, although with respect to the traditional view (according to which the mind, whatever it is, has to be carried along comfortably inside the head), the position presented here is classifiable as a kind of externalism, from a more liberal perspective, it is a better form of internalism. In fact, both the structure and the content of the experience remain inside the extended body. However, here, the boundaries of the subject are extended beyond those of the body, thereby permitting to the perceived world to be internal to the mind, without requiring an incorporeal mind. 5. Intuitive arguments and empirical proofs against the spread mind I will mention here a few of the main commonsensical reasons to dismiss such an intuition. Identification between the body of the subject and the subject. It is commonly held that the subject must somehow be identified with her/his body. Since Galileo, the body is considered to host the machinery relevant for the occurrence of the mind. The alternative has always been that of considering an immaterial soul. Yet, there is nothing immaterial in considering the larger set of physical processes whose fulcrum is the body of a subject. Perceptual center of gravity. Our eyes, our ears, our mouth and nose are all centered in the skull. This fact gives a powerful feeling of being where our senses are centered – that is inside the skull. However, such feeling is no more authoritative than the feeling that our planet cannot rotate since we

88 always feel to be standing up. In fact, our brain could be located somewhere else without any sensible difference in our conscious experience (Dennett 1978). Social skills. When we look at someone, it is convenient to identify that someone with her/his body. Analogously, we feel that others identify us with our body. Children learn to recognize themselves in mirror (Staller and Sekuler 1976; Robinson, Connell et al. 1990). Yet what they recognize is their body, and not themselves. Confusion between necessity and sufficiency. Since the body is a necessary part of the subject, it could never be absent. Whenever there is a subject, there is the body of the subject. This is not a proof that the subject is identical either to her/his body or to a part of it. Rather it is a proof (by induction) that the body is necessary for the subject. We should not derive unwarranted conclusion out of this simple fact. For instance, there are no examples of environment-less bodies (brain in a vat). So we do not really know whether a body is sufficient to host a conscious subject. “Brain-invat cases have always been seriously undefined. Until the scenario is much better fleshed out, we can’t say what the brain’s intentional contents would be. Simply to assert that they are the same as yours begs the question” (Lycan 2001, p. 34). Common usage of words vs. exact meaning. In everyday life, it is useful to identify a subject with her/his body. However such identification is more a practical than a theoretical one. It could be something more akin something like the “rising sun”. We all know that the sun does not literally rise anywhere. Yet the common usage of words could endorse some wrong intuition where the exact meaning of terms is unclear, as it happens with consciousness. Although such vague commonsensical views can lurk in the background of the evaluation of a proposal, they are not serious theoretical challenge. I listed them to bring them in the open of a discussion. I will now consider more serious empirical and theoretical drawbacks. 5.1 Spatial limits of a process It has been objected that the whole process cannot be taken to be a good candidate for phenomenal experience since it extends on an excessive spatial dimension. The red apple could be many meters from me. A mountain could even be at several kilometers. The moon is a 1 light second from the standard human observer. Yet this is a very unsound argument,

89 since there is no known law of nature by which processes must have a upper bound limit for their spatial length. On the other hand, neural processes are spatially extended too. Their spatial extension goes from a few microns (the smaller neurons) up to the scale of the whole brain. As far as we know, there is no upper limit. Of course, neurons are constrained to a limited space due both to the practical constraints of a mobile organisms and to the speed limitation of signal axonal transmission. However, in principle, if we accept spatially extended physical processes in the case of neural activity, why should we reject other physical processes only because of their different spatial size? In short, since neuroscientists accept to consider Pcortical although is spatially extended, there is no reason they should reject Pextended for the same fact. 5.2 Temporal limits of a process Here the same argument applied above goes. Neural processes are extended in time. A neural process of object recognition, from the sensory input to the final triggering of some object correlated neural firings, takes approximately 300-400 msec. In physical terms, it is a lot. Plenty of separate events can take place in that amount of time. If we consider the larger physical process, we are outlining here, it is only slightly longer than the neural part. The increase in the total duration is the amount of time necessary for the nervous signal to be transmitted from the retina to the early visual cortex: a few msec. For the sake of completeness, it should be considered also the time needed for light to travel from the object to the retina, but such time is really infinitesimal. Furthermore, as in the case of space, there is no upper limit to the duration of a process. In our brain processes are constrained by the fact that the organism has to react with reasonable speed to incoming stimuli. But there is no reason to suppose that if it were possible to slow down a brain, consciousness should disappear. There is no known law of nature that constrains phenomenal experience to a certain temporal dimension. In short and as above, since neuroscientists accept to consider Pcortical although is temporally extended, there is no reason they should reject Pextended for the same fact. 5.3 Material discontinuity Many neuroscientists could find more congenial to accept only processes mediated by the same kind of chemical reactions and kind of cell. The

90 argument, if there is any, is more or less the following. While Pcortical takes place among cortical neurons, there is a discontinuity once you step out of the brain. Alternatively, a neuroscientist could accept all kind of neurons and the discontinuity is at the receptor. Besides, at the receptor there is another kind of discontinuity, psychologically very demanding: the discontinuity between organic material belonging to one organism and other material (either inorganic or organic of some other organism). After all the psychological need for a material uniformity is the what pushed Camillo Golgi in believing that neurons were interconnected by means of sharing the same intracellular fluid (eventually Ramón y Cajal showed that neurons are, in fact, separate cells). Yet this is a very weak argumentation. A physical process does not depend on the tassonomic uniformity of its constituents but rather on the causal transmission of a certain amount of energy (Reichenbach 1956; Salmon 1980/1993; Dowe 1999; Dowe 2000). As a trivial example, consider a mechanical wave passing through various medium before dissolving into thermal noise. It could pass through concrete, steel, air and remain the same physical process. After all, isn’t the same thing that occurs in neural processes? As every schoolboy knows, neurons are separate and autonomous cells. They communicate by spreading chemicals between synapses. What takes place inside the axon is very different from what happens between neurons. What matters is the propagations of a signal and not the material used for its propagation. Once more, since neuroscientists accept to consider Pcortical although is mediated by discontinuous materials, there is no reason they should reject Pextended for the same fact. 5.4 Counterexample from indirect perception In this category I consider a huge range of various phenomena such as hallucinations, dreams, after images, hallucinations, phosphenes, mental images. There is no space left to consider all these phenomena since each will deserve at least a whole paper. Here I will focus only on dreams mainly because they are the most often quoted case of phenomenal experience occurring without direct contact with the external world. During a dream, the brain seems indeed sufficient to sustain phenomenal experience identical or very similar to actual perception. This fact has been so convincing that many scientists consider direct perception as a special case of dreaming, namely a case where the external stimuli control that

91 special dream that we take as the real world (Edelman 1989; Revonsuo 1995; Lehar 2003; Metzinger 2003). However fascinating this idea could be, there are no final evidence as to its soundness. First, we are not sure what a dream is. And, above all, there is no proof that a neural tissue could be sufficient to host the occurrence of a dream without being part of an extended network of physical processes. The confusion is here due to the fact that it is taken for granted that a dreaming brain is physically separate from the environment. Is it completely correct? Is the brain ever disconnected from the external world? I have argued elsewhere that it is not the case, due to very basic physical considerations (Author in press). Consider a simpler physical system: a bottle of water. You rotate the bottle of water. As a result, the liquid inside will keep moving for a while. Suppose that, after a powerful rotation, you close the bottle inside a box. Is the bottle disconnected from the environment? Yes and no. Of course, the bottle is disconnected from what is taking place in her surrounding after the inboxing. However, are the events inside the bottle (the rotation of the liquid) autonomously produced by the bottle? Of course not. From the point of view of the physical processes involved, what is going on inside the bottle is causally continuous with external events occurred a few moments ago. Similarly, the brains of real subjects reporting having dreams are never disconnected by their past environment. Even if they are sleeping in a hermetically closed room, their brain is still the result of many past events that are causally continuous with the dream-correlated neural activity. Instead of considering normal perception as a special case of dream, dreams could be seen as a delayed and disordered case of perception. Another fact that could support this view is provided by the sever limitation of dreamed mental content during dream. As far as we know, dreams are made of phenomenal building blocks that are always the result of direct contact with the corresponding physical phenomenon in the external world. In dreams as well as in other cases of mental imagery, the brain seems incapable of autonomously producing new phenomenal content but only to recombine them. Systematic studies of dream content showed a remarkable lack of novelty in dream with respect to real life (Domhoff and Schneider 2008). The overall finding of several studies is that "dreaming consciousness" is "a remarkably faithful replica of waking life" (Snyder 1970, p. 133). Not only in dreams it seems that there are no complete novel mental content, but even their combination is seldom really unusual, a condition referred to as bizarreness of dream, i.e. any events

92 outside the conceivable expectations of waking life. In most surveys the majority of dreams were rated as having little or no bizarreness (Snyder 1970; Domhoff and Schneider 2008). Specific studies emphasized “the rarity of the bizarre in dreams” (Dorus, Dorus et al. 1971). A convincing example is offered by born blind patients that seem unable to have any phenomenal experience of colour. It is surprising that there is a widespread and unsupported belief that there are subjects that, although systematically deprived of some sensory modalities, are able to mentally conceive them somehow. A simple case is represented by the alleged ability of born blind subjects capable of dreaming colours1. This is, at best of my knowledge, largely dismissed by actual data (Pons 1996; Ittyerah and Goyal 1997; Kerr and Domhoff 2004). Real born blind subjects do not dream any colour. Real born blind subjects do not have mental imagery with colours. They do not experience the phenomenal experience we have when we open our eyes and see a coloured world. There is some confusion in the literature since many alleged born blind subject are neither blind nor born with that condition. Truth is that many classified born blind became blind at a very early stage in their development (a few months, a few years), but had some kind of contact with light-related phenomena. However, if we set aside all the dubious and vague cases, it seems that without some residual sight, no congenitally blind subject ever reported a mental colour of any kind – whether dreamed or imagined. As reported by a detailed study on 372 dreams from 15 blind adults “those blind since birth or very early childhood had (1) no visual imagery and (2) a very high percentage of gustatory, olfactory, and tactual sensory references” (Hurovitz, Dunn et al. 1999, p. 183). Of course completely born blind subject can have mental imagery of various kinds. They can experience shapes and forms. This is coherent with the kind of phenomenal externalism I advocate here. In fact, a born blind subject has plenty of physical contacts with shapes and forms by means of her other senses. For instance, she can touch a ring and feel the circular shape. It has been suggested that since patients who lose their primary visual cortex to strokes continue to experience visual dreams, the activity in V1 is unnecessary for dreaming (Koch 2004, p. 109). As a result, consciousness should be located innermost. Apart from the fact that the neurobiology of dreaming is still the theatre of heated discussions (Domhoff 2005), 1

For instance, it has been reported that a Turkish painter, allegedly born blind, is capable of mentally feeling colors (Kennedy and Juricevic 2003). Yet there is no evidence that he can have any phenomenal feelings associated with colors.

93 If this were confirmed for every sensor modality, it would support the fact that mental content is not generated inside the brain, but rather is the result of physical continuity with external phenomena. A possibility offered by phenomenal externalism is that the difference between unconscious processes and conscious processes (either direct perception or dreaming or whatever) lies in the existence of a physically causally continuity with real events in the environment, no matter how long and complex. In short, instead of being a counterexample of phenomenal externalism, dreams could offer a convincing test bed for phenomenal externalism. On the other hand, also in normal perception we are not in contact with the instantaneous environment, which is the environment taking place at the exact time of our neural activity, but rather with a temporally proximal environment. In dreams, due to the relative physical and functional isolation from such temporally proximal isolation, there could space left for a temporally more extended spatial continuity with the environment. 6. Advantages of phenomenal externalism I hope I have made it clear that, at least, there are no compelling reasons to reject phenomenal externalism from the start. We need to collect more empirical data and to carry on more dedicate experimentations such as to explore a neglected option. Yet, on a more positive note, why should we take the trouble to consider phenomenal externalism? What are the advantages that could we derive from it? In fact there are many possible advantages. I will mention a few. A different candidate for phenomenal experience (hard problem). So far, consciousness research mostly focused on neural activity. Adopting a process and externalist view, different and more promising physical processes could be identified. A solution to the representation problem. A crucial aspect of the mindbody problem understands how the brain represents the external world. Adopting an internalism approach entails to postulate the existence of intermediate representations physically distinct from what they represent. Phenomenal externalism could allow to get rid of the problem of representation all together, since there would be no need to have representations in the classical sense. What we would perceive would be identical with what is perceived – the vehicle of representation and what is represented being one and the same. An easier approach to mental causation. If mental processes were identical with physical processes, they would have causal antecedents and

94 postponents. They would belong to the physical causal flow. These processes would not face the epiphenomenalist risk since they would not be a property emerged out of an otherwise already coherent causal network. They would be already a part of the causal network. 7. Conclusion The spread mind theory tries to outline a radical ontological twist. The spread mind suggests that most of current problems in dealing with phenomenal experience are due to a series of unwarranted ontological premises. The most obnoxious of them is probably the assumption that phenomenal experience stems out of neural processes alone. As it has been argued by other authors, there is not a magic threshold dividing what takes place inside our nervous system and what takes place outside. Indeed there is a continuous flow of causal processes seamlessly going from the environment to the brain and backward. After all, neural activities are instantiated by series of action potentials distribute in time and in space. They are processes spanning a spatiotemporal region. Once we accept a neural process as something that is not located at a precise point in time and space, there ought not be any difficulty in accepting other kinds of physical processes. References Adams, D. and K. Aizawa, (2008): The Bounds of Cognition, Singapore: Blackwell Publishing. Bennett, M. R. and P. M. S. Hacker, (2003): Philosophical Foundations of Neuroscience, Malden (Mass): Blackwell. Burge, T., (1979): "Individualism and the Mental" in: French, Uehling and Wettstein, Eds, Midwest Studies in Philosophy IV, Minneapolis, University of Minnesota Press: 73-121. Byrne, A. and M. Tye, (2006): "Qualia ain't in the Head" in: Noûs 40(2): 241-255. Crick, F. and C. Koch, (1995): "Are we aware of neural activity in primary visual cortex?" in: Nature 375: 121-123. Crick, F. and C. Koch, (1998): "Constraints on cortical and thalamic projections: the no-strong-loops hypothesis" in: Nature 391(15): 245-250. Dehaene, S., J.-P. Changeux, et al., (2006): "Conscious, preconscious, and subliminal processing: A testable taxonomy" in: Trends in Cognitive Sciences 10: 204-211.

95 Dennett, D. C., (1978): Brainstorms: philosophical essays on mind and psychology, Montgomery: Bradford Books. Dennett, D. C., (1995): Darwin's dangerous idea: evolution and the meanings of life, New York: Simon & Schuster. Domhoff, G. W., (2005): "Refocusing the neurocognitive approach to dreams: A critique of the Hobson versus Solms debate" in: Dreaming 15: 3-20. Domhoff, W. G. and A. Schneider, (2008): "Studying dream content using the archive and search engine on DreamBank.net" in: Consciousness and Cognition 17: 1238-1247. Dorus, E., W. Dorus, et al., (1971): "The incidence of novelty in dreams" in: Archives of General Psychiatry 25: 364-368. Dowe, P., (1999): "The Conserved Quantity Theory of Causation and Chance Raising" in: Philosophy of Science 66: 486-501. Dowe, P., (2000): Physical Causation, Cambridge: Cambridge University Press. Eddington, A. S., (1929): The Nature of the Physical World, New York: The MacMillan Company. Edelman, D., B. J. Baars, et al., (2005): "Identifying hallmarks of consciousness in non-mammalian species" in: Consciousness and Cognition 14: 169-187. Edelman, G. M., (1989): The Remembered Present: A Biological Theory of Consciousness, New York: Basic Books. Faux, S. F., (2002): "Cognitive Neuroscience from a Behavioral Perspective: A Critique of Chasing Ghosts with Geiger Counters" in: Behavior Analyst 25: 161-173. Gallagher, S., (2005): How the Body Shapes the Mind, Oxford: Oxford Clarendon Press. Gibson, J. J., (1979): The Ecological Approach to Visual Perception, Boston: Houghton Mifflin. Hameroff, S. R., (1998): "Quantum computation in brain microtubules? The Penrose-Hameroff 'Orch OR' model of consciousness" in: Philosophical Transactions of the Royal Society of London B 356: 1-28. Heisenberg, W., (1958): Physics and Philosophy, New York: Harper & Row. Honderich, T., (1998): "Consciousness as existence" in: A. O'Hear, Ed., Current Issues in Philosophy of Mind, Cambridge, Cambridge University Press: 137-155.

96 Honderich, T., (2006): "Radical Externalism" in: Journal of Consciousness Studies 13(7-8): 3-13. Hurovitz, C. S., S. Dunn, et al., (1999): "The Dreams of Blind Men and Women: A Replication and Extension of Previous Findings" in: Dreaming 9(2/3): 183-193. Ittyerah, M. and M. Goyal, (1997): "Fantasy and reality distinction of congenitally blind children." in: Percept Mot Skills 85((3 Pt 1)): 897-8. Kennedy, J. M. and I. Juricevic, (2003): "Haptics and projection: Drawings by Tracy, a blind adult" in: Perception 32: 1059-1071. Kerr, N. and G. W. Domhoff, (2004): "Do the blind literally "see" in their dreams? A critique of a recent claim that they do" in: Dreaming 14: 230233. Kim, J., (1995): "Dretske's Qualia Externalism" in: Philosophical Issues 7: 159-165. Kim, J., (1998): Mind in a Physical World, Cambridge (Mass): MIT Press. Koch, C., (2004): The Quest for Consciousness: A Neurobiological Approach, Englewood (Col): Roberts & Company Publishers. Koch, C., (2010): "An answer to the riddle of consciousness" in: Scientific American 303(3): 76. Kreiman, G., C. Koch, et al., (2000): "Imagery neurons in the human brain" in: Nature 408: 357-361. Lehar, S., (2003): "Gestalt Isomorphism and the Primacy of Subjective Conscious Experience: A Gestalt Bubble Model" in: Behavioral and Brain Sciences 26(4): 375-444. Lycan, W. G., (2001): "The Case for Phenomenal Externalism" in: J. E. Tomberlin, Ed., Philosophical Perspectives, Vol. 15: Metaphysics, Atascadero, Ridgeview Publishing: 17-36. Manzotti, R. and P. Moderato, (forthcoming): "Is Neuroscience the Forthcoming 'Mindscience'?" in: Behaviour and Philosophy. Mendola, J., (2008): Anti-Externalism, Oxford: Oxford University Press. Metzinger, T., (2003): Being no one: the self-model theory of subjectivity, Cambridge (Mass): MIT Press. Muller, J., (1840): Handbuch der Physiologie. Nagasawa, Y., (2002): "Externalism and the Memory argument" in: Dialectica 56(4): 335-346. Noë, A., (2004): Action in Perception, Cambridge (Mass): MIT Press. Noë, A. and E. Thompson, (2004): "Are There Neural Correlates of Consciousness?" in: Journal of Consciousnesss Studies 11: 3-28.

