An Emergentist Argument for the Impossibility of Zombie Duplicates Reinaldo Bernal Velasquez
To cite this version:
Reinaldo Bernal Velasquez. An Emergentist Argument for the Impossibility of Zombie Duplicates. 2016. halshs-01425545
HAL Id: halshs-01425545 https://halshs.archives-ouvertes.fr/halshs-01425545 Preprint submitted on 3 Jan 2017
HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Bourses Fernand Braudel IFER
An Emergentist Argument for the Impossibility of Zombie Duplicates
Reinaldo José Bernal Velasquez N°120 | décembre 2016
Some influential arguments in the metaphysics of consciousness, in particular Chalmers’ Zombie Argu- ment, suppose that all the physical properties of com- posed physical systems are metaphysically necessitated by their fundamental constituents. In this paper I argue against this thesis in order to debate Chalmers’ argu- ment. By discussing, in non-technical terms, an EPR system I try to show that there are good reasons to hold that some composed physical systems have properties which are nomologically necessitated by their funda- mental constituents, i.e., which emerge in the sense of the socalled ‘nomological supervenience’ views.
Working Papers Series
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 2/21
An Emergentist Argument for the Impossibility of Zombie Duplicates
Reinaldo José Bernal Velasquez
Décembre 2016
The author Reinaldo José Bernal Velasquez hold two 5 years bachelors, in Physics and Philosophy, from the Uni- versidad de Los Andes (2000). His final essay discussed the problem of the completeness of quantum mechanics. It was supervised by Dr. Jaime Ramos and Dr. Ramon Fayad. He did a 3 years Master in Philosophy at the Universidad Nacional de Colombia (2005). His graduate work addressed the problem of consciousness and was done under the supervision of Dr. Jaime Ramos. Afterwards, he did a MSc in Philosophy and History of Science at the LSE (2007). His dissertation concerned the realism-antire- alism debate. Recently, he did a PhD thesis under the direction of Max Kistler at the IHPST (CNRS/ ENS/Paris 1). He graduated in December 2011. The jury was composed by Ned Block, Jérôme Dokic, Pierre Jacob, Max Kistler, David Papineau and Jaime Ramos. A revised version of this thesis was pub- lished by Ontos Verlag (2012)/De Gruyter (2013). The text
During 2015 Reinaldo José Bernal Velasquez was a postdoctoral researcher at the Institut Jean-Nicod, thanks to a Fernand Braudel IFER scholarship. He worked within the research group “conscience et ipséité”, directed by Dr. Uriah Kriegel. His research interests mainly involve the Philosophy of Mind and the Philosophy of Science; in particular, the metaphysical questions within these fields. Citing this document
Reinaldo José Bernal Velasquez, An Emergentist Argument for the Impossibility of Zombie Duplicates, FMSH-WP-2016-120, décembre 2016.
Les Working Papers et les Position Papers de The Working Papers and Position Papers of la Fondation Maison des sciences de l’homme the FMSH are produced in the course of ont pour objectif la diffusion ouverte des tra- the scientific activities of the FMSH: the © Fondation Maison des sciences de l’homme - 2015 vaux en train de se faire dans le cadre des chairs of the Institute for Global Studies, diverses activités scientifiques de la Fonda- Fernand Braudel-IFER grants, the Founda- Informations et soumission des textes : tion : Le Collège d’études mondiales, Bourses tion’s scientific programmes, or the scholars [email protected] Fernand Braudel-IFER, Programmes scien- hosted at the Maison Suger or as associate tifiques, hébergement à la Maison Suger, research directors. Working Papers may also Fondation Maison des sciences de l’homme Séminaires et Centres associés, Directeurs be produced in partnership with affiliated 190-196 avenue de France d’études associés... institutions. 75013 Paris - France Les opinions exprimées dans cet article n’en- The views expressed in this paper are the http://www.fmsh.fr gagent que leur auteur et ne reflètent pas author’s own and do not necessarily reflect http://halshs.archives-ouvertes.fr/FMSH-WP nécessairement les positions institutionnelles institutional positions from the Foundation http://wpfmsh.hypotheses.org de la Fondation MSH. MSH.
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 3/21
Abstract
Some influential arguments in the metaphysics of consciousness, in particular Chalmers’ Zombie Argument, suppose that all the physical properties of composed physical systems are metaphysically necessitated by their fundamental constituents. In this paper I argue against this thesis in order to debate Chalmers’ argument. By discussing, in non-technical terms, an EPR system I try to show that there are good reasons to hold that some composed physical systems have properties which are nomologically necessitated by their fundamental constituents, i.e., which emerge in the sense of the socalled ‘nomological supervenience’ views. Keywords Philosophy of Mind, Zombie Argument, Phenomenal Consciousness, Physicalism, Emergence, Bell’s Theorems
Un argument émergentiste pour l’impossibilité des zombies
Résumé
Des arguments parmi les plus influents dans la métaphysique de la conscience, en particulier « l’argument des zombies » de Chalmers, supposent que toutes les propriétés physiques des systèmes physiques composés résultent avec une nécessité métaphysique de leurs constituants fondamentaux. Dans cet article je remets en question cette idée et l’argument de Chalmers. A travers la discussion, en termes non techniques, d’un système EPR, j’essaye de montrer qu’il y a de bonnes raisons pour soutenir que certains systèmes physiques composés ont des propriétés qui résultent avec nécessité nomologique de leurs constituants fondamentaux, c’est-à-dire, qui émergent dans le sens des théories de « survenance nomologique ». Mots-clefs
Philosophie de l’esprit, argument des zombies, conscience phénoménale, physicalisme, émergence, théorème de Bell
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 4/21
Sommaire
Introduction 5 Physicalism, supervenience and related notions 6 Nonlocality and NS-Emergence 9
The EPR Argument 9
Bell’s EPR experiment 11
Bell’s Theorem 11 Microphysicalism and the EPR experiment 12 Back to Chalmers’ Zombie Argument 14 Conclusion 17 Acknowledgement 17 References 17
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 5/21
Introduction And, halmers (Chalmers D., 1996)advanced (premise 2) “[H]igh-level facts and laws are a still influential argument, labelled entailed [with logical necessity] by all the ‘The Zombie Argument’, to state that microphysical facts (perhaps along with consciousness is not a physical prop- microphysical laws)” (p. 71). erty1. He claimed that a world identical to the Then, actual one in all its physical aspects, but where C (conclusion 1) A world W* microphysically there is no phenomenal consciousness (Nagel T., 1974; Block N., 1998), is conceivable. Thereby he identical to the actual world W “will have the concluded that, in the actual world, consciousness same macroscopic structure as ours, and the is not a physical property. He wrote: same macroscopic dynamics” (p. 73). 