97 O' Regan, K., (2001): "What it is like to see: a sensorimotor theory of perceptual experience" in: Synthese 129: 79-103. O' Regan, K. and A. Nöe, (2001): "A sensorimotor account of visual perception and consciousness" in: Behavioral and Brain Sciences 24(5): 939—1011. Pons, T., (1996): "Novel sensations in the congenitally blind" in: Nature 380(11): 479-481. Putnam, H., (1975): Mind, language, and reality, Cambridge: Cambridge University Press. Pylkannen, P., (2007): Mind, Matter and the Implicate Order, Berlin: Springer. Rees, G., G. Kreiman, et al., (2002): "Neural Correlates of Consciousness in Humans" in: Nature Reviews 3: 261-270. Reichenbach, H., (1956): The Direction of Time, New York: Dover. Revonsuo, A., (1995): "Prospects for a cognitive neurosciences of consciousness" in: Behavioral and Brain Sciences 18: 694-695. Revonsuo, A., (2010): Consciousness. The Science of Subjectivity, Hove: Psychology Press. Robinson, J. A., S. Connell, et al., (1990): "Do Infants Use Their Own Images to Locate Objects Reflected in Mirrors?" in: Child Development 61(5): 1558-1568. Rockwell, T., (2005): Neither ghost nor brain, Cambridge (Mass): MIT Press. Salmon, W., (1980/1993): "Causality: Production and Propagation" in: E. Sosa and M. Tooley, Eds, Causation, New York, Dover: 154-172. Snyder, F. W., (1970): "The phenomenology of dreaming" in: L. Madow and L. Snow, Eds, The psychodynamic implications of the physiological studies on dreams Springfield (Ill), Thomas: 124-151. Staller, J. and R. Sekuler, (1976): "Mirror-Image Confusions in Adult and Children: A Non Perceptual Explanation" in: 89 2(253-268). Strawson, G., (2005): "Why physicalism entails panpsychism", in: Danish National Research Foundation. Tonneau, F., (2004): "Consciousness Outside the Head" in: Behavior and Philosophy 32: 97-123. Tononi, G. and G. M. Edelman, (1998): "Consciousness and Complexity" in: Science 282: 1846-1851. Tononi, G. and C. Koch, (2008): "The neural correlates of consciousness: an update." in: Annals of the New York Academy of Sciences 1124: 23961.

98 Varela, F. J., E. Thompson, et al., (1991/1993): The Embodied Mind: Cognitive Science and Human Experience, Cambridge (Mass): MIT Press. Whitehead, A. N., (1925): Science and the modern world, New York: Free Press. Zeki, S., (2001): "Localization and Globalization in Conscious Vision" in: Annual Review of Neuroscience 24: 57-86. Zeki, S. and A. Bartels, (1999): "Toward a Theory of Visual Consciousness" in: Consciousness and Cognition 8: 225-259.

Taking the Hard Problem of Consciousness Seriously: Dualism, Panpsychism and the Origin of the Combination Problem Michael Blamauer (Vienna) About the nature of the external world we have at the onset nothing but hypotheses. Before we test them in any very exact way, we may with safety try to understand them. Perhaps what seemed the wildest of them all may turn out to be the very best.1 Josiah Royce, “Mind and Reality” 1. From dualism to panpsychism The “hard problem of consciousness”, made prominent by David Chalmers over the last fifteen years (cf. Chalmers 1995, 1996, 2002) essentially concerns the idea of fundamental mental properties and the concept of naturalistic dualism. The present paper aims to discuss two important implications of this notion of the hard problem. The first regards the principle difficulty of limiting the scope of fundamental properties. The second regards the combination problem. I will argue firstly that naturalistic dualism entails panpsychism, by demonstrating that the fundamentality of mental properties entails their ubiquity. Secondly, I will show that the core dualistic assumptions presupposed in the formulation of the hard problem within the standard materialist framework might be the origin of the combination problem.2 In the last section I will discuss two alternative positions – substance dualism and metaphysical idealism – with the primary aim of showing that a reformulation of panpsychism in one of these frameworks may have the advantage of sidestepping the combination problem. 1

Royce (1882), 35. I wish to thank Galen Strawson for the remark that the formulation of the “hard problem of consciousness” is essentially entailed in the assumption of the “standard materialist framework”. This remark substantially influenced my thoughts on the combination problem. 2

100 The “hard problem of consciousness” is chiefly concerned with the challenge of finding a place for subjective experience within the materialist metaphysics of our world. The major difficulty here lies in the fact that reductive explanation of subjective phenomena is impossible.3 Reductive explanation means here that a phenomenon (Q) can be fully captured in terms of or reduced to some more fundamental phenomena (P), so that P = Q. (Cf. Lewis 1994 and Chalmers 1996, 70) David Chalmers developed a powerful argument against reductive positions on consciousness such as physicalism or a posteriori materialism (Cf. Chalmers 1996, 52ff. and 131ff.). The basic idea is that the meaning of consciousness is determined in a different manner than the meaning of other scientific concepts, like water = H2O. To take an example, if identity in the aforementioned sense would hold for consciousness (and its underlying physical processes), an atom-for-atom physical or functional duplicate of mine (a thing indiscernible from me by methods of physical science) would have identical phenomenal experiences. But even if I cannot make sense of a duplicate of certain physical systems, like a lake lacking the property of being wet or being frozen under certain circumstances, I can of course make sense of a physical or functional duplicate of mine lacking conscious experience. But if there is no necessity in the relationship between the occurrences of subjective experiences and physical (or functional) brain states they cannot be identical, because identity is a relation that requires necessity (cf. Kripke 1980). Hence reduction in the mentioned sense is impossible. The same argument holds for the idea of consciousness as an emergent property of the brain. The concept of emergence here is taken similarly to the way we take liquidity to be an emergent property of a group of H2O molecules. Liquidity is a property of water, but liquidity is not a property of H2O molecules. If we try to understand the relation between the property of liquidity and the non-liquid components of the system, we find that liquidity arises due to some law-like behavior of the components under certain conditions (temperature, pressure, etc.). Liquidity therefore depends (in a strong sense) on the law-like behavior of the H2O molecules – it reduces to it. Insofar as a reductive approach to consciousness is rejected, a more radical kind of emergence has to be assumed, because it does not 3

Chalmers (1996) and Searle (1992) give a fairly broad overview on the major arguments against materialism. Further reading on classical arguments is to be found in Chalmers (2002).

101 intelligibly reduce to something more fundamental. Yet such a kind of radical (unintelligible) emergence seems to imply a kind of magic or miracle, and it would not provide any explanatory advantage over weak emergentism or physicalism (cf. Strawson 2006, 18). This is the reason why Strawson rejects the concept of brute emergence in the mentioned sense.4 What follows from the impossibility of reduction within this argumentative framework? An apparent consequence is that consciousness must be viewed as something fundamental. Now, due to the claim of irreducibility concerning consciousness within the scope of materialism and the “hard problem”, the following points establish the framework of arguments: 1. What physics say about nature is true. (Physical properties are taken as fundamental properties of reality.) 2. Consciousness is a fundamental and irreducible fact of reality. (Phenomenal properties are taken as fundamental properties of reality.) 3. Our universe consists only of one kind of stuff. This stuff has two fundamental kinds of properties: physical properties and mental properties. 4. These properties correspond to a set of fundamental laws that correlate the two kinds of basic properties with each other. These four assumptions form the pillars of a position that David Chalmers calls “naturalistic dualism” (which is in fact a variety of non-reductive materialism) (Cf. Chalmers 1996, 123ff.). Naturalistic dualism takes consciousness seriously yet at the same time accepts the physical facts of physical science as fundamental facts of reality. However, these assumptions seem to imply a further assumption due to the relationship between (1), (2) and (3), namely: 5. The fundamental features of reality must be ubiquitous. The reason for assuming (5) is that if we take physical properties to be ubiquitous, we must assume this of mental properties as well and in the same way. There seem to be no grounds to presuppose that something 4

A similar rejection of the concept of consciousness as a brute emergent phenomenon was already prominently featured in William James’ Principles of Psychology (1890). (Cf. James 1890/1998, 145ff.)

102 fundamental would initially appear at a certain functional or organizational level. William Seager famously pointed out that [i]t is disturbing that consciousness can be an absolutely fundamental feature of nature while being dependent upon particular systems satisfying purely functional descriptions […].No other fundamental feature of the world has this character, or a character even remotely like it. It is rather as if one declared that ‘being a telephone’ was a fundamental feature of the world, generated by a variety of physical systems agreeing only in fulfilling the relevant, highly abstract, behaviourally defined functional descriptions. […] Of course, seeing that consciousness is a truly fundamental feature we cannot ask how it is that all and only systems meeting certain functional descriptions are conscious, yet this idea does seem to deepen rather than mitigate the mystery of the generation problem. (Cf. Seager 1995, 275)

It should be noted that this statement was addressed against Chalmers’ “principle of organizational invariance”, which is constitutive for the restriction of the fundamentality claim of consciousness in order to define the framework of naturalistic dualism and to establish a border to panpsychism. Chalmers introduced this principle because he clearly saw that if we take mental properties to be fundamental and ubiquitous in the same way we do for physical properties, then the fifth assumption pushes the case for panpsychism (cf. his own remarks on panpsychism in Chalmers 1996, 293ff.). This structure of reasoning provides a rather good argument for panpsychism and is indeed originally found in some notable papers on this topic (Cf. e.g. Nagel 1979, Seager 1995 and Strawson 2006): If we accept the irreducibility of consciousness to pure physical or functional states, and if we are doubtful on the topic of brute emergence, we have a strong argument for panpsychism. Panpsychism is – in short – typically understood to be the view that conscious experience is a fundamental as well as ubiquitous characteristic of our universe, equal to physical properties like mass, charge and spin. In the current discussion some hold it as a possible coherent thesis about the mind-body relationship. According to this definition, every concrete thing has both physical and mental aspects. Galen Strawson takes panpsychism to be the position that holds “that the existence of every real concrete thing involves experiential being even if it also involves non-experiential being” (Strawson 2006, 8). Thus the scope of experience to be found in the

103 universe may range from the micro-experiential level of ultimate particles to the macro-experiential level of human beings. Yet the assumption of consciousness as something widespread, as well as the talk of micro- and macro- experience seems to obscure the meaning of the concept. If we speak of the consciousness of a dog, a cat or a mouse – even if it is quite obvious we are not equating the term with “human consciousness” – we end up with a rather confused idea of what exactly we mean by it. As a matter of fact, the matter becomes even more confused the deeper we descend the “ladder of evolution”. Due to this difficulty, I think it is important to keep at least the most fundamental characteristics of consciousness in mind when we try to consider it on the lowest organizational levels of the universe. This is even more important if we are attempting to arrive at panpsychism as a sound theory of the human mindbody relationship. This means that the kind of consciousness we ascribe to ultimates must at least provide the possibility of generating a full-blown version of these features at the level of more highly organized animals. In the current debate, the fundamental characteristic of consciousness, which is likewise associated with the hard problem, is its experiential phenomenal character. But despite the comprehensive discussions of the nature of phenomenal consciousness, one might still feel uncomfortable with this as an exclusive characterization. This is because narrowing the concept of consciousness as a dimension of subjectivity solely to its experiential phenomenal character seems to leave out some of its fundamental aspects. Therefore, I intend in the following to take a deeper look at the problem by arguing that consciousness is not just an empirical fact, but also a transcendental feature of our world.5 The term “transcendental” is used here in the classical sense as the view that the subject’s conditions of possibility of experiencing objects are simultaneously the conditions of the possibility of the appearing objects themselves. Furthermore, I assert that consciousness as a dimension of subjectivity is primarily intentional, and characterized by some basic form of selfhood and experiential feeling. In the context of the following outlines, intentionality is taken in the sense of always being involved with something other than itself. Whereas consciousness represents things in a certain way, things like tables, chairs and knifes do not represent things in a certain way. They are not about anything. On the contrary, tables, chairs and knifes are rather things being 5

In recent years, Rowlands (2003) has referred to this interpretation of phenomenal experience as transcendental rather than as an empirical feature of our world.

104 represented in conscious experience of them. Yet conscious acts do not only involve the consciousness of something, but also an immanent awareness of the experiencing subjects themselves. This accompanying consciousness of the subject itself is no additional act directed towards an empirical person, but rather the core self-givenness of every intentional act. Consciousness does not appear to itself in the same way objects do. Rather, it is in itself immediately self-conscious by representing the world in a certain way. There is no additional experience. This brings us to the topic of selfhood. The term selfhood is taken in the sense that everything that is perceived or experienced, is perceived or experienced by a subject of experience. The concept does not thereby refer to higher cognitive or reflexive features such as a self-model, a self-concept, a higher-order reflexive awareness of one’s own mental states or internal system status, and the like. Rather, the argument is for an experience-immanent structural feature of first-personal presence or mineness of experience. The unfortunately untranslatable, though correct, German formulation of this idea would be that experiences are “jemeinig”. However, I don’t think it makes sense to speak of experience that is not present to anybody.6 And of course, coming to the third and last of the basic features of consciousness, every perception or experience of something is steeped in some exclusive feeling which is the essence of what it is like for me to see, hear, smell, do, etc. In the following, it is assumed that every one of the aforementioned features entails problems for panpsychism, especially concerning the challenge of making sense of tiny subjects summing. 2. The combination problem In light of the characteristics of the outlined subjective dimension of consciousness, panpsychism raises several problems.7 One of them is the so-called combination problem, first noted by William James in his Principles of Psychology (1890). It concerns the fact that we currently have no idea as to how a combination of micro-experiences could produce fullblown conscious experience like ours. Although James does not refer to panpsychism per se, since his critique was directed only at so-called 6

Galen Strawson has also recently put this idea forward. Cf. Strawson (2009, 63ff.). A comprehensive overview of the problems of panpsychism can be found in Seager (1999, 216-252) and Seager / Allen-Hermanson (2005). 7

105 “mind-dust” theories, he points out the logical incomprehensibility of the idea of subjects summing (Cf. James 1890/1998, 158 and 160; also Seager 1999, 242). Hence, even if one could make sense of the idea of atoms having conscious experience, there seems to be no easy answer to the question of how lower single states of subjective experience could be combined to result in higher (and even more complex) states of consciousness, especially with respect to the subjective dimension of consciousness. The problem starts when we attempt to explain how multiple streams of consciousness – e.g. the streams of every single unit that constitutes my brain – can combine to constitute one single perspective (namely mine) on the object. The perspectival relationship of the ultimate to its environment must in some sense be constitutive in its having an intentional object present and hence in the object’s condition as three-dimensional thing. How then, can zillions of these perspectives combine to form my single perspective on an object? The same applies to the concept of selfhood: You cannot arrive at a single first-person-perspective from the fusion of many first-person-perspectives. For the sake of argument, let us imagine that every single particle has a basic kind of “self” or “I”. Even if it does not see or hear because it has no ears or eyes, it has a specific feeling for its environment. It stands in an intentional – though very primitive – relationship to it. Now, combine a multiplicity of these different experiencing “selves” or “I’s”. You won’t ever arrive at yourself as an experiencing single “I” or “self” via a combination of different other “I’s” or “selves” (Cf. James 1890/1998, 158ff.). And again, the same holds true for a single subjective experience. My feeling (severe) pain is not constituted by the fusion of a multitude of (slightly) pained tiny subjects (cf. Goff 2006). Treating the combination problem by evaluating the concept of panpsychism, it seems Thomas Nagel was right in his programmatic paper of 1979. There, he stated that the acceptance of the premises that lead to panpsychism is more coherent than their negation, but in the end, panpsychism as a whole turns out to be rather problematic. Yet how could it be that all the presented arguments coherently lead to panpsychism, but culminate in such a seemingly serious problem? 3. On the origin of the combination problem If all of the reasoning concerning panpsychism ultimately leads to the combination problem, it seems plausible that the error might be found along the way. Hence, a recapitulation of the outlined assumptions,

106 arguments and consequences is in order. I will go about it by critically discussing the basic assumptions and major steps of the foregoing argumentation, which has led us to panpsychism and the combination problem. Let us begin by examining the background of the non-reductive approaches to consciousness under discussion today, which were the source of our reasoning. This is the background of the so-called “hard problem”, which can be interpreted as the name for the main point of dispute between reductionists (physicalists or functionalists) and non-reductionists (dualists or panpsychists) in regards to the nature of consciousness. Whereas reductionists (physicalists or functionalists) claim – in short – that mental phenomena are identical with physical or functional processes and hence reducible to it, non-reductionists (dualists or panpsychists) hold that consciousness is something fundamental besides the fundamental physical aspects of reality. Generally speaking, the standard materialist framework entails the formulation of the “hard problem”, and this problem indicates the difficulty of finding a space for consciousness within the physical domain of our world. And given the fundamentality claim concerning consciousness within this framework, one is naturally driven to a dualaspect position better known as property dualism.8 It is essential to bear this in mind when we return to the combination problem later. Now, as I have argued above, it is apparently only a small step from property dualism to panpsychsim. The reason for this is the principle difficulty of limiting the scope of fundamental mental properties to within the overall framework. And the reason for this in the first place is that consciousness is not a vague concept.9 Moreover, following William Seager, it turns out to be rather incoherent to artificially limit mental properties to appearing only in systems with a certain complex and finegrained functional architecture while generally taking physical properties to be ubiquitous. It is more coherent to view mental properties as ubiquitous as well, above all when facing the problems of emergentism (cf. Strawson 2006, 12ff.). 8

David Chalmers labels his position “naturalistic dualism”, even though it is a kind of property dualism. By “naturalistic” he wishes to emphasize the fundamentality of both the physical and the mental, and likewise indicate the – albeit yet unknown – lawful and natural relationship between these two distinct domains. 9 Philip Goff developed the argument that the non-vagueness of the concept of consciousness leads to a kind of panpsychistic ontology. He presented this idea at the workshop, “The Mental as Fundamental” at the University of Vienna in May 2010.

107 These outlines urge us to address the question of the status of physical properties in relation to mental properties and, moreover, the distinct mode of being of both. For these reasons, Strawson infers a position he calls “microphsychism” (Cf. Strawson 2006, 24ff.). Micropsychism, like property dualism, presupposes the existence and fundamentality of a mindindependent, physical world. According to Strawson, micropsychism is a position similar to property dualism insofar as it takes mentality to be a fundamental property of physical systems. But it avoids the problem of emergence by not only ascribing mentality to certain complex systems, but by taking it to be the property of (at least some) fundamental particles as well: Real physicalists must accept that at least some ultimates are intrinsically experience-involving. They must at least embrace micropsychism. Given that everything concrete is physical, and that everything physical is constituted out of physical ultimates, and that experience is part of concrete reality, it seems the only reasonable position, more than just an ‘inference to the best explanation’. (Strawson 2006, 25)

Now, it follows that the concept of property dualism and its subsequent development as micropsychism are obviously the origin of the combination problem. This diagnosis is based on two strong reasons: 1. The ontological separation of the mental and the physical, by taking both as fundamental properties of reality (which are neither reducible, nor emergent). 2. The transfer of this fundamental separation from the ontological macro- to the ontological micro-level (to save the intelligibility of emergence). To arrive at this diagnosis, we must merely draw the logical conclusions from the foregoing passage: If the subjective dimension of consciousness is nothing we can think of in terms of generation, then micropsychism, as the idea of (some) fundamental particles having physical properties as well as mental properties in the aforementioned sense, is a dead position. Within such a framework, panpsychism (as well as micropsychism) lacks an explanatory advantage over any dualistic position – on the contrary, it must deal with counterintuitive consequences and additional ontological dead weight. To put it differently: In the case portrayed above, the mind-body problem has only been transferred from the macro-level of human

108 consciousness to the micro-level of particle-consciousness merely to avoid the problems of emergentism.10 After evaluating the recapitulation of arguments, the standard materialist framework as the primary source of property dualism turns out to be the origin of the combination problem. It has been argued that this kind of dualism is also manifested in the sort of panpsychism that later faces the unsolvable combination problem. Hence, if we want to overcome the combination problem we have two alternatives: Either we abandon panpsychism and turn to a completely different position, with other premises etc. Or, if we wish to uphold panpsychism as a sound theory, we must challenge the background assumptions of the standard materialist framework by finding an alternative position that meets the different requirements. In other words: If we take the mental to be fundamental, it is not necessarily self-evident that we should take physical properties to be fundamental in the same sense. However, precisely this is presupposed in the formulation of the “hard problem” and, moreover, the mindindependence of the physical domain is in fact an unquestioned premise of the most prominent arguments voiced in this discussion (e.g. the zombieargument or the knowledge-argument). Actually, I believe a fully elaborated position that meets all the outlined requirements is currently lacking. Nevertheless, I view a few positions as possible candidates for further consideration with regard to the combination problem. In the following, I will outline two of them. 4. Two alternative strategies for avoiding the combination problem If the combination problem is in principle unsolvable within the scope of the materialist framework, it seems the only viable strategy is to sidestep it – either by switching the framework or by abandoning panpsychism per se. There are at least two alternative approaches to property dualism, with quite different background assumptions. Both take consciousness seriously, yet without running into the combination problem. On the one hand, there is substance dualism, which avoids the combination problem by avoiding panpsychism as such. On the other hand, there is – at first glance, the more radical approach of – idealism, which avoids the combination problem by constructing a different framework for panpsychistic ideas. 10

Dean Zimmermann (2006, 115) referred precisely to such a kind of “micropsychism” as a rather “bizarre theory”, which nevertheless “qualifies as compositional dualism” even if it seems “to be a kind of materialism”.

109 Let me begin with the first candidate: substance dualism.11 In the following discussion of this doctrine my major focus is on its relevance to the combination problem. I will not discuss the typical problems related to this view.12 In his book on Cartesian dualism, John Foster presents the five basic claims of substance dualism. It is astonishing that they are in most respects similar to the basic claims of panpsychism. He lists them as follows (Foster 1991, 1): 1. 2. 3. 4. 5.

There is a mental realm. The mental realm is fundamental. There is a physical realm. The physical realm is fundamental. The two realms are ontologically separate.

Of course, one can easily see that it is claim (5) that is essential to the idea of substance dualism. Contrary to panpsychism and even to property dualism, substance dualism views the mental and the physical as possibly existing independently of one another. Every single mind is something that exists merely through and by itself, without further dependence on a physical body consisting of physical ultimates. Foster points out that substance dualism views the mind as being (i) conceptually fundamental, as well as (ii) metaphysically fundamental. According to (i), “No mental statement is amenable to a non-mentalistic analysis”; and according to (ii), “No mental fact is non-mentally constituted” (Foster 1991, 8). And the same holds true for the physical realm. So again, contrary to property dualism, which faces the problem of emergence when attempting to make intelligible the occurrence of mental properties at a certain level of functional or physical complexity, substance dualism avoids it by simply holding that the mind is neither an emergent phenomenon, nor identical with physical processes (neither token- nor type-identical), since “facts or states of affairs, cannot be identical if their ontological ingredients are different” (Foster 1991, 9). In Descartes’ view, the mind is simple and essentially indivisible. He takes it to be a substance in the classical Aristotelian sense of having 11

In my discussion of substance dualism I primarily follow Forster (1991), Crane (2003), Zimmermann (2006) and Robinson (2007). 12 Which are in fact the interaction problem and the pairing problem. For a comprehensive and detailed discussion of the different problems of substance dualism cf. Zimmermann (2006), Robinson (2007) and Lycan (2009).

110 independent existence (in addition to being persistent and propertyinstantiating).13 Minds are unities in a core sense, and each individual mind is considered to be an individual mental substance. This was one of Descartes’ basic characteristics of the mind, posited to differentiate it from material (extended) objects: whereas objects can be divided into smaller parts and, vice versa, constructed of such parts, the mind is an indivisible unity, though a unity of different and complex experiences. Substance dualism is a “radically non-physicalistic account of what exists or occurs within the mind” (Foster 1991, 202) and it furthermore views all appearances as belonging to a basic mental subject in the sense that “mental events are always and necessarily events concerning the mentality of these subjects” (Foster 1991, 205). However, this talk of the mind as a substance does not imply that the mind is to be regarded as a “thing” in the same way we regard tables, knifes and chairs as things. Substance in this context simply means that the mind is something fundamental in the sense of not ontologically relating to anything other than itself. Moreover, the mind is not to be considered a thing since things are contents of representational states and the mind is first and foremost no such content. On the contrary, the mind represents things in a certain way due to its intentional character. Obviously, because of this radical separation of the mind-sphere from the physical world of material objects, substance dualism faces nothing like a combination problem, since the mind is not understood as something generated or composed. A position different to, and even more radical than, substance dualism would be one that denies the aforementioned metaphysical separation by (either entirely or merely partially) absorbing the physical realm into the fundamental dynamics of the mind. As a consequence, the physical realm would no longer be fundamental in the same sense as the realm of the mind, but rather be taken as derivative of fundamental mental processes or objects. Nowadays, such a position would be labeled classically idealistic. Most of these positions could also be interpreted as panpsychistic insofar as they consider the mental to be something fundamental and ubiquitous in their overall philosophical systems (e.g. those of Leibniz, Berkeley or the

13

Crane (2003), referencing Peter Simons (1998), indicates that the concept of substance has little relevance in current ontological discussions since it is challenged by other ontological categories like events, particulars, etc.

111 German Idealists).14 It must firstly be stated that even though idealism is treated today as a rather dead position, it is obviously at no explanatory disadvantage to realistic, materialistic or dualistic positions. Indeed, quite the contrary is the case, although I will not attempt a vindication of idealism in the following. As in the preceding section, this position will only be discussed with regard to the combination problem. A major aim of idealistic approaches was to overcome substance dualism as a doctrine claiming the existence of two different realms of being, which in their view was in some way an unintelligible distinction. One of the major critics of dualism was Leibniz.15 In his Discours de métaphysique (1686/2002, 20ff.) he offers a challenging critique of Descartes’ concept of extended substances and the notion of their divisibility. Even if the mind is taken to be an indivisible substance, the extended world (i.e. the material world made of matter) is in principle divisible ad infinitum. Leibniz’s point of criticism was that if we consider matter to have this character, it would be unintelligible how an infinite number of infinitely small units could merge to result in a being of a certain shape and certain size, such as a table, knife or even a human body. So Leibniz assumed that objects, which are mere aggregates, must be made up of simples, which are by definition indivisible. There follows the distinction between mere aggregates and real unities (Cf. Rescher 1979, 77ff.). In Leibniz’s philosophy, these indivisible simples or individual substances are called monads (Cf. Leibniz, 1714a/2002, 152ff.; 1714b/2002, 110ff.). One might now ask: What is the quality of a monad that makes it the indivisible constituent of everything that exists? Remembering Descartes’ definition of the mental, one can easily infer that a monad must be something mental in order to meet the specified requirements. And this is what Leibniz’s idealism is about. However, idealism is only the umbrella term for a number of positions that range from a Leibnizian kind of pluralistic idealism, to Berkeleian phenomenalistic idealism, to the transcendental idealism of Kant and, later, the German Idealists. In general, the basic assumption of the idealistic position is that reality is epistemologically and (depending on the kind of 14

Cf. Seager/Allen-Hermanson (2005), who label at least Leibniz and Berkeley as panpsychistic idealists. But certainly, e.g. Schelling’s System (as representative of German Idealism) should be interpreted as panpsychistic idealism as well (cf. Blamauer 2006). 15 In my discussion of Leibniz’s idealism and his critique of Descartes’ notion of substance I mainly follow Leibniz (1686/2002), Leibniz (1714a/2002), Leibniz (1714b/2002) and Rescher (1979).

112 idealism) metaphysically mind-correlative or mind-dependent. Of course, this assumption carries with it a real challenge, as it runs contrary to our everyday assumption about reality, which, in a nutshell, is that there exists a physical, mind-independent as well as metaphysically fundamental material world. But this should not concern us here, since we are only addressing the question of alternative strategies that avoid the combination problem. In this context, a hypothesis worth exploring could be as follows: • (H) All facts about states and objects of reality are facts about experiential or mental states or about relations between or within experiential or mental states. (H) is conceptually connected with the following two claims: 1. Consciousness is epistemologically and metaphysically fundamental. 2. Mind and world are not distinct metaphysical realms. According to the basic notion of idealism, consciousness must be considered the constitutional basis of the objective world. Hence, mind and the objective world must be dynamically interrelated to each other due to their identical constitutional basis. Now, what consequences result for our main questions concerning the mind-body problem and the combination problem? Obviously, there is no longer a mind-body problem in the sense of the aforementioned “hard problem”, since within the scope of idealism, there is no more question of how the mind could fit into a supposedly fundamental physical framework. There is only one fundamental basis – be it interpreted dynamically as a process or more statically as a substance; no more question remains of how two epistemologically and ontologically different kinds of properties, substances, states, particulars or events could relate to each other. There is only mind, and that is the only immediately given, real fact. If we now examine the combination problem in light of this view, it seems to vanish into thin air. Within the Leibnizian framework, there are only monads or aggregates of monads and a hierarchical organizing principle (Cf. Leibniz 1714a/2002, 156 and Rescher 1979, 110ff.). Every real unit is a monad, even if it assembles a set of monads beneath itself as the hierarchically superior monad and hence organizational principle. In the case of the mind-body problem, this means that even if the body is an aggregate of monads, it has

113 a hierarchical organization where the highest principle is the apperceiving subject. This subject, of course, is an indivisible unit by itself, and hence there is no combination problem. Within a phenomenalistic (Berkeley), or transcendental (Kant, German Idealists), framework, there is only the transcendental subject, or transcendental subjectivity, as the constitutional principle of an objectively given world. According to Descartes’ skeptical hypothesis outlined in the 2nd Meditation, we cannot be sure whether we perceptually grasp things as they really are or rather merely as they are present to an experiencing subject. Kant’s transcendental project stands firmly in this tradition when it assumes that gaining objective knowledge about the external world is necessarily mediated by the subjective conditions under which we are aware of them. Due to the fact that knowledge of external things necessarily requires experience, and experience always means that things appear for a subject and only under the subject’s conditions of experience, we have no access to things as they are in themselves, but rather only to how they are present for us in experience. If we examine the question of objectivity from this transcendental point of view, the mode of being of the objective turns out to always be related to subjective conditions of conscious experience. Being, in the sense of being-an-object-for, turns out to be relative to a subject;16 it depends on the subject’s point of view. In contrast, consciousness does not appear for itself in the same way objects do. It rather is in itself immediately self-conscious in representing the world in a certain way. This is the reason why all attempts of the subject to comprehend herself in terms of objectivity must fail – at least within the transcendental framework. And obviously, no combination problem occurs either.

5. Conclusion Wild and airy indeed! But why so? Mind-Stuff was a worse hypothesis, because, when you tried to express all its consequences, it became unintelligible. The ordinary uncritical Atomism is a worse hypothesis, because we never get from it the least notion of how this eternally existent matter may 16

It must be mentioned that the former, epistemologically-driven, critical transcendentalism of Kant was transformed into a rather strong ontological idealism by the German Idealists, especially by Fichte and Schelling, as a result of the alleged unintelligibility of things-in-themselves.

114 look and feel when nobody sees or feels it. The mystical “one substance with two faces” is worse, because that is no hypothesis, only a heap of words. (Royce 1882, 40)

Keeping within the scope of Royce’s thoughts, I think there is no better reflection of what I have tried to outline than these words. I began with a rejection of materialism and embarked on a quest for a suitable explanation of consciousness. I wished to take the fundamentality claim of conscious experience seriously and was naturally driven towards panpsychism. I attempted to show that panpsychism faces the combination problem only due to certain core dualistic background assumptions resulting from a rather questionable scientific worldview, which I classified as the standard materialist framework. This position also forms the background to the idea of “one substance with two faces”, nowadays better known as property dualism or as the derivative form of micropsychism. So in aiming for a solution to the combination problem entailed by the standard materialist framework, I have presented two alternative positions that both sidestep the combination problem by taking the mental as truly fundamental: one that avoids the problem by abandoning panpsychism as such – substance dualism; the other sidestepping it by constructing a different ontological framework: idealism. However, these positions may also have problematic consequences. One might now ask: Shouldn’t we better give up on these questions given all these troubles? I do not think so. To be continued…17 References Blamauer, M. (2006): Subjektivität und ihr Platz in der Natur, Stuttgart: Kohlhammer. Chalmers, D. J. (1995): “Facing Up to the Problem of Consciousness.” In: Journal of Consciousness Studies 2 (3), pp. 200-219 Chalmers, D. J. (1996): The Conscious Mind: In Search of a Fundamental Theory, Oxford et al. Chalmers, D. J. (2002): “Consciousness and its Place in Nature.” In: Idem (Ed.), Philosophy of Mind. Classical and Contemporary Readings, Oxford et al. 17

I thank Pierfrancesco Basile, Wolfgang Fasching, and Georg Schiemer for helpful comments on former drafts of this paper. The Austrian Science Fund (FWF) financed this study in the context of the research project “Taking the Hard Problem of Consciousness Seriously – Naturalistic Dualism and the Consequence of Panpsychism”.

115 Crane, T. (2003): “Mental Substances.” In: O'Hear, A. (ed.), Minds and Persons. Cambridge University Press. Foster, J. (1991): The Immaterial Self. A defence of the Cartesian dualist conception of the mind. London / New York: Routledge. Goff, P. (2006). “Experiences Don’t Sum.” In: Freeman, A. (ed.), Consciousness and ist Place in Nature. Does Physicalism entail Panpsychism?, pp. 53-61. James, W. (1890/1998): The Principles of Psychology Vol. I. Thoemmes Press u.a. (Nachdr. d. Ausg. New York 1890). Kripke, S. (1980): Naming and Necessity. Cambridge Mass.: MIT Press. Leibniz, G. W. (1686/2002): “Discours de métaphysique / Metaphysische Abhandlung.” In: idem., Monadologie und andere metaphysische Schriften (Französisch-Deutsch), Hamburg: Meiner, pp. 2-109. Leibniz, G. W. (1714a/2002): “Principes de la Nature et de la Grâce fondés en Raison / Auf Vernunft gegründete Prinzipien der Natur und der Gnade.” In: idem., Monadologie und andere metaphysische Schriften (FranzösischDeutsch), Hamburg: Meiner, pp. 152-173. Leibniz, G. W. (1714b/2002): “Monadologie.” In: idem., Monadologie und andere metaphysische Schriften (Französisch-Deutsch), Hamburg: Meiner, pp. 110-151. Lewis, D. (1994): “Reduction of Mind.” In: Guttenplan, S. (Ed.), A Companion to Philosophy of Mind, Oxford: Blackwell, pp. 412-430. Lycan, W. (2009): “Giving Dualism its Due.” In: Australasian Journal of Philosophy 87 (4), pp. 551-563. Nagel, T. (1979): Panpsychism. In: Idem, Mortal Questions, Cambridge. Rescher, N. (1979): Leibniz. An Introduction to his Philosophy. Oxford: Basil Blackwell. Robinson, H. (2007): “Dualism.” In: Stanford Encyclopedia of Philosophy, http://plato.stanford.edu/entries/dualism/ (23.06.10) Rowlands, M. (2003): “Consciousness: the transcendentalist manifesto.” In: Phenomenology and the Cognitive Sciences 2, pp. 205-221. Royce, J. (1882): “Mind and Reality.” In: Mind, Vol. 7, No. 25, pp. 30-54. Seager, W. (1995): “Consciousness, Information and Panpsychism.” In: Journal of Consciousness Studies 2 (3), pp. 272–88 Seager, W. (1999): Theories of Consciousness. London u. a.: Routledge. Seager, W. / Allen-Hermanson, S. (2005): “Panpsychism.” In: Stanford Encyclopedia of Philosophy, http//plato.stanford.edu/entries/panpsychism/ (02.12.09)

116 Searle, J. R. (1992): The rediscovery of the mind. Cambridge, London: MIT. Simons, P. (1998): “Farewell to substance: a differentiated leave-taking.” In: Ratio. New Series 11, pp. 253–252. Strawson, G. (2006): “Realistic Monism: Why Physicalism Entails Panpsychism.” In: Freeman, A. (Ed.), Consciousness and ist Place in Nature. Does Physicalism entail Panpsychism?, Exeter u. a., S. 3-31. Strawson, G. (2009): Selves: an essay in revisionary metaphysics. New York: Oxford University Press. Zimmerman, D. (2006): “Dualism in the Philosophy of Mind.” In: Encyclopedia of Philosophy 2nd Edition, London, pp.113-122.

Mind Space: Toward a solution to the combination problem David Skrbina (Dearborn/Michigan) I begin with two fundamental, and perhaps obvious, observations. First, there is no more basic fact of existence than that of subjective experience—which is the central feature of that which we loosely call ‘the mind.’ Mentality, experientiality, qualia, qualitative experience—I take these as synonymous, and as representative of this one most-basic truth of reality. Each person knows this truth more deeply and more intimately than anything else. It is an ancient truth, one which crosses many divisions within philosophy. It is the common thread that links disparate thinkers and times, from Anaxagoras and Parmenides to Berkeley and Descartes and many others. Normally one would expect that no reasonable person could question this fact, although certain eliminativists, analytic philosophers, and hard-core materialists come perilously close to doing so. Secondly, I take it that some form of metaphysical monism must be true. This view has widespread support today, if only because few are willing to defend the alternatives—particularly full-blown (classical) substance dualism. The problems of dualism are well-known: interactionism, the causal closure of the physical world, and various untenable theological implications, all conspire to make it virtually unacceptable. In abandoning (true) substance dualism, we are of practical necessity left with various forms of monism.1 The leading candidates would include: conventional physicalism (taken as synonymous with materialism), neutral monism, and so-called ‘property dualism,’ which I prefer to call dual-aspect monism. Neutral monism is an intriguing option, but we lack a well-articulated version at present. Hence I take it that dualaspectism is the leading challenger to the dominance of physicalism. If we accept, then, that mind or experientiality is the most basic fact of existence, and that some form of monism must obtain, then our path forward narrows. First of all, if the mental is primary, then the physical—the material—is secondary; it is something of which we are less certain, and less able on which to make definitive pronouncements.

1

No one in the past century, at least, has been willing to defend any form of pluralism, and thus I leave that option aside.