1. In our world, there are conscious experiences. And yet, 2. There is a logically possible world physic- (conclusion 2) There would be no conscious ally identical to ours, in which the positive experiences in W*. facts about consciousness in our world do not 2 The purpose of this paper is to contest Chalmers’ hold 2. Argument. To do so, I shall not question the main 3. Therefore, facts about consciousness are fur- intuition behind it: that facts about consciousness ther facts about our world, over and above the cannot result (panpsychism aside) from micro- physical facts. physical facts in virtue of a metaphysical neces- sitation relation5. Instead, I will argue against 4. So materialism is false (Chalmers D., 1996: the acceptance of premise 2—a central tenet of 123)3. ‘microphysicalism’—by favouring a ‘nomological The question is if ‘there is’ a metaphysically pos- supervenience’ variety of physicalist emergentism sible world W* which is an exact physical replica (NS-emergentism)6. The argument I will advance of the actual world W and yet lacks conscious- is grounded on Bell’s (Bell J., 1964) Theorem, ness. For the physicalist, there is no such world. which concerns the correlations in EPR experi- He takes consciousness to be a physical property ments. I should emphasise that this argument will and, therefore, he would claim that an exact rep- be presented without technical terms or formulas; lica of W in every physical aspect would neces- no training in quantum mechanics is required to fol- sarily include consciousness. However, this claim low it. alone would beg the question as Chalmers pre- To be sure, in the philosophy of science arguments sented it. Chalmers argued for the possibility of for emergentism based on quantum entangle- the zombie world’ W* with a line of reasoning ment and critiques to microphysicalism have a that is based on a ‘global supervenience’ claim: 7 relatively long history . But note, first, that in the If, philosophy of mind microphysicalism remains as (premise 1) Facts about consciousness are not a prevalent view among physicalists. In Papineau’s microphysical facts4. words, “many […] philosophers seem to think of physicalism as some kind of commitment to the 1. The Zombie Argument is a version of ‘the Conceivability primacy of the microscopic. In their view, physic- Argument’ advanced by (Kripke S., 1980) and others. See alism doesn’t just say that everything is physical. (Chalmers D., 2010). 2. Chalmers strongly relates logical modality with meta- physical modality: “[...] the metaphysically possible worlds mers D., 2002) and (Chalmers D., 2010). are just the logically possible worlds (and [...] metaphysical 5. This intuition frequently underlies as well arguments for possibility of statements is logical possibility with an a pos- panpsychism. See, e.g., (Strawson G. et al., 2006). teriori twist)” (Chalmers D., 1996: 38). 6. Emergentism is not always conceived as a form of phy- 3. In (Chalmers D., 2002; Chalmers D., 2010) Chalmers sicalism. Indeed, (Crane T., 2001) claims that it is incom- presents the argument with an alternative form that stress- patible with physicalism, and some authors, e.g. (Nida-Rü- es the entailment from conceivability to possibility: “If it is melin M., 2006), propose emergentist views as forms of conceivable that there are zombies, it is metaphysically pos- dualism. sible that there are zombies” (Chalmers D., 2010: 107), and 7. For an extensive discussion of microphysicalism see argues for the validity of this entailment. (Hüttemann A., 2004). Concerning arguments for the exis- 4. This premise amounts to the exclusion of panpsychism, tence of emergent properties based on quantum entangle- which Chalmers discusses in (Chalmers D., 1996), (Chal- ments, (Humphreys N., 2008) is among the most influential.
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 6/21
It also says that everything is microscopically differing in their B-properties”10(Chalmers D., determined” (Papineau D., 2008: 126). Secondly, 1996: 33). A-properties constitute the ‘superven- note that the majority of arguments for emergent ience base’ of B-properties. He introduces this properties in physics concern weak emergence or concept in order to elaborate the aforementioned epistemological emergence, and are rarely applied premise 2 and conclusion 1. This definition con- in a straightforward manner in the metaphysics cerns properties, but analogous definitions can of consciousness8. This paper is an attempt to con- be stated to relate other kinds of items like enti- tribute to the debate of The Zombie Argument ties, structures, facts, events, and laws. I will use by showing, with a relatively simple case, why ‘supervenience’ in this more general sense. and how NS-emergentism threatens its validity. Supervenience is a reflective, transitive and I shall not advance an original interpretation of non-symmetric relation. Now, let us distinguish, Bell’s Theorem or a new view on emergentism. as is usually done, between ‘logical supervenience’ Surely, the interpretation of quantum mechanics (L-supervenience) and ‘nomological superven- and, in particular, the interpretation and signifi- ience’ (N-Supervenience) depending on whether cance of Bell’s Theorem, are open questions that the modal operator in ‘possible situations’ quan- are still highly debated. However, it is fair to claim tifies over logically or nomologically possible that there is wide agreement on the idea that will worlds respectively. N-supervenience must be be central: Bell’s Theorem plus the empirical distinguished from nomological supervenience adequacy of quantum mechanics provide strong emergence (NS-emergence). As we will see, in reasons for believing in the existence of nonlocal NS-emergence there must be a fundamental law effects in nature9. of nature linking the emergence base with the emergent property. By contrast, N-Superven- In the following sections, first, I will introduce ience admits a metaphysical necessitation relation the central notions that I will be using. Second, as the link between the supervenience base and I will describe a two-particle EPR experiment the supervenient property11. and introduce Bell’s Theorem. Third, I will pro- pose an emergentist interpretation of nonlocal Physicalism correlations. Forth, I shall elaborate on the con- sequences of the previous interpretation of Chal- Imagine a world W* defined as an exact physical mers’ ZombieArgument and address five plaus- replica of the actual world W: Every physical entity of W, with its properties and history, has ible objections. 12 a qualitatively identical counterpart in W*12 . If Physicalism, physicalism is true, then for W* to be a physical replica of W is for it to be a replica tout court: supervenience and related Everything that exists or obtains in W has an notions identical counterpart in W*. All the organisms, persons, institutions, facts, events, etc., of W, are In this section I will advance the definitions of replicated in W*. On the contrary, if physicalism supervenience’, ‘physicalism’, ‘microphysicalism’ is false, then W and W* differ: some entities of and ‘NS-emergentism’ I shall adopt. W lack their counterpart in W*. Accordingly, I Supervenience take physicalism to be the view that all the enti- Chalmers defines ‘supervenience’ in a canon- ties inhabiting the actual world, their properties, ical way, as follows: “B-properties supervene on and all the facts and events involving them, have A-properties if no two possible situations are a physical nature in the following sense: identical with respect to their A-properties while (Physical Nature) An item has a physical nature if it L-supervenes on a set of physical items. 8. For the distinctions weak/strong emergence and ontolo- Here, ‘physical items’ includes physical enti- gical/epistemological emergence, see (Chalmers, D. 1996) and (O’Connor T., Wong H., 2012) respectively. Arguments ties, their properties, and their spatiotemporal for the existence of emergent properties in physics that, explicitly or arguably, present ontological views, include (An- 10. About supervenience see, e.g., (McLaughlin B., 1995). derson P., 1972), (Humphreys N., 1997), (Sewell G., 2002), 11. This is so because a metaphysical necessitation relation (Mainwood P., 2006) and (Howard D., 2007). holds in every nomologically possible world. 9. For a survey of the current state of opinion regarding 12. By ‘the history’ of a physical entity I mean its complete- Bell’s Theorem, see (Bell M., Gao S., forthcoming). trajectory relative to some frame of reference of space-time.
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 7/21 histories. Clearly, if physicalism holds the totality differ. Analogously to the case when physical- of physical items in our world constitutes a suffi- ism is false, W and W** may differ regarding, say, cient supervenience base for everything there is in social phenomena. But now, unlike the case when it, including in particular social reality in Searle’s physicalism is false, W and W** can also differ in sense (Searle J., 1995). physical respects: some composed physical enti- ties of W may lack an identical counterpart in I shall not attempt to provide a complete account W**. of what it is for an item to be physical. This would require an extensive digression with little chances Accordingly, I take the central thesis of micro- of success13. Intuitively, it is clear that a proton, a physicalism to be, using Chalmers’ words, that molecule and electric charge are physical items. “high-level facts and laws are entailed [with By contrast, cheques and prices are not, even logical necessity] by all the microphysical facts though they have a ‘physical nature’ if physicalism (perhaps along with microphysical laws)” (Chal- holds. I will briefly come back to this question mers D., 1996: 71). Now, the ‘microphysical facts’, latter. strictly speaking, are those that exclusively involve simple particles. Therefore, microphysicalism can Physics, being an empirical science, is open to be stated as the view that all the entities inhabit- substantial revisions. But let us accept the stan- ing the actual world, their properties, and all the dard atomistic picture it proposes: Physical enti- acts and events involving them, have a micro- ties are either (simple) particles or systems com- physical nature in the following sense15: posed by them, and the laws of physics govern (or describe) the way these entities interact with (Microphysical Nature) An item has a micro- each other14. I will label ‘basic properties’ the ones physical nature if it L-supervenes on a set of that are instantiated by particles; ‘high proper- basic physical items. ties’ the ones that are systemic, i.e., instantiated ‘Basic physical items’ exclusively includes (simple) by physical systems as a whole; and ‘non-physical particles, their properties, and their spatiotemp- properties’ the remaining ones (which have a oral histories. Clearly, if microphysicalism holds physical nature if physicalism holds). Note that the totality of basic physical items in our world some properties can be attributed as high or as constitutes a sufficient supervenience base for basic, for instance mass and electric charge, while everything there is in it including, besides social others can only be attributed as high, for instance constructs, composed physical systems with their temperature and rigidity. (high) properties and relations. Microphysicalism Note that the microphysical reality on which, Imagine a world W** defined as an exact micro- according to microphysicalism, everything physical replica of the actual world W: All of the L-supervenes, should not be conceived as involv- most basic physical entities of W, together with ing items not belonging to the lowest level of their properties and history, have qualitatively physical ontology. Otherwise, microphysicalism identical counterparts in W**. If microphysicalism would lose what makes it appealing in the first is true, then for W** to be a microphysical replica place: the idea that everything can be ontologic- of W is for it to be a replica tout court: Every- ally reduced in terms of simple entities, with their thing that exists or obtains in W has an identical (basic) properties and relations, without residue. counterpart in W**. In addition to persons, insti- Moreover, if ‘the microphysical’ were to include tutions, etc., in W** are replicated all the com- items beyond the basic ones, a sharp distinction- posed physical systems of W together with their between this fundamental level and higher ones properties, e.g., the macroscopic objects that are could hardly be justified. rigid and have a given temperature. On the con- Microphysicalism is intuitively clear and has trary, if microphysicalism is false, then W and W** the advantage of proposing an austere ontology.
13. For a definition of ‘the physical’ aimed at avoiding Hem- 15. For (Hüttemann A., 2004) microphysicalism can be pel’s dilemma see, e.g., (Papineau D., 2008). articulated in at least three different ways: as a thesis of 14. I use ‘particles’ in a wide sense. In quantum mechanics ‘micro-determination’ (that concerns properties), as a the- there is a wave-particle duality, and quantum field theory sis of ‘micro-government’ (involving laws), and as a thesis provides a unified frame for particles and fields. For instance, of ‘micro-causation’. With the inclusion of a few plausible electrons are treated as excitations of an electron field, and assumptions it is possible to derive these theses from the the excitations of the electromagnetic field are photons. articulation in terms of L-supervenience I advanced.