118 Physicality is the inferred reality, whereas mind is the experienced reality.2 The question then follows: Is the physical derived from the mental? If so, we arrive at a form of idealism. If not, then the physical must have something like equal standing with the experiential. Along with most others, I have a hard time accepting the former, for much the same reason that pure (mechanistic) materialism is unacceptable: in the words of Nagel (1974: 446), “we do not at present have any conception of how it might be true.” Thus we are driven to accept the equiprimordial standing of the mental and the physical, yielding a monism that has irreducible experiential and physical qualities—which is none other than dual-aspect monism.3 We can accept this despite the fact that the mental appears to us as the more fundamental reality; that is, the mental is epistemologically primary, though without being ontologically primary. The vast majority of philosophers, in accepting conventional (nonexperiential, mechanistic) physicalism, must thereby adopt two unsavory positions: (a) they must reject the powerful intuition that it is the mental that is the more fundamental reality, rather than the physical; and (b) they must embrace some form of mental emergentism—thereby incurring the burden of explaining how the experiential can arise from, or emerge from, a wholly non-experiential physical reality. Intuitions can be rationally overridden, but the thesis of emergentism is a different matter altogether. Galen Strawson, in his contemporary classic “Realistic monism: Why physicalism entails panpsychism” (2006, 2009), gives an incisive formulation to the notion that brute emergence of mind is, quite literally, incomprehensible. His conclusion: “I think it is very, very hard to understand what it is supposed to involve. I think it is incoherent, in fact…” (2009: 41)—that is, how can it be that something like experience can possibly emerge from that which is utterly devoid of experience? I won’t recount the whole case against emergence here, but I do want to emphasize three salient (and interrelated) aspects of this problem. In order to be comprehensible, any theory of emergence must offer 2

This obviously flies in the face of conventional materialism, in which the physical is the ultimate reality and the mental is derivative, or emergent—as I will explain below. 3 Of course, there could be (many) other, equally-fundamental aspects of the one monistic reality, or even infinitely many—as Spinoza presumed. I hold this as an open conjecture, but one which does not substantially affect the argument to follow.

119 explanations for what I have called the historical, phylogenic, and ontogenic questions. First: If, at some time in the distant past, mind (experientiality) did not exist at all, and now it does, then we would like to know roughly when in the course of evolution it appeared, and why. Presumably there was a reason for mentality, some evolutionary advantage, that facilitated its sudden—and I emphasize ‘sudden’— appearance in the natural world. The emergentist should be able to offer at least a tentative proposal regarding when and why mind first appeared on this planet. Second, of the wide variety of organisms that exist on Earth today, we would like to know which ones possess this experiential quality, and which do not (the phylogenic question). Few today would defend the view that humans alone are enminded; Griffin (2009) gives compelling empirical data for nonhuman minds. But once we extend mind to nonhumans, the question is begged: How far do we go? Either we go ‘all the way,’ and include all nonhumans, or we draw a line somewhere—with appropriate justification, of course. But there’s the rub; Velmans (2009), for example, demonstrates the intractable difficulty of drawing such lines in the phylogenic order. And in truth, no one thus far has been able to establish a defensible breakpoint. Without such demarcation, the way is open for some form of panpsychism. The third question, regarding ontogenics, is best illustrated by a developing fetus. If a fertilized, one-celled human egg has no mind, but a nine-month-old newborn does, when in the process of development does mind appear? How many cells does the fetus need, how many functioning neurons, in order for it to begin experiencing—not ‘like a human,’ but simply at all? Again, one can imagine that there is no clear line to be drawn in the process of ontogenesis that would mark off the sudden emergence of mind. And without such a line, experientiality goes back to the egg—and likely beyond. All three questions are at present unanswered, on the emergentist thesis. All are related to the basic issue of the requisite structural framework— physiological, biological, or physical—that can support mind, consciousness, and experientiality. What this is, we do not know. It could simply be a matter of ignorance; and indeed this is the standard claim: ‘There must be a line somewhere, but we do not know where it is, nor even how to determine this.’ But the faith remains that the line is there, after all. The panpsychist counterclaim is that this question is epistemologically indeterminate because it is ontologically

120 indeterminate; in other words, the line can’t be drawn because it’s not there. Once we break out of the human sphere, the ball is rolling and it is very hard to stop. (Perhaps we can sympathize with the nice, clean distinctions of Descartes’ day…) Given these problems with emergentism, a panpsychist dual-aspectism is certainly a viable option. And if emergence is found to be fundamentally incoherent, then dual-aspectism becomes a clear favorite. Under this presumption, the one monistic reality is intrinsically experiential, and thus panpsychism obtains. I will take this to be the case. *** At this point, an important issue arises regarding the compositional nature of the mind. In our own first-hand experience, mind is a singular sense of conscious awareness, characterized by a rich and diverse sensory imagery. But on the dual-aspect thesis, experiential mind also resides at the level of physical ultimates. This implies two immediate questions: (1) How many levels of mind exist between the ultimates and ourselves? (Or for that matter, ‘above’ ourselves?); and (2) What is the relationship between these ultimate / intermediate minds, and our singular, higher-order self? These points need to be addressed by any adequate theory of panpsychism. The usual assumption is that, on the panpsychist view, higher-order minds are composed of lower-order ones, in much the same way, or in an analogous way, that macro objects are composed of physical (atomic) ultimates. Composition of physical ultimates is relatively straightforward; the mass of the whole is the sum of the masses of its ultimate parts. And the physical characteristics of the whole are entailed by the properties of its ultimates, and the corresponding physical laws. However, the notion that a higher-order mind could be the result of a composition of mental ultimates—or we might say, of the mental characteristics of physical ultimates—is mysterious, and threatens to undermine panpsychism at its foundation. We lack a conception of how this could be true. Thus, the combination problem—which some (e.g. Strawson) call the composition problem—is perhaps the most significant issue facing panpsychist theories today. We need to be clear as to what precisely the problem is. At a minimum, we have the problem that the human body,

121 with its singular sense of mind and consciousness, is composed of myriad atomic particles, each with their own corresponding experiential lives. What is the relationship between these lesser selves and my unitary sense of consciousness? Am ‘I’ composed of these atomic selves? Do I supervene on them? Is there a causal relationship between myself and them? How can billions of minds unify into one? And perhaps most importantly, how can billions of incredibly simple minds combine to form one rich, diverse, multi-modal consciousness? To further complicate the picture, I must allow for the possibility that the various organs, subsystems, and other composite structures in my body may also possess singular subjective selves—or at least participate in mental processes. Could this really be possible? William James certainly thought so: “To sum up, mental facts can compound themselves… My arm-feelings can be, though unnoticed… They can also be noticed and cooperate with my eye-feelings in a total consciousness of ‘my arm’” (in Perry 1935: 765). Some two centuries earlier, Leibniz and Diderot observed similar points. Leibniz wrote, “I even maintain that something happens in the soul corresponding to the circulation of the blood and to every internal movement of the viscera, although one is unaware of such happenings…” (1704/1996: 116). Diderot said, “In all seriousness, [I] believe that the foot, the hand, the thighs, the belly, the stomach, the chest, the lungs, the heart, have their own particular sensations…” (1769/1937: 85). As well Schopenhauer believed that the parts of the body were individual and discrete ‘objectified wills’.4 In support, he cites a passage from Parmenides: “for in each and all men ‘tis one thing thinks—the substance of their limbs” (in Aristotle, Metaphysics, 1009b20). And in a similar vein, Nietzsche wrote: “our body is only a social structure composed of many souls” (1886; sec. 19). On the most liberal reading of the panpsychist thesis—which I adopt— the body indeed has innumerable lesser selves: organs, cells, 5 macromolecules, proteins, atoms, and so on. All of these (except the atomic ultimates) are themselves composed of lesser selves, and all

4

See World as Will and Idea, section 20, for an especially vivid description. Process philosophers take issue with this, holding to a much more restrictive form of panpsychism (‘panexperientialism’) that excludes aggregates and artificial composite structures. For a further discussion of the problems with this view, see my (2008).

5

122 participate in higher-order minds—with ‘my self’ serving as the peak of this mental hierarchy. The two problems cited above become clear; to reiterate: (1) How do selves ‘combine’ at all?, and (2) How can my higher-order self consist of, ultimately, a combination of billions of incredibly simple atomic selves? Let me interject here that I do not see the combination problem as a formal objection to panpsychism—even though it is often stated as such.6 In fact it is no objection at all to panpsychism per se. More properly understood, it is a call for details. At best, it may stand as an objection to certain specific forms of panpsychism that may fail to adequately account for this relationship. But in general we should treat it as a (reasonable) demand for explication. *** In order to get a better handle on this issue, it’s probably best to take a quick look at how it was addressed by our predecessors. The earliest discussion of the topic occurs in Leibniz, in his first elaboration of the idea of the ‘dominant monad’ (1686).7 Certain objects, he said, have “a thoroughly indivisible and naturally indestructible being”; it is these things that possess a single, higher-order, dominant monad, which functions as the ‘soul’ (or in certain cases, mind) of the thing. (We recall that every individual monad was, for Leibniz, formed on the model of the soul.) But exactly how this process of domination or unification occurs, he does not say. Nor does he offer any criterion by which to evaluate degree of divisibility or destructibility. In the mid-18th century, Kant briefly addressed Leibniz and the combination problem in a footnote of his early work Traume der Geisterseher (Dreams of a Spirit-Seer; 1766). He is discussing the concept of an ‘inner reason’ for the efficient causality found in matter, when he adds this comment: Leibniz says that this inner reason...is the power of conception [i.e. intelligence], and later philosophers received this undeveloped thought with 6

Including by myself: see my book Panpsychism in the West (2005: 264). Though the term ‘monad’ would not appear until 1698, in the work On Nature Itself. But the basic concept was already present in Leibniz’s letter of 1686.

7

123 laughter. But they would have done better if they had first considered whether a substance of the nature of a simple particle of matter is possible without any inner state. [If so, they would have to] think out another possible inner state than that of conceptions... Everybody recognizes [that] even if a power of obscure conceptions is conceded to...matter, it does not follow thence that matter itself possesses power of conception, because many substances of that kind, united into a whole, can yet never form a thinking unit. (p. 54)

Evidently, then, Kant believed mental combination to be impossible. Unfortunately he does not elaborate. Just three years later, Diderot published his Le Reve de d’Alembert (1769), observing that a swarm of bees functions like a single organism even though it is nothing more than a myriad of individual creatures in close contact: “This cluster is a being, an individual, an animal of sorts” (1769/1937: 67). Likewise, the human body is a kind of swarm of organs, “[which] are just separate animals held together by the law of continuity in a general sympathy, unity, and identity.” These individually-sentient organs then somehow combine to create the whole human. This happens, Diderot said, through “continual action and reaction;” that is, through constant exchange of matter and energy. This is sufficient for unification to occur; “it seems to me that contact, in itself, is enough” (ibid: 76). In 1890, James’ Principle of Psychology tackles the issue, coming down against the possibility of combination. It is “logically unintelligible” he says, because combination cannot create an utterly new feature of reality—namely, a single unified mind. We recall this well-known passage: Where the elemental units are supposed to be feelings, the case is in no wise altered. Take a hundred of them, shuffle them and pack them as close together as you can (whatever that might mean); still each remains the same feeling it always was, shut in its own skin, windowless, ignorant of what the other feelings are and mean. There would be a hundred-and-first feeling there, if, when a group or series of such feeling were set up, a consciousness belonging to the group as such should emerge. And this 101st feeling would be a totally new fact… (1890/1918: 160)

But James ultimately changed his mind. In his A Pluralistic Universe (1909), he offers a lengthy chapter on the “compounding of consciousness.” Empirically, combination must happen; otherwise “it

124 makes the universe discontinuous” (p. 206)—that is, there would be no higher-order minds at all, when in fact there obviously are. We ourselves are proof that combination must occur. After James, it would be nearly 100 years before this problem was again taken seriously. Especially since the mid-1990s, we see growing attention paid to it, in conjunction with a generally increasing awareness and acceptance of panpsychism.8 *** So, how can we move ahead on this ‘problem’ that is, after all, a call for details? Consider again the physical realm. All the complexity of the physical world, and the macro-scale objects that populate it, derive (as far as we know) from a small number of very simple particles: quarks, leptons (i.e. electrons), and a handful of force particles (photons, gluons, and other bosons), together with the laws of physics. That the complexity of the human brain, for example, can arise from the motions and interactions of a vast number of a few types of simple subatomic particles, is astounding. In effect, size matters. Size brings hierarchical complexity, which induces greater variation. Size allows simple ultimates to jointly create intricate patterns of behavior in space and in time. With a large number of constituent parts, an object can assume a large number of different states and conditions, which vastly outnumber the limited states of its elements. In essence, each ultimate, and each level of structural hierarchy, expands the realm of possible states of the whole. Two factors, then, are critical: large numbers of particles, and a sophisticated hierarchy of structural complexity. Fewer particles, or lesser complexity, would yield a vastly simpler object—that is, an object with a much smaller universe of possible states. Conversely, the more particles and the richer the hierarchical complexity, the larger the space of possible states. This ‘space of possible states,’ sometimes called a configuration space, embodies the physical complexity of the object. Among macro-scale objects, living organisms exhibit a particularly wide range of possible states and behaviors. Compared to nonliving objects of similar mass (i.e. of similar number of elementary particles), they 8

For other contemporary discussions of the combination problem, see for example: Seager (1995), Skrbina (2005), Coleman (2006), Goff (2006, 2009), Basile (2008, 2009), and Harman (2009).

125 have a more complex and more dynamic internal structure, and thus a much larger configuration space. Compare a human being and a large rock. Each may possess, say, 80 kg worth of particles, but the state space of the human far exceeds that of the rock. In each case, though, such a space exists, and it arises from a few types of very simple particles. I would further suggest that each of these two driving factors—number of ultimates, and degree of structural complexity—opens up two primary ‘axes’ in configuration space. An object’s space has a width given by the number of ultimates in it, and a depth given by the degree of structural complexity. The volume of the total space is, then, ‘width x depth.’ An 80 kg person and an 80 kg rock both have equal state space width, but the human has a vastly greater depth (due the complexity of a living being), and thus a much greater overall space of possible states. The mental life of a rock is wide but not deep. An amoeba, by contrast, is deep but (relatively) narrow. Notions like configuration space, or state space, are accepted and widely used today, at least within the physical sciences. They have proven extremely useful in depicting the behavior of complex objects in a conceptually simple manner. I believe that we need something similar in the psychological sciences, i.e. in the philosophical study of mind and consciousness. Consider the following: If experience is a core property of matter, along with such things as mass, charge, spin, and so forth, and if the physical characteristics yield a large and complex spatio-temporal configuration space, then an analogous effect should occur with experience. That is, an object should exist both in a physical configuration space, and simultaneously in an ‘experiential space’—a mind space, if you will. Just as simple physical ultimates can yield an extraordinarily complex configuration space, so too may we conjecture that a large number of experiential ultimates, combined in a complex mental hierarchy of experiential structures-within-structures, can yield a complex mental space. The experiential ultimates, each very simple but large in number, open up a large and complex mind space for the collective whole. An image that comes to mind is a camping tent. The tent is held up by many individual ropes, each of which, by itself, can do little to open up a space within. But each rope, staked in a different direction, opens up another ‘dimension’ of space inside the tent. Many simple ropes, united together but pulling in different directions, collectively open up a large

126 internal space. In an analogous fashion, each (experiential) atomic particle and each (experiential) level of structure of a given object opens up a new dimension in its overall mental space. In this way, simple ultimates can, in truth, ‘combine’ to form a complex, higher-order mind. Furthermore, there must be a one-to-one correspondence between physical configuration space and mind space—because both refer to the same object. As an object’s physical state changes—that is, as it moves within its configuration space—so too must it move within its mind space, because mind is linked to the very same ultimates. There can be no physical change without mental change; every physical change occurs simultaneously with experiential change. Thus, we have no dualist interaction problem. Mind and body change together, in parallel, simultaneously. Leibniz’s “preestablished harmony” was perhaps not far from the mark after all. But what about the connection, if any, between configuration space and mind space? They have one common dimension: time. Physical objects exist in space-time, and mental objects exist in, let us say, qualia-time. But it is the same temporal axis in each case. Time is the common parameter between the physical world and the mental world. In effect, it is a bridge between the two, and unifies the two realms into a single, comprehensive space. Thus, the complete object, as a mind-body whole, exists in a unified sphere that we might call the space-time-qualia (STQ) complex. And since there is a one-to-one correspondence between changes in space-time and changes in qualia-time, the temporal dimension serves as a line of symmetry, reflecting the changes in one to the changes in the other—in a non-causal manner. Time is the axis around which the physical and mental worlds revolve. The STQ complex, then, considered as a whole, is required to encompass the full range of existential reality. Only this complex maps out the complete range of conditions of a given object. Finally, we still have the question of unity. How can billions of individual experiences yield one complex mind? Why is it one mind? Let us ask the same question in physical space: How can billions of physical elementals combine to yield one complex physical object? Actually, this question is deceptively simple, but the short answer is that the particles adhere sufficiently tightly, and persist together long enough, for us to declare a given configuration as an object. Because the particles are closely linked—because they co-participate strongly— something called ‘the whole’ exists, and interacts with the world.

127 In an analogous way, these same particles, and their same hierarchical complexity, map out a mind space—a space of possible mental states—of which something called ‘the whole’ moves within. Just as long as, and to the same extent that, a given configuration of particles exists as one object, so too, and in the same degree, does there exist a singular mental object, or mind. This is hard for us to grasp because the physical whole is objective—accessible to third party observation—whereas the mental whole, or the mind of the object, is subjective and thus necessarily hidden from view. And in truth we might be fully justified in denying its existence at all, except for one bothersome fact—our own mind. It is our own singular mental existence which assures us, in the most positive manner possible, that individual physical objects possess individual internal experiential states. The existence and unity of mind is neither more nor less certain than the existence and unity of objecthood. I have portrayed objects as clear-cut entities, but this is far from the case. Physical things are ‘fuzzy,’ in the sense that, at the atomic scale, quantum fields tail off to infinity. Additionally, other particles and forces in the universe exert a small but determinate presence within any given object. And, particles are constantly joining and leaving objects, particularly living ones (particles of food, water, air, etc). Objects are thus best understood as fuzzy, quasi-persistent temporal patterns of mass/energy. On the above theory, minds must be understood in the same way. Mind has an indeterminate reach in the universe, potentially able to encompass everything. It has ‘fuzzy edges’ that tail off into the unconscious, the subconscious, and other esoteric states. And it is perhaps best understood as a quasi-persistent pattern in time, assuming only rough self-similarity combined with long-term stability.9 The combination problem is, in many ways, the toughest challenge for panpsychism. But it is also the key to its success. Any viable and wellarticulated panpsychist theory of mind will, of necessity, resolve the combination problem. It is a kind of acid test, and one that should spur us on to think more deeply about a concept of mind that finds experientiality residing at the very heart of the physical world.

9

Elsewhere (2009) I have employed the terminology of dynamical systems theory, and argued that the mind is best viewed as a strange attractor pattern in state space.