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 8/21
Imagine we have to create an exhaustive inven- NS-Emergentism tory of all the items that ever existed or obtained in the actual world. If microphysicalism is true, As mentioned, I will be concerned with NS-Emer- gence, which is an ontological and synchronic view. when we include all the basic physical items the 17 inventory is complete. It is undoubtedly useful I shall adopt McLaughlin’s definition17 : forepistemological purposes to include items that If P is a property of w, then P is emergent are not basic, but these would be redundant from if and only if (1) P supervenes with nomo- the ontological point of view1616. logical necessity, but not with logical neces- Certainly, microphysicalism looks appealing and sity, on properties the parts of w have taken is easily taken as being a central tenet of con- separately or in other combinations; and (2) temporary natural science. Indeed, for Chalmers some of the supervenience principles linking (Chalmers D., 1996; Chalmers D., 2010) mater- properties of the parts of w with w’s having P ialist monism is committed to microphysicalism. are fundamental laws […] A law L is funda- For instance, he writes: “It is widely agreed that mental if and only if it is not metaphysically materialism requires that P necessitates all truths” necessitated by any other laws, even together (Chalmers D., 2010: 110, my italics), where “P with initial conditions. (McLaughlin B., 1997: [is] the conjunction of all microphysical truths p. 39) about the universe” (Chalmers D., 2010: 110, my NS-Emergence is ontological in the follow- italics). However, the argument I will present, ingsense: the emergent properties are conceived if sound, shows that contemporary natural sci- as something ‘over and above’ the correspond- ence warrants the plausibility of the denial of ing emergence bases18. Even though their exist- microphysicalism. ence depends on the existence of the bases, they One may wonder if it is not the case that the are further constituents of reality. Ontological truth of microphysicalism entails the truth phys- views, in general, are motivated by the intuition icalism and if, conversely, the truth of physicalism that, even though there surely are tight relations does not entail the truth of microphysicalism. It between high properties and basic ones, and even should be clear that the truth of microphysical- though we can better understand higher-levels ism logically entails the truth of physicalism: if with models at lower-levels, some high proper- ties— the emergent ones—are (at least) as real as microphysicalism is true, a microphysical replica 19 of W is also a physical replica of W. In contrast, the basic ones are considered to be . whether the truth of physicalism entails the truth NS-Emergence is synchronic because the emer- of microphysicalism is an empirical question: that gent properties are conceived as coexisting with a microphysical replica of W is also a physical rep- the corresponding emergence bases. Note that lica of W is not an a priori true statement. Indeed, synchronic relations of emergence cannot be I will motivate on empirical grounds the thesis causal (at least in orthodox views, where causes that this statement is false. After showing that always precede their effects) in contrast with dia- NS-emergentism is incompatible with micro- chronic relations. physicalism, I will argue, on the basis of contem- Now, notice that microphysicalism is incompat- porary physics, for the plausibility of the thesis ible with the existence of NS-emergent proper- that there are NS-emergent physical properties ties. In fact, these views entail respectively: in nature. If this thesis is right, the worlds W* and W** (as defined before) do differ: There are NS-emergent properties in the actual world W 17. McLaughlin’s (McLaughlin B., 1997) view is an elabo- which are replicated (by definition) in W* but not ration of Van Cleve’s (Van Cleve J., 1990) view. Historically, these are reminiscent of Broad’s (Broad C. D., 1925) emer- (for empirical reasons) in W**. gentism. 18. In McLaughlin’s definition of NS-Emergence, the ‘emergence base’ of property P is constituted by the “proper- ties the parts of w have taken separately or in other combi- nations” (McLaughlin B., 1997: 39). 19. According to Hüttemann (Hüttemann A., 2004), the attractiveness of microphysicalism derives precisely from 16. Microphysicalism is not committed to epistemological the success of micro-explanation. Besides, it should be noted reductionism, i.e., it does not entail the possibility of redu- that McLaughlin (McLaughlin B., 1997) himself does not cing to basic physical theory every other scientific theory. assert that the emergent properties he defines do exist.
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 9/21
(MPH) Every high property is logically better claim to the status of emergent proper- necessitated by basic items. ties than do any of the other properties else- where in nature so far nominated for the prize (EMG) Some high properties are not meta- (Howard D., 2007: 150)23. physically but nomologically necessitated, via fundamental laws, by basic items. The EPR Argument In the following section I will argue in favour In a famous article (Einstein A. et al., 1935), Ein- of EMG on the base of contemporary physics. stein, Podolsky, and Rosen claimed that quantum Certainly, it is not possible to prove thereby that mechanics is incomplete and thus unacceptable. microphysicalism is definitely flawed, since any They conceived a Gedanken experiment and scientific theory can be false. However, micro- discussed the solution that quantum mechanics physicalism is motivated by the acceptance of would provide. In Bohm’s (Bohm D., 1951) ver- contemporary theories in natural science, and 24 sion, an ‘EPR experiment’ is as follows : thus it is legitimate to argue against it using con- temporary physics. Nonlocality and NS-Emergence In order to argue for EMG I will elaborate a case with the EPR experiments and Bell’s Theorem20. It has three significant advantages: First, it involves determination relations within microphysics, avoiding problems that appear in attempts to relate properties that belong to signifi- Figure 1. Bohm’s version of an EPR experiment. cantly different ‘levels’ (e.g. the microphysical and the macrophysical) or to different sciences (e.g. Initially, there is a source of electrons and two physics and biology)21. Second, this case avoids Stern-Gerlach magnets. These systems are spa- the causal exclusion problem (Pepper 1926; Kim tially separated, they are not connected by any 1992, 1999, 2005)22. We will notice that quan- further system, and the space between them is tum entanglement is not epiphenomenal and empty. An electron is a particle that has a ‘spin’ yet poses no problem of causal overdetermina- property that, when measured, will always take tion. Third, the existence of EPR correlations is one of two possible values: +½ or -½. A Stern- strongly supported by empirical evidence. Indeed, Gerlach magnet is a device for measuring the spin Howard writes: in a given direction depending on its orientation. How and why the properties of a pair of pre- A row of the EPR experiment, divided in three viously interacting and, therefore, entangled- consecutive intervals, proceeds as follows: quantum systems fail to supervene on the - (Δt1) Both magnets are oriented in the same properties of the two individual systems taken direction α. The source contains a system separately is perfectly well understood and composed of two interacting electrons that today routinely demonstrated in the labora- has a null total spin: ∑S = 025. tory, as in experimental tests of Bell’s theorem […] By my lights, the quantum correlations characteristic of entangled joint states have a 23. Howard (Howard D., 2007) proposes a definition of supervenience for the physical context, and then advances a definition of ‘emergence’ as the failure of supervenience. 20. For a thorough and illustrative exposition without tech- Unfortunately, he does not elaborate on this conception of nicalities of EPR experiments and Bell’s Theorem, see (Mer- emergence and he does not show exactly why and how quan- min D., 1985). tum entanglement would violate supervenience. 21. Indeed, within physics itself the nature of the relation 24. EPR experiments can involve different types of particles between the underlying quantum world and the macrosco- and concern different properties. They are habitually dis- pic systems is an open issue referred to as ‘the measurement cussed under Bohm’s version because it illustrates the situa- problem’. tion in a simpler manner than the original EPR setup and 22. Replies to the causal exclusion argument can be found, makes the question of locality more salient. e.g., in (Loewer B., 2001), (Shoemaker S., 2002), (Wood- 25. According to quantum mechanics since the electrons are ward J., 2008), (Kistler M., 2009) and (Campbell J., 2010). interacting they have become ‘entangled’.