128 References Basile, P. (2008): “Is mental composition impossible in principle?” In: Chromatikon IV. Basile, P. (2009): “Back to Whitehead?” In: Mind That Abides (D. Skrbina, ed.). Benjamins. Coleman, S. (2006): “Being realistic.” In: Journal of Consciousness Studies, 13(10-11). Diderot, D. (1769/1937): “D’Alembert’s Dream.” In: Diderot: Interpreter of Nature. Lawrence and Wishart. Goff, P. (2006): “Experiences don’t sum.” In: Journal of Consciousness Studies, 13(10-11). Goff, P. (2009): “Can the panpsychist get around the combination problem?” In: Mind That Abides (D. Skrbina, ed.). Benjamins. Griffin, D. (2009): “Windows on nonhuman minds.” In: Process Approaches to Consciousness in Psychology, Neuroscience, and Philosophy of Mind. (Weber and Weekes, eds.). SUNY Press. Harman, G. (2009): “Zero-person and the psyche.“ In: Mind That Abides (D. Skrbina, ed.). Benjamins. James, W. (1890): Principles of Psychology. Dover. James, W. (1909/1996): A Pluralistic Universe. University of Nebraska Press. Kant, I. (1766/1900): Dreams of a Spirit-Seer. Macmillan. Leibniz, G. (1686/1989): Letter to Arnaud. In: Philosophical Essays (Ariew and Garber, eds.). Hackett. Leibniz, G. (1704/1996): New Essays on Human Understanding. Cambridge University Press. Nagel, T. (1974): “What is it like to be a bat?” In: Philosophical Review, 83(4). Nietzsche, F. (1886): Beyond Good and Evil. Penguin. Perry, R. (1935): Thought and Character of William James. Little, Brown. Schopenhauer, A. (1819/1995) World as Will and Idea. J. M. Dent. Seager, W. (1995): “Panpsychism, information, and consciousness.” In: Journal of Consciousness Studies, 2(3). Skrbina, D. (2005): Panpsychism in the West. MIT Press. Skrbina, D. (2008): “On the problem of the aggregate.” In: Chromatikon IV. Skrbina, D. (2009): “Minds, objects, and relations: Toward a dualaspect ontology.” In: Mind That Abides (D. Skrbina, ed.). Benjamins.

129 Strawson, G. (2006): “Realistic monism: Why physicalism entails panpsychism.” In: Journal of Consciousness Studies, 13(10-11). Strawson, G. (2009): “Realistic monism: Why physicalism entails panpsychism (revised).” In: Mind That Abides (D. Skrbina, ed.). Benjamins. Velmans, M. (2009): “The evolution of consciousness.” In: Process Approaches to Consciousness in Psychology, Neuroscience, and Philosophy of Mind. (Weber and Weekes, eds.). SUNY Press.

There is No Combination Problem Philip Goff (Hertfordshire) Emergentist and reductive forms of panpsychism We can divide things that exist into two categories: metaphysically heavyweight entities and metaphysically lightweight entities. The heavyweight entities are those into which nature is carved in and of itself; the lightweight entities are mere shadows cast by our discourse. To put it less metaphorically: the heavyweight entities are the entities upon which truths depend, the lightweight entities are those entities which depend on truth. For those enemies of realism who hold that truth is grounded in coherence or usefulness, all entities are lightweight: the sentence 'The moon exists' is grounded not in the existence of some object, the moon, but in the coherence the proposition expressed by the sentence has with other propositions, or in the usefulness of believing that proposition. But those who believe that truth is ultimately grounded in the world need not take all entities to be heavyweight. It is highly plausible to think that the truth of the sentence 'There is a party at Bill's' does not require of the world that there be some thing, Bill's party; it requires merely that there be people having a good time at Bill's. If this is the case, then Bill's party is metaphysically lightweight: its existence is grounded in, rather than itself being something that grounds, truth. For reasonable realists the same is surely true of wars, street fights, the European Union and Beethoven's 5th. I take physicalism to be a form of realism according to which the only metaphysically heavyweight objects are the non-conscious objects of fundamental physics; all other objects, such as tables, planets, rocks and conscious subjects, are mere shadows cast by our language. I take the defining commitment of panpsychism to be to the heavyweight existence of mereological ultimates with conscious experience. We can, at least in the first instance, distinguish between two distinct ways to go on from that initial commitment. Let us call 'panpsychists reductionists' those who take these conscious ultimates to be the only metaphysically heavyweight things: sentences about the existence and conscious experience of macrolevel subjects, e.g., 'Bill is feeling anxious', are made true not by macrolevel subjects instantiating phenomenal qualities but by facts about conscious ultimates, just as 'There is a party at Bill's house' is made true not by a party but by facts about partiers. For panpsychist reductionists macro-level conscious subjects are metaphysically lightweight, in

132 something like the way the physicalist takes all conscious subjects to be metaphysically lightweight. In contrast, we can define 'emergentist panpsychists' as panpychists who believe that both conscious ultimates and macro-level conscious subjects are metaphysically heavyweight: 'Bill is feeling anxious' is made true by some single subject of experience instantiating the phenomenal quality of feeling anxious (and not by any more fundamental state of affairs). Taking macro-level conscious subjects to be heavyweight does imply substance dualism; it is coherent to suppose that certain physical objects, wholly composed of physical parts, such as brains or central nervous systems, are metaphysically heavyweight. If a brain is metaphysically heavyweight, then, although it is wholly composed of physical parts, it is utterly irreducible to those parts; we can call such a composite physical object an 'emergent whole'. Emergenist panpsychists identify the subject of Bill's experience with some emergent whole in the physical world.1 An emergent whole is irreducible to its parts in the sense that truths about its existence and nature are not made true by facts about its parts. Contrast again with Bill's party. Truths about the existence and nature of Bill's party, truths such as 'There is a party at Bill's tonight' or 'Bill's party was wild!', are made true by entities at a more fundamental level, i.e. people (having a good time). Emergent wholes are not like that: truths about their nature and existence, such as 'Bill thinks therefore he is', or 'Bill is feeling anxious', are made true by facts about the emergent whole itself and its qualities, and not by facts about its parts. The physicalist and the reductive panpsychist think that all macro-level entities are reducible to micro-level facts, in the same way that Bill's party is reducible to facts about partiers. In contrast, the emergentist panpsychist holds that macro-level conscious subjects are emergent wholes. Against reductive panpsychism I think reductive panpsychism is a position that is difficult to maintain. One becomes a panpsychist because one is metaphysically serious about 1

I remain neutral on the question of whether 'the subject of Bill's experience' is identical to Bill. It would be plausible to think so if we took the subject of Bill's experience to be identical to a whole organism. It is less obvious what we should say if we think that the subject of Bill's experience is identical to a part of an organism, for example a brain. In this case, I suspect it is indeterminate whether Bill's uses of the first person pronoun refer to the organism or the brain, but obviously these are deep matters which deserve more attention (see Strawson 2009 for a little more detail on these issues, and a defence of the view I incline towards).

133 consciousness. It seems to me that this metaphysical seriousness ought to be credited either to both macro-level and micro-level subjects alike, or to neither. Just as we have divided objects, so we can divide properties, into heavyweight and lightweight. Heavyweight properties constitute the joints of nature; to the extent that objects share heavyweight properties they genuinely resemble, that is, they resemble in and of themselves. Lightweight properties are merely shadows cast by predicates. An electron and a horse do not genuinely resemble, do not resemble in and of themselves, in virtue of both possessing the property of being either an electron or a horse. But it is plausible to think that two electrons genuinely resemble, resemble in and of themselves, in virtue of both being negatively charged. If these intuitions are correct, being negatively charged is a heavyweight property, whilst being either negatively charged or a horse is a lightweight property. Lightweight properties are those which are dependent on the satisfaction of predicates, whilst heavyweight properties are those upon which the satisfaction of predicates depends. An electron has the property of being negatively charged or being a horse in virtue of satisfying the predicate 'being either negatively charged or a horse', which it in turn satisfies in virtue of being negatively charged. When it comes to the property of being negatively charged, however, it is plausible to think that things go the other way round: the electron satisfies the predicate 'being negatively charged' in virtue of being negatively charged. Physicalists take phenomenal properties to be lightweight: objects have phenomenal properties in virtue of satisfying phenomenal predicates, which they satisfy in virtue of more fundamental physical facts.2 The minimal commitment of panpsychism is that the consciousness of fundamental particles is a heavyweight property: fundamental particles satisfy phenomenal predicates in virtue of having phenomenal properties rather than the other way round (from now on I will think of mereological ultimates as fundamental particles for ease of exposition). It follows from this commitment that two fundamental particles, in so far as they share phenomenal qualities, genuinely resemble each other. But 2

Can't physicalists identify phenomenal properties with physical properties, e.g. pain with c-fibres firing? Of course they can, but on my understanding of physicalism, the physicalist is obliged to take such higher-level physical properties to be lightweight; only the properties of fundamental physics, e.g. mass and charge, are metaphysically heavyweight.

134 macro-level conscious subjects like Bill and Ben also share phenomenal qualities.3 It seems to me unintelligible that the sharing of determinates of some determinable X can, in one instance, constitute genuine resemblance, whilst in another instance fail to constitute genuine resemblance, for example, it is not intelligible to suppose that two spherical things genuinely resemble in so far as they are spherical, whereas two cuboid things do not genuinely resemble in so far as they are cuboid. We have then an argument from the basic commitment of panpsychism to panpsychist emergentism: Premise 1: The consciousness of fundamental particles is a heavyweight property, and so any two fundamental particles, in so far as they share phenomenal qualities, genuinely resemble. Premise 2: Bill and Ben share some phenomenal qualities, e.g. what it's like to see red. Premise 3: For any four objects O1, O2, O3, and O4, and any determinable X, if O1 and and O2 genuinely resemble in so far as they share a single determinate of X, and O3 and O4 share a single determinate of X, then O3 and O4 genuinely resemble in so far as they share a single determinate of X, and hence in sharing a single determinate of X they share a heavyweight property. Conclusion: Bill and Ben, in so far as they share phenomenal qualities, genuinely resemble, and hence in their sharing of phenomenal qualities, Bill and Ben share heavyweight properties. I conclude, therefore, that the panpsychist is obliged to take phenomenal qualities to be metaphysically heavyweight. Even if the argument above is sound, couldn't the panpsychist be an emergentist about phenomenal qualities, but not about conscious subjects, i.e. the objects that instantiate phenomenal qualities? I find this kind of hybrid position unintelligible. It entails that a lightweight object instantiates a heavyweight property, which would be analogous to Bill's party instantiating negative charge. How can a linguistic shadow instantiate a joint of nature? To drop the metaphor: I don't think we can make sense of the idea that an object dependent on the truth of our discourse can instantiate a quality upon which the truth of our discourse depends. 3

Bill and Ben might be organisms or might be brains of organisms depending on what we identify macro-level subjects of experience with.

135 I conclude, therefore, that panpsychists are obliged to be emergentists about conscious subjects. Why there is no combination problem Let us return to the canonical expression of the combination problem in William James: Take a hundred of them [feelings], shuffle them and pack them as close together as you can (whatever that may mean); still each remains the same feelings it always was, shut in its own skin, windowless, ignorant of what the other feelings are and mean. There would be a hundred-and-first-feeling there, if, when a group or series of such feelings where set up, a consciousness belonging to the group as such should emerge. And this 101st feeling would be a totally new fact; the 100 feelings might, by a curious physical law, be a signal for its creation, when they came together; but they would have no substantial identity with it, not it with them, and one could never deduce the one from the others, nor (in any intelligible sense) say that they evolved it. (James 1983, 162)

What does James mean when he says that the 101st feeling would not have a 'substantial identity' with the 100 feelings, that it would not have 'evolved' from them? I suggest that we can capture the spirit of what James meant in terms of the framework developed above. I interpret James as claiming that the relationship between the 101st feeling and the 100 feelings is not like the relationship between Bill's party and his guests. Once you shuffle round Bill's guests in a certain way – give them drinks, pump up the music, encourage them to engage in inane party banter – you've thereby got a party. The fact the guests are arranged in this way, and the fact that there is a party, are not metaphysically distinct states of affairs. The fact that the guests are so arranged makes it true that there is a party. On my interpretation of James, he is struggling to make sense of an analogous relationship between the hundred feelings and the hundred-andfirst-feeling. James can't see how the fact that there are a hundred feelings arranged in a certain way could make it true that there is a hundred-andfirst-feeling. These seem to be metaphysically distinct states of affairs, even if, as a matter of brute fact, one emerges from the other, perhaps on account of some 'curious physical law'. In other words, what James is struggling to find intelligible is panpsychist reductionism, the view that truths about the existence and nature of highlevel conscious subjects are made true by truths about lower-level subjects. He has no concerns about the intelligibility of panpsychist emergentism,

136 the view that facts about the existence and nature of high-level conscious subjects, as a matter of brute fact or natural law, arise from facts about the existence and nature of micro-level conscious subjects. The combination problem, at least as understood by James so interpreted, is a problem only for panpsychists reductionists, not for panpsychists emergentists. If I am right that all panpsychists ought to be emergentists in any case, it follows that panpsychists have nothing to worry about as regards the combination problem. The special phenomenal composition question So we panpsychists have to be emergentists. As such we don't have a combination problem. But we do have a pressing question to answer. I call this question, in homage to van Inwagan,4 'the special phenomenal composition question', and it goes like this: The special phenomenal composition question: Under what conditions do fundamental particles combine to form a higher-level subject of experience? I take there to be five places metaphysicans should look for answers to metaphysical questions: 1. Truths we know with Cartesian certainty via our immediate acquaintance with conscious experience 2. Conceptual truths 3. Common sense 4. Empirical data 5. Cost-benefit analysis of theoretical virtues I don't believe, contra Descartes, that Cartesian certainty, in and or itself, can rule out any answer to the special phenomenal composition question. Cartesian certainty tells me that there exists at least one conscious thing, but it cannot tell me whether or not that conscious thing is metaphysically simple or composed of physical parts.5 However, on the plausible 4

See van Inwagan 1990. So I disagree with Descartes that we can know that the mind is indivisible. When I introspect, the subject of my experience does not present itself as a thing with parts. However, I see no way of ruling out the possibility that, when introspecting, I am conceiving of the subject of my experience in terms of only part of its nature. Perhaps 5

137 assumption that cost-benefit-analysis will yield the result that the identification of conscious subjects with composite physical objects is preferable to the identification of conscious subjects with either Cartesian egos or fundamental particles, then Cartesian certainty and cost-benefit analysis working together can rule out one answer to the special phenomenal composition question: Nihilism about phenomenal composition: Fundamental particles never come together to form a macro-level conscious subject. If there is at least one conscious subject, i.e. the subject of my experience, and that conscious subject is identical to a composite physical object, then there is at least one occasion on which fundamental particles come together to form a conscious subject, i.e. when they come together to form the subject of my experience. What about conceptual truths? I can't see how conceptual truths can rule out any of the answers to the special phenomenal composition question (except ones which are obviously contradictory, e.g. fundamental particles form conscious subjects under condition X, and don't form conscious subjects under condition X).6 The following answer seems perfectly coherent, although exceedingly unlikely (for cost-benefit reasons to do with inelegance and arbitrariness): Small blue tablism about phenomenal composition: Fundamental particles form conscious subjects when they form small blue tables. What about common sense? I think it's pretty clear what answer common the subject of my experience has non-phenomenal as well as phenomenal aspects, such that if I were to conceive of it in terms of its whole phenomenal and non-phenomenal nature, it would be evident that it has parts. I find the idea of an object that instantiates both phenomenal and non-phenomenal aspects forming a unified nature quite beautiful, even though we can form no positive conception of such a thing. If the brain is the bearer of consciousness, and I believe it is, it must be a very beautiful thing. 6 Slight qualification: if metaphysical vagueness is incoherent, then conceptual truths can rule out the common sense answer (and probably small blue tablism as well, given the vagueness of the concept of the concepts involved in this answer). Also, perhaps some non-obvious incoherences may constrain answers to the special phenomenal composition question, e.g. if actual infinites are impossible, then the thesis that macrolevel consciousness requires an infinite number of ultimates will be inconsistent with realism about macro-level consciousness.

138 sense would give to the special phenomenal composition question: Commonsense answer: Particles form a conscious subject when and only when they form organisms (or a subset of organisms, or the brains/central nervous systems of organisms; I will ignore these alternatives for the sake of simplicity). Unfortunately, I think we have very strong reason to reject this answer, as I will now try to explain. The boundary between the organic and the nonorganic is vague. There are what we can call 'organic borderline cases' – cases where there is no fact of the matter as to whether or not we have a human organism – at the beginning and end of an organism's existence. In any particular case, there is no utterly precise point in time, after which we have a zygote, and before which we had only sperm and egg. Similarly, in each particular case, there is no utterly precise point in time after which we have a corpse and before which we have a living body. Given our macroscopic concerns, this vague boundary is barely discernible. But if we were looking at a complete description of the fundamental particles composing a human organism during, and slightly before and after, its existence, there would be no precise arrangement of particles which constituted the beginning and end of the human's existence; there would be borderline cases. If the commonsense answer to the special phenomenal composition question is correct, it follows that there are 'phenomenal borderline cases', cases where there is no fact of the matter whether or not we have a conscious subject. If the existence of an organism is necessary and sufficient for the existence of a conscious subject, and if it's sometimes vague whether or not we have an organism, it follows that it's sometimes vague whether or not we have a conscious subject. But for emergentist panpsychists, both micro and macro-level conscious subjects are metaphysically heavyweight entities: nature is carved up in and and of itself into both micro and macro-level conscious subjects. If we couple this commitment with a commitment to phenomenal borderline cases, we reach the view that the world in and of itself is vague. Emergentist panpsychism, in conjunction with the commonsense answer to the special phenomenal composition question, entails metaphysical vagueness. However, it is extremely plausible that vagueness is a semantic, rather than a metaphysical, phenomenon. To explain the fact that Bill is tall, we don't

139 need to suppose that there is some fuzzy, indeterminate state of affairs of Bill's neither having nor lacking some quality; we need only suppose that the meaning of the predicate 'is tall' is indeterminate such that there is no fact of the matter as to whether or not it applies to someone with Bill's exact height. On such an explanation, it is language, rather than the world, which is indeterminate. Such semantic explanations of vagueness are extremely plausible.7 Indeed, Horgan and Potrč have powerfully argued that metaphysical vagueness is incoherent.8 Vagueness is a massive area, and I don't have space here to adequately defend the semantic theory of vagueness. My own judgement it that the attractiveness of semantic theories of vagueness, and the dubious intelligibility of metaphysical vagueness, give us very strong reason to want to avoid any metaphysical theory which is committed to metaphysical vagueness.9 If we are to avoid metaphysical vagueness, and if we are to take macrolevel subjects to be heavyweight (as I have argued we must as panpsychists), then we cannot accept common sense's answer to the special phenomenal composition question. What about empirical data? Perhaps a non-panpsychist might try to argue that the behaviour of a system can constitute evidence as to whether or not that system is conscious, but it is difficult to see how the panpsychist could claim this, given that she thinks that fundamental particles are conscious. And so, having exhausted all other possible sources of metaphysical truth, we find ourselves having to turn to cost-benefit analysis of theoretical virtues to find the answer to the special phenomenal composition question. The two most theoretically elegant answers to the special phenomenal composition question are 1. Nihilism about phenomenal composition: Fundamental particles never come together to form a macro-level conscious subject. 2. Universalism about phenomenal composition: All arrangements of fundamental particles result in their composing a macro-level conscious subject.

7

See for example, Dummett 1978: 260, Fine 1975, Russell 1923. Horgan and Potrč 2008, Ch. 2. 9 It is worth noting that the phenomenon that philosophers call 'vagueness' is very different from the kind of indeterminacy postulated by standard interpretations of quantum physics. 8

140 We have already ruled out nihilism. Thus, by a process of elimination, we end up with universalism. Universalism entails that there is a conscious object formed of the particles composing my teeth, the particles composing your toe nails, and the particles composing Venus. Isn't this view wildly counter to common sense? Of course it is, but the commonsense view cannot be true, assuming reality in and of itself is not vague, and so we must look elsewhere for the truth. Having exhausted all other sources of metaphysical knowledge, we find we must turn to cost-benefit analysis, and this criterion of theory choice clearly favours universalism and nihilism. Assuming the falsity of nihilism, our only rational choice is universalism.10 Conclusion Panpsychists should be emergentists, from which it follows that (i) they should stop worrying about the combination problem, (ii) they should be universalists about phenomenal composition.11 References Fine, K. (1975): 'Vagueness, truth and logic', Synthese, 30: 265-300. Horgan, T. and Potrč, M. 2008. Austere Realism, Cambridge, MA., London: MIT. Lewis, D. (1986): On the Plurality of Worlds, Oxford: Basil Blackwell. James, W. (1983): The Principles of Psychology, Cambridge, MA., London: Harvard University Press. Russell, B. (1923): 'Vagueness', Australasian Journal of Philosophy, 8492, reprinted in Keefe, R. and Smith, P. 1996. (Eds.) Vagueness: A Reader, Cambridge, Mass.: MIT, 61-8. Strawson, G. (2009): Selves: An Essay in Revisionary Metaphysics, Oxford, New York: Oxford University Press. Van Inwagan, P. (1990): Material Beings. Ithaca and London: Cornell University Press.

10

My argument for universalism about phenomenal composition is very similar to David Lewis's argument for universalism about composition (Lewis 1986: 212-13). 11 I am very grateful to all the participants at the 'The Mental as Fundamental' conference in Vienna for helpful comments.

Panpsychism as Paradigm Freya Mathews (Victoria) I come to panpsychism not from an attempt to explain consciousness but from an attempt to explain the world – to explain what kind of world this world of ours is, and consequently how we ought to fit into it. My entry point has been, in other words, not through the philosophy of consciousness but through the philosophy of nature, philosophy of nature construed as the attempt to understand the metaphysical make-up of the world at large. The philosophy of nature has become urgently relevant in our time because the environmental crisis is pointing to a misalliance between humanity and its world. Panpsychism has thus for me been a project of metaphysical rehabilitation conceived within the discourse of environmental philosophy. The argument that I shall put forward in this chapter however focuses strictly on the metaphysical rather than the environmental. The gist of the argument is that a holistic or cosmological version of panpsychism, according to which the universe as a whole is the ultimate locus of mind, or of mind-like properties, can function as a rival to materialism, materialism being understood as the view which denies that mind, or any mind-like property, inheres in an essential way in matter or in other fundamental elements of physical reality. Moreover, I shall suggest that, in relation to materialism, cosmological panpsychism functions not merely as a rival theory but as a rival paradigm. I make this suggestion because materialism generates a number of intractable anomalies, anomalies that have become so entrenched in the philosophical tradition of the West as to seem inevitable. We perhaps forget that they are anomalies, and treat them instead as the very substance of metaphysics. It might be in consequence of this conflation that we have concluded that metaphysical questions are in principle undecidable, and consequently not worth pursuing. Tackling these intractable questions from the viewpoint of an alternative paradigm might then have implications for metaphysics itself as a discipline. I would like to emphasize that it is panpsychism in a holistic or cosmological guise, as exemplified in, for instance, Spinoza or Schelling, rather than in the more “distributed” guise favoured by process theorists, that can serve as an anomaly-dissolving alternative to materialism. Perhaps there are parallel arguments for the distributed panpsychism of process theory, but this is not a question I shall be taking up here.

142 I shall address four specific metaphysical anomalies. In each case I shall argue that these are anomalies for materialism but are far less problematic for cosmological panpsychism. The arguments as I present them here will be very abbreviated but can be found in more developed form elsewhere in my work.1 1. Problem of realism, or of the appearance/reality distinction 2. Problem of why the universe hangs together, or, more narrowly, the problem of causation 3. Problem of why there is something rather than nothing 4. Problem of the origin of the universe, or of a beginning to time Of course, the hard problem of consciousness, which I have not listed, is also a pre-eminent anomaly for the materialist paradigm, an anomaly which panpsychism can make some claim to solve. But if it can be shown that materialism harbours other anomalies, and that cosmological panpsychism solves, or at least softens, these, this independent evidence for panpsychism strengthens it as a contender in the case of the hard problem. Moreover, a sense of the cosmological reach and origins of consciousness will provide a new and illuminating context for the investigation of our own human consciousness. In both these respects then exploring cosmological panpsychism as paradigm is relevant to the hard problem of consciousness. 1. Problem of Realism I would like to start with the problem of realism, though I don’t need to spend too much time on this as it was anticipated by both Schopenhauer and Bertrand Russell, and has been reviewed in contemporary discussions. (Skrbina 2009) The problem may be set out as follows: the language of physics affords us no way of intrinsically characterizing the difference between a real physical entity and a merely apparent (perhaps illusory) one. Take the property of solidity, for instance. It is common sense to suppose that a body is real if it is solid, but solidity cannot be characterized in intrinsic terms: there is nothing we can identify in a solid body itself that renders it solid. Solidity is rather defined extrinsically in terms of impenetrability: a body is solid if it can keep other bodies out. But as an account of the realness of a body, of its actually occupying space as opposed to merely 1

See Mathews 1991 and 2003 in particular.

143 appearing to occupy it, this is clearly question-begging: a body is solid, in the sense of real, only if the bodies it keeps out are themselves already solid. There is no reason why an order of illusory bodies should not be such that they appear to keep one another out. Their doing so however will not render them solid, in the sense of real. All the properties assigned to matter in physics – mass, momentum, charge, and so on –turn out, like the common sense property of solidity, to be defined in this extrinsic way, and as such may be question-begging as accounts of what it is for a body to be really there – for it to be really real as opposed to merely apparent or illusory. The only way we could ever distinguish the real from the illusory, conceptually speaking, is to ascribe to the real some form of reflexive interiority or presence-to-itself, such as we ourselves enjoy. We have no problem conceptually distinguishing a real person from an illusory one, because we know that though the illusory person may look the same as the real person, there is nothing going on in her. In the real person, by contrast, there is definitely something going on. It may not necessarily be thought – the real person may, for example, be asleep; but there is in her an unceasing psychophysical movement towards an end, the end of selfexistence, which, though it may not be conscious at all times, is definitely intentional, self-directional, conative. The real person is psychophysically self-activated and present to herself, through introspection, in a way that the illusory person, qua mere appearance, is not. There is, in other words, a way we can describe the real person that clearly and definitively marks her off from any illusory counterpart. To ascribe a comparable self-presence, or reflexive interiority, to matter generally seems to be the only means available to us by which we might escape from the conceptual impasse of the appearance/reality distinction. This is a means unavailable to materialists, with their denial of mindlike properties to matter. It follows that only panpsychists – who are indeed prepared to ascribe mindlike properties to matter – can solve this conceptual anomaly of the appearance/reality distinction at the heart of materialism. There is a second, epistemological form of the problem of realism, but it would take me too far afield to detail it here.2 I shall instead explain why I think the conceptual version of the argument from realism suggests not only panpsychism per se, but a global or cosmological form of panpsychism. 2

See Mathews 2003, chapter 2.

144 Argument to global panpsychism Let us allow that the argument from realism has shown that matter is endowed with interiority – we might call this its subjectival dimension. However, within our accepted Western framework for thinking about matter, viz physics, the domain of physical reality is no longer conceived as merely coextensive with that of matter but includes forces and fields and even space itself. In relation to these nonmaterial aspects of physical reality, the problem of realism still arises: how are we to distinguish between real and merely illusory light, for instance? Can we say of light that it too has a subjectival dimension? If the argument from realism is to succeed, an extension of subjectivity to physical reality generally, rather than its restriction merely to matter, does seem to be required. But to take this step forces us to face the question of the relation of subjectivity to the subject. Subjective experience is, after all, the province of a subject. However, a subject, understood as a centre of subjectivity, is necessarily an indivisible unity: there are no scattered subjects, and the boundaries between subjects are not nominal. The individuation of subjects, or centres of subjectivity, is objectively determined: a thought objectively belongs to you or me; it is not up to a third person, qua knower, to decide where the boundaries of our respective subjectivities will be drawn. However the individuation of objects, at the macro-level, at any rate, is not consistently objectively determined in this way. Matter is not really, in any ontological sense, parcelled up into convenient units or packages, despite the plethora of discrete artefacts in our own daily life that suggest that it is. Indeed, many of our individuations – of rocks and mountains, for instance - have basically nominal status. We accordingly have to ask whether the physical realm could be externally divided up so as to correspond with an internal differentiation into a manifold of individual subjects. Where does the subjectivity of a mountain range, for instance, lie? In the individual mountains or in the range as a whole? In the rocks that make up the mountain, or in the underlying crust from which the rocks have become detached? Where do individual mountains end and the range begin, and how many of the individual rocks are still attached to the crust, and to what extent is the range an extension of the crust? Are there many subjects here or one, and if one, is that one the earth’s crust as a whole, or perhaps the entire body of the planet? Matter at the macro-level is not given in indivisible unities the way subjectivity is. And the question of how to divide reality up into subjects, or centres of subjectivity, to which materiality is subjectively present, becomes acute

145 when the attribution of subjectivity is extended to the physical realm generally. For there is not even any intuitive presumption that the nonmaterial dimensions of the physical realm – field or wavelike processes, for instance – can be carved up into units. There may be a wave-particle duality at the micro-level of physics, but the wave-like aspect is as much a part of the fabric of reality as the particle-like aspect is. And the wave-like aspect does not lend itself to differentiation into units. How then, again, can physical reality as a whole, inclusive of its wave and field-like aspects, be externally differentiated consistently with an interior differentiation into a manifold of subjects? The most effective way for a panpsychist to reconcile the internal ontological unity and indivisibility that properly belongs to subjects with the generally merely nominal unity of physical entities may simply be to adopt a holistic approach to physical reality. If physical reality as a whole, under both its material and non-material or field-like aspects, is seen as constituting a genuine, indivisible unity, then it could itself perhaps be regarded as a subject, or field of subjectivity, to which the entire differentiated physical manifold is subjectively present. In this case, while matter generally could be said to be present to itself, objects individually could not be said to be so. That is to say, when we regard the universe as a whole as the prime locus of subject-hood, we face a combination problem in reverse: how are the entities, objects or beings we normally regard as distinct subjects to be individuated within such an all-encompassing holism? The problem is not, as it is for process philosophers, how are compound consciousnesses to be built up out of simple ones, but rather how can local, individual subjects come to differentiate themselves within the matrix of a global mind. (For extensive discussion of the combination problem, see Skrbina 2009.) In order to address this “combination problem in reverse”, we need to sketch in the outlines of cosmological panpsychism, or at least some version of it. Since I do happen to have a version to hand, let me provide such a sketch, and in the process explain how this theory deals with the combination problem in reverse, or, to give it a name of its own, the individuation problem. Cosmological panpsychism: a version I shall start with cosmological panpsychism under its physical aspect. From this point of view, physical reality as a whole, including both its material and its non-material aspects, such as space and electromagnetic or

146 gravitational energy, forms an unbounded, indivisible, substantival (though not in the first instance material) plenum. This plenum is construed geometrodynamically, as a dynamic extended substance – space – in a continuous process of expansion and internal self-differentiation. The model is the age-old one of water (shades of Thales here)(how nice it would be if the very first philosopher got it basically right!): the universe may be compared with a vast ocean coursed continually by currents and waves, some of which interfere to become vortices which hold their structure for long enough to give the appearance of independent or enduring existents. (W. K. Clifford articulated this “space theory of matter”, as a metaphysical template, long before Einstein and later John Archibald Wheeler gave it mathematical form as geometrodynamics. (Clifford 1876, 125-6) And although geometrodynamics has not yet been accepted into physics, due to lack of experimental confirmation, it still makes eminent sense as a metaphysical template.) This geometrodynamic plenum is holistically rather than aggregatively structured, and those internal differentia which are not only stable in their configuration, but actively self-realizing, qualify as what I call selves. Selves are defined, in systems-theoretic terms, as systems with a very special kind of goal, namely their own self-maintenance and self-perpetuation. On the strength of their dedication to this goal, such self-realizing systems may be attributed with a drive or impulse describable as their conatus, where conatus is understood in Spinoza’s sense as that “endeavour, wherewith everything endeavours to persist in its own being”. (Spinoza 1951, Part III, Prop VI, Proof) Selves then enjoy a real though relative individuality even though they exist in the context of an undivided whole. Since they proactively seek from their environment the resources they need to actualize and maintain their structure while at the same time resisting causal inroads into their integrity, they count, ontologically, as individuals, even though they are not separate substances, but disturbances within a global substance. Moreover, the interference patterns which create these relatively stable configurations in the plenum are relational: it requires a very special “geometry” in the surrounding field to create the conditions for such selfperpetuating “vortices”. The paradigmatic instances of selfhood, in the present sense, are of course organisms, constituted in the relational matrices of ecosystems. The systems-theoretic criteria of selfhood – selfregulation, homeostasis, goal-directedness and equifinality – may also turn out to apply to higher order biological systems, such as ecosystems and the

147 biosphere. Indeed, it may be argued that the cosmos itself satisfies these criteria, since it is necessarily self-actualizing and self-regulating, and its self-structuring follows the relational dynamics of systems. (The details of this argument can be found in Mathews 1991.) Overlaying systems theory on geometrodynamics then, we arrive at a universe which is One, substantivally speaking, but which also selfdifferentiates, selectively, into a relative Many, the Many being those “interference patterns” which correspond to self-realizing systems, or selves. The systems criterion, set in a geometrodynamic framework, thus renders inanimate matter mere backdrop to the true individuals, the selves, in this scheme of things. Rocks and clods of clay and grains of sand are, from this point of view, not really things in their own right, but rather knotty bits of the matrix or plenum. Matter is thus properly described in mass terms: earth, rock, sand, water, sky, air, etc. The only genuine individuals that populate this world are selves. Having briefly reviewed this cosmology under its physical aspect, let us now consider it under its subjectival – which is to say its explicitly panpsychist – aspect. We can see that, even described in physical terms, this cosmology, consisting of a One which self-differentiates into a Many, points towards panpsychism, inasmuch as the One is already described as a self-realizing system, and hence as imbued with the mind-like property of conativity. But it is under the duress afforded by the Argument from Realism that we are really driven to interpret this view in panpsychist terms. Considering the present cosmology under its subjectival aspect then, what presents is an extension, a field, which, while it appears to observers (observers embedded within the field itself) as a geometrically dynamic space, is experienced from within as a field of subjectivity, a great, internally differentiated field of impulse, of intrinsic activity, of felt expansions, swellings, dwindlings, contractions, surges, urges and so forth. Impulses within this field follow certain patterns: they gather and unfold, constellate and dissipate, in objectively patterned ways, just as our own subjective impulses do. This “lawlikeness” of motion – of impulse – in the primal field may be read as correlative with physics. To read subjectival process this way, as that which is manifested externally in the lawlikeness of physics, is not to deny the characteristic freedom or spontaneity of subjectivity: the characteristic patterns of subjectival movement may be acknowledged without this implying that subjectival processes are strictly predetermined. An analogy might be helpful here. Consider the movement

148 of a snake. Snakes move in a determinate way: they slither or slide, in wave-like fashion; they do not hop, gallop or fly. Yet this does not entail that the direction a particular snake chooses to take on a given occasion is predetermined. Order in the sense of the large-scale patterning of motion does not preclude small-scale variations of “direction” within such patterns. This degree of freedom within the framework of a larger order is presumably correlative with quantum mechanical “give” in physical systems. Quantum mechanics has revealed aspects of physical reality, such as wave-particle duality, complementarity and nonlocality, which, while compatible with the large-scale patterning of classical physics, allow, at the micro-level, for the indefiniteness, diffuseness, indeterminism, and sudden resolutions and dissolutions that are typically associated with mental or subjectival processes. In other words, the largescale patterning of impulse in a subjectival field may manifest externally as the lawlikeness of classical physics while an element of choice or spontaneity within that patterning may be read as correlative with quantum physics. The primal field then, from the present point of view, exhibits aspects of both the traditional physical and the traditional mental or subjectival, without being reducible to either. If this is how the global field of subjectivity may be imagined from within, from the viewpoint of the One as Subject, the question of how finite selves embedded in this larger Subject may be imagined from within remains to be considered. How can relatively distinct subjectivities, the subjectivities belonging to the differentia we have identified externally as self-realizing systems in the geometrodynamic matrix, form within the field of a larger consciousness? How do such subjects manage not to be absorbed, experientially, into the larger field, and how does the larger Subject, of which they are a part yet from which they also differentiate themselves, experience them? A psychoanalytic analogue might provide a model in this connection. Psychoanalytic theory is, of course, premised on the idea of the unconscious – that certain aspects of mental life in human beings, particularly those aspects associated with ego, are conscious, while other aspects are unconscious. Amongst the unconscious components of psyche are enduring constellations of psychophysical energy which never surface into ego consciousness yet which nevertheless may be active in the psychic life of a person. In certain circumstances these constellations may eclipse the ego as engine of agency and overtake the person in question, driving their behaviour. Carl Jung called such constellations of psychophysical

149 energy autonomous complexes, to indicate that though they are in reality part of the psyche, such complexes are unaware of this fact and to a certain extent are self-organized to lead a “split off” life of their own. Though the analogy is not perfect, it does, I think, suggest a way in which a selfrealizing system might become a relatively distinct and self-directing centre of subjectivity within the larger field of the One as Subject. And just as a person whose psyche harbours an autonomous complex is not consciously aware of the way the complex feels to itself, but is nevertheless psychically organized so as to make room for the complex, and perhaps made uncomfortable by the “gaps” that the complex leaves in her consciousness, so the One as Subject may feel the effects of finite centres of subjectivity in the field of its own larger subjectivity, even though it may not be able actually to experience the way such finite selves feel to themselves.3 These are very brief remarks, and much remains to be explained, but I hope I have said enough to show in a preliminary way how the individuation problem – which is to say, the combination problem in reverse – may be resolved in the context of the present cosmological version of panpsychism. Having completed my sketch of this version of panpsychism, let me now return to the anomalies, those problems which remain intractable for materialism but which dissolve, or at least soften, in the context of panpsychism. I have already reviewed the Argument from Realism, and explained how panpsychism can give conceptual content to the distinction between appearance and reality in a way that materialism, anomalously, cannot. Let us now turn to the question of why the universe hangs together, where this can also be read, more narrowly, as the problem of causation, as originally formulated by David Hume. 2. Why does the universe hang together? Materialism, insofar as it remains particulate, analyzes physical reality in terms of patterns of motion exhibited by elementary physical units, where these patterns are given by laws that are logically extrinsic to the units in question. In other words, the laws could be different; they could in principle cease to hold. To the extent that physics continues to source 3

Many thanks to Craig San Roque for helping me to understand, in psychoanalytic terms, the role of autonomous complexes in psychic life.