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 10/21
- (Δt2) The source emits the electrons. Elec- ‘locality’ that they deemed non-negotiable would tron e1 travels towards magnet M1 and elec- be violated29. tron e2 towards magnet M2. Let us consider two disjoint closed volumes - (Δt3) Electrons e1 and e2 encounter mag- in physical space, A and B. Locality says that a nets M1 and M2 respectively, and each one is change in the properties of physical systems deflected either in directionα or in direction enclosed in A cannot determine any change in -α. Thereby, a measurement (in directionα ) of the properties of physical systems enclosed in B, the corresponding spins is obtained: S1 = +½ unless there is some signal going from A to B. By if e1 is deflected in directionα , S1 = +½ if e1 a ‘signal’ I mean some wave, or some travelling is deflected in direction –α, and likewise for particle, or something of the sort. In a slogan, the e2. I will label ‘M1/e1’ and ‘M2/e2’ the events principle of locality precludes ‘spooky actions at a of e1 and e2 interacting with M1 and M2 distance’. Now, the holistic picture violates local- respectively26. ity for the following reason: If, during Δt2, the values for S1 and S2 are not determined, at the Quantum mechanics predicts two possible out- moment of the measurement (during Δt3) there comes for each row of this experiment: {S1 = +½; must be a physical relation between M1/e1 and S2 = -½} or {S1 = -½; S2 = +½}. The combinations M2/e2 which ensures that the correlations obtain, {S1 = +½; S2 = +½} and {S1 = -½; S2 = -½} are i.e., that only the outcomes {S1 = +½; S2 = +½} excluded. This is not surprising, since recall that and {S1 = -½; S2 = +½} occur. But, ex hypothesi, ∑S = 0. It also predicts that the two possible out- no physical system connects M1 with M2 and, comes are equiprobable. moreover, the measurements M1/e1 and M2/e2 Now, it happens that quantum mechanics repre- can be made simultaneously. No signal could have sents the state of the system composed of e1 and travelled from M1/e1 to M2/e2 or vice-versa. e2 with a ‘wave function’ that does not attribute In contrast to the holistic picture, the EPR group individual values for S1 and S2 during Δt1 and proposed an intuitive and straightforward explan- Δt2; e1 and e2 are represented as being in ‘entan- ∑ 27 ation: Given that S = 0 and Sn = ±½, only two gled’ states . However, for Einstein et al. S1 and outcomes are possible. The outcome {S1 = +½; S2 S2 must always have definite values. Thus, they = -½} obtains when, during Δt2, e1 is in the state claimed that quantum mechanics provides an 28 S1 = +½ and e2 in the state S2 = -½, and analo- incomplete representation of reality . gously for {S1 = -½; S2 = +½}. Accordingly, the In fact, quantum mechanics provides a ‘holistic’ measurements during Δt3 only revealed the states picture of the experiment: During Δt2 electrons that the electrons were in while travelling towards e1 and e2 are not in states with a definite value the detectors. for their spins, but in a ‘superposition’ of states: e1 Theories along the lines of the last picture are is in a superposition of [S1 = +½; S1 = -½], with called (deterministic) ‘local-hidden-variable- each of these sub-states having the same prob- theories’ (LHVT)30. These theories, first, sup- ability of obtaining, and analogously for e2. Now, pose that the quantum mechanical description of when the measurements M1/e1 and M2/e2 occur, the state of the system composed by e1 and e2 is S1 and S2 acquire definite values that are corre- incomplete in the following sense: there are ‘ele- lated as indicated. Einstein et al. claimed that this ments of reality’ (Einstein A., et al., 1935) that picture resulted from the incompleteness of the are not represented in the theory (the ‘hidden- theory. If this picture were correct, a principle of variables’) which account for the correlations. Second, these theories aim at explaining the cor- 26. In some interpretations of quantum mechanics these relations preserving locality. events cause the wave function to ‘collapse’ (one of them alone is sufficient for the collapse). The entanglement is thusbroken and there is a ‘decoherence’. 27. Technically, a ‘wave function’ is a function defined in a 29. Entangled systems also violate a principle of ‘separab Hilbert space (a complex vector space). When a wave func- lity’ which, according to (Howard D., 1985), was the primary tion, φ12, represents two systems, these are said to be entan- reason for Einstein’s unwavering belief in the incomplete- gled when φ12 cannot be formulated as the product of two ness of quantum mechanics. φ φ functions, 1 and 2, representing these systems separately. 30. (Clauser J., Horne M., 1974) generalised Bell’s Theorem 28. The details of the formulation of the argument are sup- for the case of stochastic LHVT. For a discussion see (Brown pressed for being unnecessary for present purposes. H., Timpson G., forthcoming).