150 fields and forces to particles of various kinds, it affords no logical assurance that the universe will continue to be governed by these laws, nor hence that it will continue to hang together. The laws holding the elementary units together might simply collapse, and physical chaos ensue. In other words, such a physics cannot explain the coherence of the world; this coherence remains a fortuity. In this sense, physics remains subject to Hume’s anomaly. There can, Hume pointed out, be no logically necessary connections between distinct existences. To the extent that physics posits a universe of logically distinct existences, it cannot explain why these existences are held together by the observed laws of nature. A ready solution to this problem of causation, and to the larger conundrum of why the universe hangs together, is simply to relinquish the hypothesis of distinct existences. If the universe is not fundamentally particulate, if it is instead fundamentally an indivisible unity, or plenum, then the problem of coherence disappears. A plenum, such as space considered geometrodynamically, is intrinsically internally structured in accordance with a principle of perfect point to point connectivity and hence perfect continuity. In other words, motion in a plenum will necessarily propagate smoothly and isotropically, and hence in a fundamentally lawlike way, at least until interference patterns set in. A geometrodynamic universe is necessarily a lawlike one. (Mathews 1991, chapter 2) The problems of coherence and causation then point us metaphysically in the direction of unity and holism, but this does not yet in itself necessarily entail a commitment to panpsychism. The step to panpsychism occurs when we wonder why the universe is an indivisible unity. Granted, if it is an indivisible unity, then it will be lawlike, and will hence hang together. But if it does not have to be an indivisible unity, if the fact of its indivisible unity is itself contingent, then we have still not explained, in any ultimate way, why it hangs together. This hanging together will remain a metaphysical fortuity. And, from a materialist perspective, there seems to be no more reason, ultimately, for unity than for fragmentation: atomistic chaos is as logically possible as lawlike unity. From a panpsychist perspective however, there is every reason to expect unity rather than fragmentation. Having inferred the unity of the universe from the evidence of its lawlikeness, the panpsychist explains this unity by reference to a subjectival aspect of reality. A subject, in the sense of a centre of subjectivity, is itself intrinsically, as we have already noted, an indivisible unity. Subjectivity is an extension, a field for experience – for meaning, purpose and agency - but there are no hard edges or boundaries in this

151 field; the different instances of its experience are cross-referential and inter-permeating. Meanings, for example, are differentiated only within a larger field of meanings which overlay and inform one another; purpose gives coherence and direction to all it touches, bringing different cognitive functions together into indissoluble synergies; agency, being referenced to purpose, has a similar effect. The “hanging together” that presents itself to us as the physical face of reality thus points forcefully to the conclusion that subjectivity, so clearly indivisible in its essential nature, must, at a deeper level, be integral to the nature of things. Reality hangs together under its outer aspect because that outer aspect is indicative of an inner, subjectival and accordingly indivisible, nature. It is thus the panpsychist who offers at least the outlines of an answer to the question why the universe holds together rather than falling apart. In other words, it is the panpsychist who can shed a little light on a conundrum that is sheer anomaly for the materialist. 3. Why is there something rather than nothing? Either there is or there isn’t a reason for the existence of a world, for the existence of the “something” that can be seen as a matter of fact to exist. If there is no reason, then what exists – the world as we know it – just is as it is, absolutely arbitrarily; it could have been entirely otherwise or it could have been not at all. If on the other hand there is a reason for the existence of something, then presumably that reason involves self-causation or selfcreation, since there is ultimately nothing other than what-exists that could bring what-exists into existence. But self-causation, upon analysis, turns out to have consequences: it turns out to entrain a certain nature for the thing that is self-creating. This nature is, as we shall see, consistent with panpsychism but not materialism. Allow me, very briefly, to explain. A universe, which is self-creating, must also exhibit certain other attributes. Firstly, it must be imbued with causal power. The reason for this is that a Humean account of causation cannot work in the case of self-causation: self-causation cannot be analyzed, Hume-wise, as a contingent relation of succession and contiguity such that every time event of type A occurs it is, contingently but invariably, followed by an event of type B. Since in self-causation cause and effect are the same “event”, no such relation of succession can be posited. Causality must in this instance either be reduced to nothing at all or it must be analyzed in terms of the action of a non-empirical “power” within the cause. Powers are (notoriously to logicians) “occult” things, or

152 aspects of things, empirically opaque, intensional, lurking within things rather than fully extensional, as materialist attributes are. (Routley 1980; Plumwood 1993) Let us add to intensionality the fact that the power of self-causation is reflexive, hinging on the ability of the thing in question to direct its powers onto itself. And to intensionality and reflexiveness let us furthermore add an element of telos – the fact that the power of selfcreation has a specific end in view, viz the self-existence of the selfcreating thing. The upshot of this thumbnail analysis of self-causation is that self-creativity looks decidedly conative, and conativity bears all the hallmarks of the intentional, where intentionality is understood, in phenomenological terms, as the province of the mental, or at any rate of the mind-like. In other words, a self-creating universe will be a conative universe. It will not be a purely materialist or mechanical universe. If conativity is understood as a category of panpsychism, then we can say that a panpsychist universe offers an answer to Leibniz’s question, why is there something rather than nothing, while a materialist universe does not. 4. Was there a beginning to time? If it is inherent in the nature of the universe – or, as we have seen, of a particular kind of universe, namely a conative one – to bring itself into existence, then this universe will presumably not come into existence at one moment rather than another. There must be a sense in which such a universe has always existed. Why so? The nature of things generally, in abstracto, can be specified outside of time, but in the case of a self-causing thing, its nature in abstracto is necessarily self-instantiating, so that the instantiation will exist as timelessly as the thing in abstract. (Heavy shades of the Ontological Argument here.) To say of the universe that it is of a finite age, as physicists currently do, does not then make sense for a selfcreating universe. There is no more reason for a self-creating universe to have come into existence X million than Y million years ago. This is not to say that it must have endured for an infinite number of years, or instants. It is rather to say that it must be, so to speak, temporally indeterminate at the edges. To explain what I mean by this, let us contrast such temporal indeterminacy with the current orthodoxy amongst physicists that postulates a “first three minutes” scenario.4 According to this scenario, the 4

This was the title of a particularly famous book by Stephen Weinberg outlining the early history of the universe. See Weinberg 1977.

153 universe began with an explosion of energy from an initial singularity and evolved rapidly thereafter, over tiny fractions of micro-seconds, till the highest of the initial high energy was dissipated, and things started to calm down, after an interval of three minutes or so. In other words, a fully temporally determinate universe burst into being at an originary moment and thereafter time immediately began ticking over, minute by minute. In supposing this, physicists are not of course supposing that the universe began at a point in pre-existent time, but that time itself originated at a certain point, a point that is now measured by how long ago it was in the past. Such a view however invites the question, why then? Why did the universe originate X billion years rather than Y billion years ago? No answer seems to be available for this question. Moreover this first-moment scenario seems implicitly inconsistent with physicists’ own account of the history of the universe. For the passage of time is discoverable through change, but metric time is a function of periodic processes: if there are no periodic processes that can function as measuring sticks for time, then time remains metrically indeterminate. In the early stages of the universe, according to the physicists’ story, there was an extremity of physical flux, a soup of protean form, as yet unfixed. In the midst of this explosive instability, no periodic processes were yet occurring. It is accordingly impossible to impose a measure of time, even retrospectively, on that period. Time was undoubtedly passing, since change was occurring, but whether the time that passed was of long or short duration would have been, and remains, impossible to say. There would have been no more reason to say of that earliest period that it lasted for three minutes than to say that it lasted for three billion years, since there was no way, even in principle, of measuring time. So although there may have been physical sequence in the unfolding of the universe, it seems wrong to say that this sequence began at a particular moment. There is no moment of origin because, during those earliest phases, there were simply no moments. The time between events was indeterminate. The most we can say is that the universe was around, in a temporally indeterminate way, and then as physical order constellated, the possibility of a temporal metric began to emerge.5 Moreover, if either a universal heat death or a “big crunch” ever 5

In order to ascertain that a given process is periodic, a significant degree of order must already obtain in the universe. Different kinds of ostensibly periodic processes need to be measured against one another, and a certain amount of physical theory capable of predicting periodicity in the processes in question must be at least tentatively formulable.

154 eventuate, then the universe might pass into a further phase of temporal indeterminacy.6 The “first three minutes” hypothesis then – which is to say, the hypothesis that posits an originary moment for the universe – seems inconsistent with physicists’ own account of evolutionary cosmology. The hypothesis of self-creation however - which, as we have seen, entails a panpsychist view of the universe – enables us to make sense of evolutionary cosmology without falling into the originary anomaly. The self-created universe emerges into metrical time from a pre-metric and hence temporally indeterminate past. In this sense such a universe is “outside” time, yet temporality can emerge within it. Perhaps this is not so different from the temporal experience of an ordinary self. There is a temporal indeterminacy about our own experience of origin. In our early life we lack any sense of the metric of time – our infancy is our “dreaming”, an ocean of temporal non-differentiation, from the depths of which a definite temporal axis eventually defines itself. There is no originary moment, only this ocean of flux out of which temporal order emerges. This is admittedly only an analogy, but perhaps it provides some intuitive support for the panpsychist hypothesis of a universe which is both timeless, because self-created, but also generative of time. In conclusion, the “hard problem” of consciousness is not the only problem that panpsychism helps to dissolve. Several of the primordial questions of metaphysics return satisfactory answers from a panpsychist perspective, not only vindicating panpsychism as an explanatory paradigm, but reinstating those metaphysical questions themselves, so long discarded as unanswerable. References Clifford, W.K. (1973): “On the Space Theory of Matter.” In: Proceedings of Cambridge Philosophical Society 2, 157. Reprinted in Kilmister, C.W. (ed), General Theory of Relativity, Pergamon Press: Oxford. Mathews, F. (1991): The Ecological Self. Routledge: London. Mathews, F. (2003): For Love of Matter: a Contemporary Panpsychism. SUNY Press: Albany. Plumwood, V. (1993): Feminism and the Mastery of Nature, Routledge: London and New York. 6

This paragraph parallels very closely a passage in Mathews 2003, p. 53.

155 Skrbina, D. (Ed) (2009): Mind That Abides: Panpsychism in the New Millennium. John Benjamins: Amsterdam and Philadelphia. Spinoza, B. (1951): Ethics. trans R.H.M. Elwes, Dover: New York. Routley, R. (1980): Exploring Meinong’s Jungle and Beyond. In: Departmental Monograph 3, Philosophy Department, Research School of Social Sciences, Australian National University, Canberra. Weinberg, S. (1977): The First Three Minutes. Andre Deutsch: London.

Not bound to feel everything? A dialog on the scope of experience. Matthias Rugel and Benjamin Andrae (Munich) In this article, we will discuss the question how many connections a thing in existence has to other things. Here, the term 'thing' is understood in a very wide sense to cover all kinds of concrete entities, be it material atoms or human souls. The discussion takes the form of a dialog between the two authors of this article, hopefully capturing the essence of the problem in a lively and entertaining way, but still providing solid theoretical arguments that are central for metaphysical investigations. As the argument progresses, 'experience' of another thing will take the center stage as a model for connection, and thus a strong panpsychist tendency is introduced. In the dialog, 'B' will argue that every individual is connected to all other individuals. He finds support for this in Aristotle, Leibniz and Whitehead. 'M' will try to argue that every individual is only connected to some and definitely not all other individuals. He backs up his thesis with Gregg Rosenberg's theory of natural individuals. M: Aristotle says in De anima (III.8.431b.22): “The soul is in a way all existing things”. What a fascinating statement. Nonetheless I cannot really believe it. Do you have an idea what it could mean? B: Hmm, Aristotle is notoriously brief in his statements, but I guess he means just that a mental entity is connected to every other entity. Otherwise it doesn't seem to make much sense (...unless he has a very strange concept of identity, but let us not go there). M: Alright. There are also nutritive and perceptual souls for Aristotle, but in this case he really thinks of a mental or intellectual soul. A mental entity may be related to every other entity. Yes, maybe that is what he means here. B: I think that from what we just said it follows that I am right now connected to the Big Bang (which happened some more than 13 billion years ago) and to the Alpha Centauri System (which is a little more than four light-years away), although I was not five minutes ago, simply

158 because I am now thinking about them. This basically means that I am connected to everything! (Or at least to everything I can think about. This statement opens the Pandora's box of some hardcore realist/anti-realist debates, but let’s not go there either.) I think that this conclusion might possibly upset quite a few of our fellow philosophers... M: Why? There is no problem. I do accept that a human can think about every entity. But that does not mean that I am connected to everything, because the relation to most entities is only a potentiality and a faculty. Actually one only ever thinks about some entities. B: But a potentiality in more than a mere connection on its own. For example, even German right now, the potentiality for me to part of reality. So via the potential to think entity is indeed connected to everything!

possibility. It constitutes a if I do not actually speak do so remains an important about everything, a mental

M: That's interesting. Perhaps a mental entity has a connection to all concrete things via its faculty to think. Maybe only the potentiality of thinking is necessarily a connection to all existing things. I'm not really sure about it: It might be that in quantum mechanics entities exist that are both particle and wave but we cannot conceive of them. But still, this may be as Aristotle says. However, I'm not really interested in the connection of thinking-about. It's a too thin connection; there is no guarantee that the thinking really grasps a real thing (and not some unicorn or some non-existent future). I agree with another opinion of Aristotle, that all thinking is derived from perceiving or simple experiencing. It seems to me that there are many existing things to which you are connected in experiencing. But Aristotle would be wrong when he would say that souls are experientially connected with all existing entities. B: By experience, do you mean sensory experience? M: As far as I can see Aristotle thinks of sensation. I take that to mean sensory experience. And that's a good starting point and the prototype of experience. (I would prefer this prototype to the prototype of the potentiality of thought. A thought seems to me constructed out of sense experience.)

159

B: Well, I doubt that sensory experience is a connection at all. For in sensory experience, I am just perceiving something, like light waves or sound waves, arriving at my sensory organs. It does not make any difference to me whether they are produced by some object I seem to perceive or as part of some illusion – so sensory experience can hardly constitute a connection to some object. This is also what Leibniz would have said, I think. He thought that we are indeed connected to everything, but only in an ideal way and not via sensory perception. “This interconnection, or this adapting of all created things to each one, and of each one to all the others.” (Monadology, §56) “But in simple substances this influence of one monad over another is only ideal, and it can have its effect only through the intervention of God; in the sense that in God's ideas one monad requires of God, and with reason, that he takes account for it when he organizes the others at the beginning of things.” (Mondaology, §51)

M: I like Leibniz and especially his Monadology, thanks for focusing on it. Leibniz really thinks that not only humans as mental entities, but also all monads are connected to all other monads in an ideal way. That includes animal souls and simple monads. At least some of them do not have sense organs. I agree with Leibniz that experience is not always sense experience. But it is also implied that the ideal connection of monads is not only thinking; Leibniz calls it perceiving. It will be non-sensory perceiving in some cases, but why is it forbidden to think that it is also sensory experience in some cases? B: It is true that Leibniz talks about perceptions. But I am fairly sure that he does not take that to mean something perceived with the sensory organs, because that would imply a direct physical interaction between monads and he explicitly denies that. Therefore I consider it certain that also for Leibniz, a connection via a sensory organ is not a real connection at all. "No external causes could ever have an influence on its [the monads] interior" ” (Mondaology, §11). M: Assuming that you are right with your rather wild statement about the status of sensory experience, the only real connection left is non-sensory

160 experience. Whitehead knows this and speaks of 'perception in the mode of causal efficacy'. An analogy for non-sensory feelings is the experience of one's own eyes while seeing and one's own skin while touching. But Leibniz denies that non-sensory perception is something causal. How can you understand it, if it is neither causal nor sensory? B: Well, I think that for Leibniz the perceptions still represent the whole monadic world correctly, because god made sure of that at the beginning of things, thus providing some kind of ideal, god-given connection. However, I agree that Leibniz' conception leaves very little space for an autonomous connections between a subject and the world, and that having a theory that includes such autonomous connections would be desirable. You are right that the best candidate for such a direct, non-sensory connection is causation. However, I think that talking about causation will immediately allow me to drive my point home that everything is connected to everything, because this is just how the interaction laws of physics behave. M: I know that the field of the some fundamental forces of physics, e.g. gravitation, is considered to be holistic. I also interpret physics as causal and not just as regular. Perhaps it is also the right position that everything is physically responsible for everything. But there is more to causation than the fundamental forces of physics. To get a grip on causality it is not useful to focus on the holistic causal responsibility but on the significance that one particle or field has to another. This happens all the time in everyday physics, where you disregard the wider environment. I would propose to give an ontological foundation to that with a theory of causal significance. Gregg Rosenberg did this in his 'A Place for Consciousness' (2004). A further advantage of this theory is that it is possible to interpret quantum entanglement as a causal phenomenon. B: Hm, in physics this disregarding of the wider environment you are talking about is just an approximation made for practical purposes, and there is no theoretical ground for it. But still, I find it very interesting that you attempt to provide such a ground. However, I do not understand what causation means beyond interaction by physical laws.

161 M: Well, I do not propose a physical theory but a metaphysical one. But it should be compatible with physics and in the best case it should shed light on the physical concepts. My theory of causation tries to clarify the concept of a thing as something that is causally relevant. It also clarifies the concept of entanglement, which I consider to be a connection with a restricted scope. B: Well, compatibility is not a problem, because no physicist will ever include the distribution of dust in some remote corner of the planet mars into his calculations. However, I strongly doubt that there can be positive influence of our metaphysics on physics, unless the theories are experimentally testable. Still, I also doubt that terms from physics are relevant for metaphysics, so I am very interested to hear your theory of causation that features a restricted scope of causal influence. M: My idea is somewhat Whiteheadian. For him there is a strong association between causation and experience. And at least in most cases he thinks of non-sensory experience. He says “A simple physical feeling is an act of causation” (Whitehead 1929, 361). This is my new meaning of causality. Causation and experience run in opposite directions. A Whiteheadian entity, a so called 'actual entity', is caused by past events. And its past is just what it experiences. B: This definition of causation is very interesting. It is also panpsychist, because every interaction will be accompanied by some experience. I am very sympathetic to that. However, if I remember Whitehead correctly, each actual entity is connected to all past events, and since the future does not yet exist, that means that for Whitehead everything is connected to everything, and my thesis still stands. M: That's right. An actual entity experiences everything that places a causal constraint on it and for Whitehead this is its whole past. There are only secondary aspects that filter the experienced entities. The scope can be restricted by the subjective aim or by conceptual prehensions. But these are operations the actual entities undertake additionally to the prehending of all past facts. So you are still right.