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 11/21
For many years it was considered that no experi- and M2. Accordingly, for each row of the experi- mental evidence could favour one of the men- ment there are four possible results corresponding tioned pictures over the other. Indeed, the EPR to the values of (σx.α) and (σz.β): (+1; +1), (+1; argument was not meant to question the empir- -1), (-1; +1), or (-1; -1), and the product (σx.α) ical adequacy of quantum mechanics; the discus- (σz.β) can take one of two values: +1 or -1. Now, sion was purely conceptual. However, in a sem- the expression <(σx.α)(σz.β)> represents the sta- inal paper John Bell (Bell J., 1964) showed that tistical distribution of the product (σx.α)(σz.β) the prediction given by quantum mechanics for over several rows of the experiment. a ‘generalised’ EPR experiment differs from any According to QMP, each one of the four pos- prediction that a local-hidden-variables-theory- sible combinations of (σx.α) and (σz.β) has an could give. This opened up the possibility for tests associated probability that depends on the angle and empirical arguments. y (the angle between α and β). Let us consider Bell’s EPR experiment three examples. First, take y = 0. This corresponds to the original EPR setup where the magnets are Bell considered a slightly different set-up for oriented in the same direction. Following QMP, theEPR experiment: <(σx.α)(σz.β)>= -1. This coincides with what we saw: Always, when one of the electrons is deflected in the direction of orientation of the magnet, the other is deflected in the opposite direction. Second, take y = π/2. Following QMP, <((σx.α) (σz.β)>= 0. This means that, statistically, there is no correlation between the results in M1/e1 and M2/e2. Sometimes we have (σx.α)(σz.β)= +1, sometimes (σx.α)(σz.β)= -1, and both situations are equiprobable. Finally, take y = π/4. Following QMP, <(σx.α) Figure 2. Generalised EPR experiment. (σz.β)>= 2½ /2 ≈ 0,7. This means that it is more probable to obtain [(σx.α)(σz.β)]= +1 than Now, there is an angle y between the orientations [(σx.α)(σz.β)]= -1. Approximately two times out of magnets M1 and M2. Under these conditions, of three both electrons are deflected either in the quantum mechanics provides the following pre- 31 directions of orientation of the respective meas- diction (Bell 1987: 15) : uring magnets, or in directions contrary to these. (QMP) <(σx.α)(σz.β)> = -cos y Bell’s Theorem The symbols σ{ x; σz; α; β} stand for unitary vec- tors, and a dot between two of them represents Bell’s theorem has different versions that result the scalar product. Respectively, vectors {α; β} from the proof of a ‘Bell inequality’32. Since these represent the orientations of magnets M1 and proofs are inevitably highly technical I shall not M2 (they form an angle y), vectors {σx; σz} repre- present one33. However, the meaning and con- sent the spin-states of e1 and e2, and the products sequences of the theorem can be presented in (σx.α) and (σz.β) represent the measurements non-technical terms. M1/e1 and M2/e2. Bell’s Theorem concerns the predictions that The products σ( x.α) and (σz.β) can take two pos- (deterministic) LHVT can provide for gener- sible values: +1 and -1. If we have (σx.α) = +1 this alised EPR experiments. According to these means that e1 was found to have its spin in direc- theories, each of the spin-states that e1 and tion α (the orientation of M1). If on the contrary e2 occupy (represented by σx and σz) is com- we have (σx.α) = -1, this means that e1 was found pletely determined independently of the other to have its spin in the direction -α (opposite to (hence locality), by inaccessible features (hence the orientation of M1); analogously for (σz.β), e2 32. Bell’s papers where he presents or discusses his theorem 31. This prediction is derived from the wave function that are compiled in (Bell J., 1987). describes the system composed by the interacting electrons 33. For an explanatory derivation and discussion of Bell’s e1 and e2. Theorem see, e.g., (Greenstein G., Zajonic A., 1997).
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 12/21 hidden- variables) that account, respectively, for They showed a systematic violation of Bell’sIn- the values of (σx.α) and (σz.β) (which represent equalities and great accuracy in the predictions of the measurement M1/e1 and M2/e2)34. quantum mechanics. The theorem, in one of its forms (Bell 1987: 37), Note that what Bell’s Theorem and the corres- results from the proof of the following inequality: ponding experiments strongly suggest is that theories that assume the existence of hidden vari- (BT) | P(α, β) - P(α, β’) | + | P(α’, β) - P(α’, ables and locality (i.e. LHVT) cannot be accurate. β’) | ≤ 2 A nonlocal-hidden-variables-theory can repro- In BT two possible orientations for each magnet duce the predictions of quantum mechanics, as is α α 36 are considered: and ’ are orientations of M1, the case with Bohm’s (Bohm D., 1951) theory . and β and β’ are orientations of M2. The function The correlations predicted by quantum mechan- P represents the left hand of QMP: ics do not depend on whether or not the measure- α β σ α σ β P ( , ) = <( x. )( z. )> ments of spin are realised simultaneously. Thus, P (α’, β) = <(σx.α’)(σz.β)> the hypothesis of the existence of unnoticed sig- nals connecting events M1/e1 and M2/e2 must be Etc. discarded (unless one is willing to sacrifice a prin- Bell’s Theorem states that any LHVT is con- ciple of relativity which says that there is an upper strained by inequality BT. Accordingly, any pre- limit for the speed of any signal connecting two 37 diction of a LHVT for the quantity on the left, spatiotemporal events) . Indeed, no ‘information’ for any values of {α; β; α’; β’}, is going to be minor can be transmitted from M1/e1 to M2/e2: from or equal to 2. the results M1/e1 it is not possible to deduce the results M2/e2, except for the particular cases with In contrast to LHVT, quantum mechanics, as 38 α = ± β (the original EPR cases) . expected, violates BT for certain values of {α; β; α’; β’}. Note that, since α.β = cos y, QMP can be In short, Bell’s Theorem shows that the system rephrased as: composed by e1 and e2 behaves in a way, described by QMP, which does not result from a causal (QMP) P (α, β) = - α.β interaction between intrinsic properties of e1 and Now, let us take, e.g., {α = 0; β = 3π/8; α’ = -π/4; β’ e239. In the next section, I will show why Bell’s = π/8}. From QMP we obtain: theorem plus the empirical adequacy of quantum mechanics undermine microphysicalism. P (α, β) = 1/2½ P (α, β’) = -1/2½ Microphysicalism and the P (α’, β) = 1/2½ EPR experiment P (α’, β’) = -1/2½ Recall McLaughlin’s definition of NS-Emer- gence, and consider, in the EPR experiment, the With these values, the inequality in BT is violated: system that is constituted by the entangled elec- | P (α, β) - P (α, β’) | + | P (α’, β) - P (α’, β’) | trons. On the one hand, this system, w, has two = (2) (2½) > 2 Several EPR experiments to test ‘Bell’s Inequal- bard J., and Roger G., 1982). ities’ have been conducted, the most famous of 36. Bohm’s theory is nonlocal in a very deep sense. In the 35 words of Greenstein and Zajonic, it “goes beyond simple which were by A. Aspect and his colleagues . nonlocality, and calls upon us to see the world as an undivi- ded whole” (Greenstein G., Zajonic A., 1997: 148). 34. According to local-hidden-variables-theories, the fact that quantum mechanics’s prediction has a statistical form, 37. For a proof that the EPR correlations are compatible namely the distribution <(σx.α)(σz.β)> = -cos y, reflects our with special relativity see (Shimony A., 1986). ignorance of the underlying mechanisms governing, on one 38. It could be claimed that in the particular cases α = ± β hand, the behaviour of e1 when it interacts with M1, and on there is information flow. However, if there were informa- the other hand, the behaviour of e2 when it interacts with tion flow in these cases there should be information flow in M2. every case. QMP does not give to the cases with α = ± β any 35. The first experiments to test Bell’s inequalities were do- privileged status. neby (Clauser J., et al., 1969) and (Freedman S., and Clau- 39. Causality is such that, if event c causes event e, c hap- ser J., 1972). Aspect’s experiments were done in the early pens before e. The idea of instantaneous and yet causal re- eighties. See: (Aspect A., Grangier P., and Roger G., 1981); lations is contrary to relativistic physics and, in general, to (Aspect A., Grangier P., Roger G., 1982); (Aspect A., Deli- current orthodoxy.