162 But now, I would like to modify the Whiteheadian theory to include a more fundamental restriction of connection. In this modified theory, the connection is still based on causality. B: Ok, since all cases we looked at up to now either resulted in everything being connected with everything – as in the case of thinking about or causal interaction in physics and in Whitehead's understanding – or in everything being connected with nothing – as for sensory experience – now you propose a new theory. I am very keen to hear it and will lean back and enjoy the show for a while... M: Let's again begin with Aristotle and his statement: “The soul is in a way all existing things”. I’d like to translate this statement in a more modern language. I'm not really interested in souls, especially not in souls separable from the body, but in natural individuals which I conceive in a panexperiential way as entities with both a physical and an experiential aspect. I'd like to reject “all existing things” in Aristotle's thesis and restrict my claim to “some existing things”. But I'd like to insist that there is a strong connection between the basic entity or the natural individual and the other existing things, in the best way there may be a kind of identity as Aristotle says. The connection should not only mean that two things are in the same world or that a mental thing can think about all other entities. The intensive connection I am thinking of has to be a kind of basic causal, bodily, experiencing or inner connection, perhaps akin to identity. So my thesis is “Each natural individual is connected in an intensive way with some but not all existing things.” My question is now: What makes the scope of an entity (especially a panpsychistic entity) restricted? There are two types of restriction. Half-hearted restriction means that primarily an entity is connected to everything but in a second step this holistic connection is restricted. A fundamental restriction does not share the holistic claim. We talked about three variants of the half-hearted restriction. Leibniz' holistic approach has the problem to individuate the different monads. Therefore, he ascribes to monads an individual point of view and a more or less distinct perception of other monads. A less distinct perception may be less relevant for the activity of a monad. Physics works in a pragmatic similar way. The fundamental forces may interrelate all entities, but only some entities are relevant to others, e.g. because the rest are so far away that their impact can be neglected. The third variant discussed is Whitehead's. He restricts, as we

163 have seen, via conceptual work of the experiencing subject which primarily prehends all the past. In the following I want to discuss some variants of fundamental restriction. My intuition (why I want to restrict this scope) is that each kind of sense experience is restricted to an environment and to a special domain of qualities. Such a restriction should also apply to wider concepts of experience. In particular it should apply to experience as an aspect of nonHumean causation. One way to conceive such a restriction of my connections would be to allow only connections to my parts and my inner structure. In another variant the restriction is performed by Aristotelian forms that may help in our problem in at least two ways: the restriction may result from the associated contents or from the powers which a substance is attributed by its essence. My favorite will be the restriction by a fundamental hierarchical causal mesh as it is provided in Gregg Rosenberg's theory of Causal Significance and his Carrier Theory of Causation. This will be read as a variant of a restriction via powers. First, let us try to restrict the scope of a subject to its inside or its bodily parts. With this supposition some work may be done on individuation: an individual may be correlated with something that makes a causal difference as a kind of downward causation. But this is no contribution to the question whether the scope of an individual is restricted or not, because nobody will deny that there are also causal connections of an individual both with its bodily parts and its environment. It may be proposed that one cannot experience every causal impact but only the impact of one's own inside. Only the inner (downward) causality may correspond to the scope of experience. This approach has at least two difficulties: First, what should be the difference between real inner experience and outer experience about which one is deceived? There is causal involvement with the inside and the outside; one seems to experience the inner and the outer world. Even if the distinction can be made a second difficulty with the theory of inner- or bodily-scope appears. If one can only experience one's inside, how is it possible to have experiential contact with the world outside at all? If an individual is in a deep sense only connected to its inside, we have a kind of solipsism. If one favors a causal theory of experience, our problem shows up similarly to a difficulty of Leibniz. Why should it be impossible that a monad in its experience causally interacts with other monads? These two difficulties are why I reject the thesis that the scope of a subject (individual) is restricted to ones inside or one's body.

164 All other proposals for restrictions, as far as I see, are variants of Aristotelian forms. In perception and thought the form plays the role of a transmitter. When I see a tree, there is the same form in me as it is really implemented in the tree. Perhaps because of this Aristotle not only says that the soul is connected with other things, but that it: “is (in a way) all existing things”. The soul being all things may be only a thin possibility, the restriction may be that only the environment and some types of the form may be associated with this special soul. Even if a thinking mental entity may be related to everything, the experiences transmitted to an animal may be very restricted. But the concept of forms as content has lethal defects, as we can see by looking at Leibniz. He cannot accept forms as a content of thinking and experiencing. He attacks the idea of causal influx, the idea that the same form that is a principle of a concrete thing can also be accidentally in the mind of an experiencing or thinking subject (Mondaology, §7). It's really not a good idea, that there is something that could be taken from a thing in the world and put into the soul for experiencing or thinking it. Leibniz denies that there has to be something in between the perceiving subject and the perceived thing. Instead there has to be direct representation (even if ultimately guaranteed by God). If Leibniz' argumentation is sound, the above proposal to restrict the scope via the forms as contents has to be rejected. There is a second understanding of Aristotelian forms that does not imply considering forms as transmitters of content. This substantial form is thought of as the essence of an individual. Some causal powers of the individual belong to this essence. Causal powers may correspond to restricted relations and only be directed to a restricted environment. For example, consider that the essence of a bird is to be a flying vertebrate animal, and that flying is a causal power 'flying to (somewhere)'. Let us concentrate on this specific power. If we connect this with the Whiteheadian idea that causation is something experiential this may be an attractive panpsychistic restriction. Besides some other experiential powers shared with other vertebrate animals, birds are special in that they can feel on their way on flying to some place, by their own power. Some other powers are missing, like the powers of rational thinking and acting that we ascribe to humans. It may be that this restriction of the mode of experience implies a restriction of its scope, but that is not necessary (as in the case of of thinking). We have said that Whitehead himself does not share the idea of a primarily restricted connection. He favors secondary restrictions

165 performed by conceptual prehensions. One can understand them as connected to a form – a Whiteheadian “eternal object”. (Leibniz also would see these “eternal objects” as a kind of “form as content” that has to be rejected. But for now we ignore this objection.) This eternal object, say the 'flying to (somewhere)', that is felt by a bird, may be classified as something that is potentially directed to everything. But the specific environment prevents this feeling in special cases, e.g. no underwater object and no treasure buried in the earth can be the aim of a birds feeling of flying to somewhere. But why not claim that flying is primarily restricted to open oxygenic space or something like this. You may claim that humans with their power to think and act rationally are not primarily restricted in scope. But it seems to me that it is not necessary for a strong connection of an individual to have a conceptual or cognitivemental aspect like rationality. There is a further reading of forms as causal powers besides instances where one feels one’s own powers. The idea is that it is not the individual exercising the causal power any more that has the experience of it, but that a different individual has that experience. The form does not have to be identified with the causal power; it may provide only the connection. Although Gregg Rosenberg does not understand himself in these Aristotelian terms, I'll try to interpret his position in such a way here. For Rosenberg the form is the causal connection between some natural individuals. Instead of a “form” he speaks of a receptive binding. This receptive binding is not itself experienced. But it implies that the bound or “formed” individuals experience each other. A causal connection between individual C and B is only possible if there is a receptive binding (or form) that binds C, B and possibly many other individuals in a causal nexus. Corresponding to each Aristotelian form there has to be some “matter” that it forms. Corresponding to each binding there has to be something intrinsic that is bound, intrinsic properties of the bound parts. In Rosenberg's theory this insight is implemented in the following way:

166

Individual A

Shared form (causal receptive binding) of A, B and C

Individual C With effective properties e, f, ...

Individual B With phenomenal properties p, q, ...

e contributes causally to B

Three individuals A, B, C bound in a shared receptive binding. This implies causal constraints between A, B and C, resp. the emphasized asymmetric relation that C causes B. The shared form implies this causal relation and implies as its 'matter' (intrinsic nature) that B experiences C. There have to be other relations of causal contribution and experiencing inside this receptive form that are not pictured here to integrate individual A in the shared form. Let there be a receptive binding implying that individual C constrains individual B. This may happen because of the constraining effective property e of C. Now, this connection cannot exist if it does not have an intrinsic carrier, i.e. a property that not only mirrors the relations in which B and C stand, but also gives them an ontological foundation (without which it cannot exist as a concrete entity). According to Rosenberg's proposal, this carrier is the phenomenal property p of B. (The main alternative would be that it is a phenomenal property of the whole causal nexus with A, B, C, …, which is bound by the form connecting them.) It is a plausible interpretation of Rosenberg that this phenomenal carrier p of B is identical to the effective property e of C. (It may be really attributed to B, but derivatively, seen purely structurally without regard to its phenomenal aspects, as a property of C.) In this case we are back to the starting thesis of Aristotle: The soul is not only connected to some things, but it is some other things because it carries one of its essential properties. Property e could not exist without being carried by p of B. There is a form as the connection itself, the receptive binding. But this form is not the form criticized by Leibniz. Rosenberg's form does not have to be the transmitter of the content. Receptive binding only provides the context for causal interaction and direct experiencing of the bound individuals. What are the implications of these considerations to the project of panpsychism? The point I'd like to make is in the title of this dialog. A

167 subject in a panexperientialist world is not bound to feel everything. There may be not only half-hearted restrictions of the scope of experience but also fundamental ones. B: Alright, that was very impressive. You convinced me that it is possible to devise a metaphysics that features a restricted scope of connectivity. There is one huge problem, though: You never gave a sufficient reason as to where exactly to draw the boundaries. Why include a connection to this entity and not to the one next to it? As long as this problem is not solved, I think I will rather stick to my idea of universal connectivity. M: You are right, that's a real problem. To get a hint of a solution look at a further feature of Rosenberg's theory, the hierarchy of natural individuals. A receptive binding does not only provide a form for the causal and experiential connections of the individuals that are its members. It is also the specific characteristic of a natural individual of a higher level. This higher level individual will also interact with other individuals of the same higher level and experience the impacts it gets from them. This is provided by a further form or receptive binding that does not work on the individuals of the lower level but on the individuals of the higher level. Corresponding to this receptive binding there is a further individual on the next higher level, etc. Shared form of a higher level A

C

E

B D

The picture shows receptivity of different levels. The receptivity between A, B and C is the same as in the last picture. Each closed line corresponds to a receptivity, connecting its member individuals and similarly as an individual itself, bound by a higher level form. The lowest level receptivity corresponds to the continuous line, the level above binding A, B and C on the left side and D and E on the right side corresponds to the dashed line and the highest receptivity to the outer dashed and dotted line.

168 This hierarchy is possible because the lower level individuals are not determined independently of their context. The end of this hierarchy is reached if the highest level individual is totally determined. This causal and experiential mesh corresponds to the hierarchy of the sciences. It is a scientific fact that on all levels other causal powers and other environments are relevant. I presuppose the hierarchical causal mesh is the fundamental order of the world but cannot really specify it more as through the hint to the hierarchy of sciences at the moment. The opposite thesis to this experiential mesh, a panpsychism with holistic experience, always seemed, at least to me, to be a strange position. The probability of panpsychism is much better if there is at least a model for restricting the scope of experience. This corresponds to the intuition of a restricted scope of sense experience in everyday life. L (from the audience): Matthias, your restriction seems to be an extrapolation from phenomenology to ontology. Is this restriction only according to phenomenology? It seems to me somewhat anthropocentric. M: My idea to restrict the scope of experience comes from phenomenology, that's right. As far as I see, you have to choose what type of connection you prefer while building an ontology. I prefer a connection modeled after sense experience to one modeled after thinking. No species of subjects seems to have all kinds of sense experience. Every sensory feeling seems to be restricted. This proves to be much less anthropocentric than when you model your theory after human thinking. With thinking it is much simpler to imagine an analogy to a thought-relation with all other objects. The holistic picture of science seems to me only a further example of a theory modeled on thinking rather than on experiencing. But with the much more intensive relation of experiencing something I'm able to adopt some fundamental restrictions of the scope of connectivity. Epiloge: More on Rosenberg's terms The dialogue forum is closed, some people who are acquainted with Rosenberg's book, are still talking. G: Why don't you understand Rosenberg's theory holistically? Is causality no transitive connection? Does not the causal mesh guarantee that every natural individual is causally connected to all other individuals? Does not the carrier theory imply that this connection has also its phenomenal inside?

169

M: I can admit that there are some holistic relations of all natural individuals, e.g. in the potentiality of thought. But this normally does not concern causality and experience. There is a special Rosenbergian case that is a nearly holistic story. There may be only one level of individuals that are connected together with only one form or receptive binding determining them all. But even this nearly holistic scenario implies no holistic experience of any of its individuals. The high-level-individual that connects all the others and corresponds to the world as a whole is not experienced. This happens simply because individuals in the Rosenbergian theory can only experience individuals of the same level connected by a higher-level-individual. (An indicator for this is that in the human case it is a phenomenological impossibility to experience a whole of which oneself is part in a way that is both an inner and an outer experience. To experience the whole means either an outer experience of all the other entities or an inner experience of unity. It is doubtful whether the whole nature is an entity at all.) I would also think that the world is not simple like this. If there is a more complex structure of the Rosenbergian causal mesh, not every individual has causal contact to all individuals of the same level in the sense of suffering effective causal constraint. Additionally, no natural individual has experiential access to individuals of other levels. Hence, effective causal constraint is not considered to be a transitive relation. The receptive binding itself could also be described as exhibiting some causal constraint on its members. But in Rosenberg's theory this constraining effect has not to be carried by a phenomenal property of the receptive binding relation. Its effective aspects are totally carried by effective properties of the individuals of the same level. In Rosenberg's model no individual can effectively constrain an individual of another level. If one were to assume such a constraint, one would end up with an interactionalism of levels. One had to accept some similar problems as the substance dualist has when he accepts an interaction of body and mind. Therefore, according to Rosenberg, no individual can experience something on a level different from itself. G: You've argued that the restriction to one's own body would end in solipsism. But with Rosenberg's theory, don't you also have a kind of solipsistic concept of phenomenal properties? Don't you make it impossible in principle that an individual has a grasp on the whole

170 universe? But at least in cognition it seems that humans do have access to some holistic aspects of the universe, the whole physical space, a whole qualitative space, etc. M: It may be that in cognition there is a connection to everything, but I doubted the stability of cognitive connections before. For Whitehead, e.g., our cognitive access to the world as a whole is mediated by abstraction and eternal objects. Rosenberg allows a direct experience of some things outside oneself. But he denies the possibility of an experience of all other things, especially the world as a whole. He could construct the connection to the world as a whole with similar methods as Whitehead. As long as there are many layers of nature, I cannot allow that an individual may grasp all individuals of the universe. At most it experiences all other individuals of its own level. But maybe the experiencing individuals of the highest level minus 1 could be in some receptive bindings that at least include every concrete being of the highest level minus 1 so that it could make sense to say that it experiences a totality of the world. But even if an experience of the whole would be possible for the nearly highest level individuals, it would not be possible for simpler individuals of lower levels. Rosenberg himself seems to speculate that there aren't any receptive bindings that bind a human individual to all other individuals of the same level. He could imagine that the highest level receptive connection only binds some parts of the brain, one of which is me (at the relevant moment). But he is far from being a solipsist who restricts experience to the space filled by a physical brain. There may be overlapping receptive fields which bring the relevant information from the outside to the experiencing individual. Some primitive isomorphisms that mirror the outer world to the phenomenal world and vice versa (Rosenberg 2004, 292) could transport the information. G: It's an interesting question how some information comes into an experience from outside a receptive binding. But it’s of no interest for our search for the objects of direct experience, since this information is no direct experience. It is also a correct reading of Rosenberg that he restricts effective causation on intra-level causation. But that's not the whole truth. Both lower and higher level individuals contribute to the causal profile of a natural individual. The lower level effective interaction does not stop in case of intra-level causation. Especially, through the causal constraints of

171 the four fundamental forces, that may be represented by the effective properties of the lowest level (or at least a low level), every higher level individual is effectively causally connected to all other individuals. M: I've admitted something like a holism of physical causality before. I have to go more into the details now. Rosenberg distinguishes between the Humean mosaic, the nomic mosaic and the causal mesh (Rosenberg 2004, 219-220). The Humean mosaic consisting of the regularities of nature is the most wide-meshed, it is contained in the nomic mosaic (containing also some uninstantiated laws), which is again integrated in the causal mesh containing receptive bindings. Therefore the connections of the wider meshed structures are also causal, but perhaps less fundamental in the causal mesh. Physics is wholly described by the Humean mosaic or in some readings of natural laws by the nomic mosaic. But a causal structure has to be constrained by a receptive binding that is not necessary implemented by the physical regularities or laws alone. Causal laws and regularities are ontologically abstractions and simplifications of receptive bindings, even if epistemologically the regularities and laws are easier to detect. Forms or receptive bindings are ontologically the much more fundamental causal structure. And it is enough for the carrier theory that the receptive bindings and the effective properties of the individuals that are directly bound by it are carried. The effective properties are carried by phenomenal properties. The receptive binding, although it was not necessary to mention up to now, is carried by the experience itself. This can be identified with the subject that experiences phenomenal properties. G: This may be as you said – although it seems to me quite misleading to call the physical interactions causal connections even if they are no receptive bindings. But here comes my second doubt on Rosenberg's theory coming from the level above. The receptive binding relations seem to be crucial to the nature of the effective properties of the bound individuals. Rosenberg calls the binding relations final causes of the bound individuals, we called them forms or (with Aristotle) “formal causes”. You have just said that receptive bindings are also carried by experiencings (of unitary fields) or subjects. It seems to me that Rosenberg's and your position is like Leibniz' position, there's a primitive coordination of individuals. But while in Leibniz' theory this coordination is a-causal, you call it causal. And, most important, while in Leibniz' theory this coordination is guaranteed by one divine subject,

172 your theory guarantees it through many subjects: all higher level individuals. One could say that Rosenberg's theory is the polytheistic version of Leibniz' monotheistic theory. Is the higher level really a subject like this, which cannot be accessed by the lower level but gives the restriction of the connections for the lower level by its subjectivity? M: That's an interesting idea. There may be some respects in which your suspicion is true. God may constrain every individual but may not be experienced herself. But remember the differences. The God of Leibniz thinks of all monads and the whole world, a carried individual subject in Rosenberg's theory may not think anything but only experience. Additionally, in the spirit of Leibniz it is plausible that God is experiencing, at least in so far as she is thinking. But a determinate individual – resp. one of the highest level – is not constrained by any other individual and therefore does not experience anything. As far as I see there is no necessity that the receptive binding itself has to be carried, resp. if its bound individuals are carried already. Its “matter” is not a corresponding subject, but the carrier of the individuals and properties that it binds. Most importantly, as long as you accept that there has to be a receptive binding (or a subject) in order to understand an interaction of the parts, the parts in Rosenberg's theory really affect and modify each other, while a monad only modifies itself.1 References Aristotle (350 B.C.): De Anima, in The works of Aristotle, translated by J.A. Smith. Oxford: Claredon Press 1931. Leibniz, G. W. (1714): Monadology, in G. W. Leibniz, Philosophical Texts, translated and edited by R.S. Woolhouse and R. Francks. New York: Oxford University Press 1998. Rosenberg, G. (2004): A Place for Consciousness: Probing the Deep Structure of the Natural World, Oxford University Press. Whitehead, A.N. (1929): Process and Reality: An Essay in Cosmology, ed. of D. R. Griffin & D. W. Sherburne, 1978, New York: Free Press.

1

Thanks to Christina Schneider and the audience of the Vienna Conference “The Mental as Fundamental” for valuable comments on a draft of this paper, especially Ludwig Jaskolla and Godehard Brüntrup.