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 13/21 parts, corresponding to e1 and e2. Each of these the significance of Bell’s Theorem. A LHVT sup- electrons occupies a spin-state (represented by σx poses that the parts of w, e1 and e2, individually and σz), and thus each part of w can be said to instantiate some properties (corresponding to the have the property of being in a spin-state40. It is hidden variables), h1 and h2 respectively, which in virtue of these properties, p1 and p2, that e1 completely determine the behaviour of the system, and e2 are deflected when they encounter mag- particularly, the property P of w. That no LHVT nets M1 and M241. On the other hand, w occu- can be accurate means that the system w has a pies a state of quantum entanglement of its con- property that does not result with metaphysical stituents e1 and e2, and thus w can be said to have necessity from the properties of its individual the high (or ‘systemic’) property P of being in this constituents. state42. It is in virtue of w having property P that Note that P is causally efficacious. Let us adopt the measurements M1/e1 and M2/e2 produce the following criteria for causality, restricted to results that have the statistical correlations that the domain of Physics: An event c causes an event QMP captures. e if and only if: (i) There is a nomological relation Now, it seems clear that P supervenes on p1 and between c and e, (ii) c temporally precedes e, and p2 with nomological necessity, not with logical (iii) there is a transfer of energy—a conserved necessity, in virtue of fundamental laws of nature. quantity—from c to e43. Now, take ‘c’ as the event First, the relation between p1, p2 and P is nomo- corresponding to the emission of the two elec- logical. In fact, there is a correlation among σx trons by the source, and ‘e’ as the conjoint event and σz, and thus a relation between p1, p2 and P, {M1/e1; M2/e2} that reveals nonlocal correla- which are contingent. The prediction that quan- tions. Clearly, there is a causal relation between tum mechanics provides via its laws, i.e. QMP, c and e. Firstly, there is a nomological relation has empirical content. Indeed, according to Bell’s between event c and e, and c precedes e. After the Theorem any LHVT would make a different pre- emission of a pair of electrons, the interactions diction. Second, P (nomologically) supervenes M1/e1 and M2/e2 occur following the pattern on p1 and p2. Two systems w and w’, which are given by QMT. Secondly, there is a transmis- both constituted by two interacting electrons that sion of energy (by means of a collision) from the instantiate p1 and p2, cannot differ with respect source of electrons to the magnets M1 and M2 to instantiating P. Third, underlying QMP there which makes them flash44. Now, the causal rela- are fundamental laws. In particular, that freely tion between c and e constitutively involves (given interacting systems become entangled is a fun- Bell’s Theorem) property P. It is this property damental principle of quantum mechanics; it is what is causally responsible for a property e has, not derived from other principles that would be namely, the correlations obtaining between M1/ more fundamental. In sum, the EPR experiments e1 and M2/e2. Accordingly, P makes a difference and Bell’s Theorem seem to show that there are in causal interactions: it has the causal power of physical systems (at least the ones involved in producing correlations45. these experiments) which instantiate properties Additionally, note that P has a downward causal that are something over and above the properties influence which, as previously announced, eludes of their constituent parts. the causal overdetermination problem. This prob- Another way to see the implications of the EPR lem arises when an effect has more than one suf- experiments for microphysicalism is to focus on ficient cause, and this is not the case withe . As we saw, even though the systemic property P causally 40. It is unusual to refer to quantum systems as having pro- influences the behaviour of the constituents of perties. These are normally said to occupy states, described the system e1 and e2, only c causes e. In fact, P’s by the corresponding wave functions. However, given the definitions proposed by Chalmers (Chalmers D., 1996) of ‘supervenience’ and by McLaughlin (McLaughlin B., 1997) of ‘emergence’, I should also talk in terms of ‘properties’. Thus, when talking about a system that is in a stateS , I will say that ‘it has the property’ of being in a state S. 43. For a defence of Conserved quantity theories of causa- 41. Recall that to say that e1 and e2 occupy spin-states σx lity see, e.g., Dowe (2000) and Salmon (1998). and σz is not to say that there is a definite value for their res- 44. In general, every device that detects X requires to be pective spins before the events M1/e1 and M2/e2 take place. perturbed by X, and this involves an energy transfer. 42. This state is represented by a wave function from which 45. Alternatively, we can say that because the EPR system QMP is derived. has the nonlocal property P, event c causes event e.
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 14/21 causal work is to constraint the behaviour of e1 that consciousness must be an emergent property, and e2 in the way described by QMT46. emerging from physical items48. In sum, the systems of the EPR experiments I would argue that the prima facie conceivabil- seem to violate MPH and hence microphysical- ity of Chalmers’ zombies results from the fact ism. There is no metaphysical necessitation from that consciousness is an intrinsic property. As (1) the fact that an electron is a particle with a Chalmers himself says when discussing Type-F spin that, when measured, is found to be such that (or ‘Russellian’) monism, “physics characterizes it can only take two opposite values with equal physical entities and properties by their relations probability, to (2) the fact that a system consti- to one another and to us. […] physics says noth- tuted by two electrons and with null total spin ing about the intrinsic nature of these entities and behaves, when the spins are measured, in accord- properties” (Chalmers, D., 2010: 133). Therefore, ance with QMP. By contrast, the systems of the “[w]e only think we are conceiving of a physically EPR experiments seem to support EMP, i.e., they identical system [when we conceive the zombie seem to instantiate a high property nomologically world] because we overlook intrinsic properties” necessitated, via fundamental laws, by basic items. (Chalmers, D., 2010: 134). It is important to note that here Chalmers refers to intrinsic properties Back to Chalmers’ Zombie of particles; if consciousness is a NS-emergent Argument property, it is an intrinsic property of composed (and very likely extremely complex) physical In the previous sections I questioned the validity systems49. of Chalmers’ Zombie Argument, by arguing that there are good reasons to consider that micro- To finish, I will briefly discuss five objections to physicalism, and thus the claim that “high-level the argument I have presented. facts and laws are entailed [with logical neces- Objection 1: The claim that consciousness could sity] by all the microphysical facts” (Chalmers be a NS-emergent property coincides with Chal- 1996: 71), are wrong. I did so by showing a case mers’ (Chalmers D., 1996) “naturalistic dualism” in favour of NS-Emergentism. view: he claims, precisely, that consciousness To be sure, I did not give reasons to consider that supervenes nomologically, but not metaphysically, phenomenal consciousness is a NS-emergent on the microphysical. property. Indeed, from the fact (if it is a fact) that Reply: Chalmers says when presenting natur- some relatively simple quantum systems do have alistic dualism: “It is therefore more natural to NS-emergent properties, nothing seems to follow consider experience as a fundamental property about the nature of consciousness. In principle, that is not a physical property, and to consider only extremely complex macroscopic biological the psychophysical laws as fundamental laws of organisms are conscious, and consciousness nature that are not laws of physics” (Chalmers D., raises distinctive philosophical questions. How- 1996: 128 – 129). Accordingly, for Chalmers the ever, notice that the fact (if it is a fact) that some psychophysical laws that would be involved in relatively simple quantum systems instantiate the nomological supervenience of consciousness NS-emergent properties asks for a non-micro- relate something physical with something that is physicalist ontology of the natural world, which not physical (in accordance with the conclusion opens up the possibility for there being emergent 47 of The Zombie Argument). This contrasts sharply properties beyond microphysical systems . Now, with the conception of NS-emergentism that I there are reasons which are independent of any advanced, which involves physical laws that relate consideration about quantum mechanics to hold physical items.
46. See Kistler (2009) for an elaboration of the idea that the causal powers of emergent properties consist in being constraints on the behaviour of entities at lower levels. 48. In (Bernal Velásquez R. J., 2012) I argue that emergen- 47. Indeed, quantum mechanics does not restrict entan- tism about consciousness is the best alternative if one is a glement to microphysical systems composed by a few par- materialist realist about consciousness. About emergence in ticles. In principle, any interacting physical systems become the philosophy of mind see (MacDonald C., MacDonald G., entangled. Of course, there is the question of why we do not 2010). experience (at the macroscopic level) any superposition of 49. In (Bernal Velásquez R. J., 2012) I argue that states, which is the core of the measurement problem. For a consciousness is an intrinsic property and that it is instantia- discussion see, e.g., (Brown H., Wallace C., 2005). ted in complex systems.
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 15/21
Follow-up to Objection 1: Suppose there are quantum entanglement. Thereby, the putative indeed psychophysical laws in virtue of which NS-emergent property P would supervene on the consciousness NS-emerges. Now, there is still the microphysical with logical necessity. possibility for consciousness to be non-physical. Reply: This strategy would make microphysical- In fact, it is arbitrary to take the psychophysical ism collapse into the view I advocate, namely, that laws as being ‘laws of physics’ and, thereby, to everything is either a physical item or L-super- affirm that consciousness is physical, instead of venes on physical items. If in global superven- taking the psychophysical laws—following Chal- ience physicalist metaphysics one includes in the mers—just as fundamental laws of nature that do supervenience base the NS-emergent property not suppose that consciousness is physical. P, by the same token one should include all the Reply: I am not sure how to understand in what different kinds of NS-emergent properties that sense a law of nature could relate something exist, and notice that there are no reasons to dis- physical with something non-physical. These card them in complex and macroscopic systems. laws are, minimally, statements that capture Thereby, firstly, one would be giving up micro- regularities in nature. Now, if there is a regular- physicalism. Secondly, if consciousness in par- ity involving a paradigmatically physical item ticular is a NS-emergent property, and as such, M and some item N, this seems to be sufficient one has to include it in the supervenience base, for N to fall in the category of having a physical this amounts to considering it as physical. nature. Why we think that, e.g., electromagnetic Objection 3: Suppose that in the actual world W waves are something physical? After all, they are there is in fact a law L in virtue of which con- not objects of experience in any ordinary sense of sciousness NS-emerges. Now, consider a world the term; we represent these waves through quite W* that lacks L, and hence consciousness, but is abstract mathematical equations. otherwise identical to W in every respect. W* is Suppose there is some item N whose meta- conceivable and would be a zombie world: every physical nature is an open question. Now, sup- physical item in W* (except consciousness) would pose we establish the existence of a law of nature have its counterpart in W. relating N with some paradigmatic physical item Reply: Ontological views of emergentism, and M. Then, thanks to our knowledge of this law, we in particular emergentism about consciousness, are able to produce or manipulate N, to explain usually conceive emergent properties as having or predict its instances, to account for its inter- causal powers. Indeed, it is common to argue actions with M (and thereby with other physical for the existence of an emergent property E on items), and so forth, just as we are able to do with the basis of Alexander’s Dictum (Alexander S., electromagnetic waves. This seems sufficient to 1920): E is real because its instantiation endows claim that N has a physical nature50. its bearer with irreducible causal powers. So, W* Certainly, if N is an epiphenomenal property, the is not really conceivable: in the absence of L it previous line of thought does not apply: it is not would differ from W in many respects, namely, possible to determine if N is instantiated, N does regarding the events directly or indirectly related not interact with anything, and so forth. But then, to consciousness in a causal way. I do not see on what grounds we could possibly Objection 4: Chalmers writes: “a microphysical claim that there indeed exists a property N, and description of the world specifies a distribution that it falls under some law of nature. of particles, fields, and waves in space and time. Objection 2: Chalmers wrote: “high-level facts These basic systems are characterized by their and laws are entailed by all the microphysical facts spatiotemporal properties and properties such as (perhaps along with microphysical laws)” (Chalmers mass, charge, and quantum wave function state” D., 1996: 71, my italics). So, let us include within (Chalmers D., 2010: 120, my italics). So, suppose the microphysical laws those that correspond to that in the actual world W there are NS-emer- gent properties, and consider a world W* that is 50. Note that this line of argument leads to the claim of a spatiotemporal microphysical replica of W, i.e., the instability of interactionism, i.e., of Type-D dualism in Chalmers’ (Chalmers D., 2002) taxonomy. The factual exis- not a time slice microphysical replica. The fullhis - tence of nomological interactions between conscious states tory of every basic item in W has its counterpart and physical states (which may include brain states) would in W*. Is it conceivable for W and W* to differ be sufficient to classify consciousness as physical.
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 16/21 in non-microphysical levels? Apparently not: In the case for emergentism that I discussed, the in all probability, spatiotemporal high levels do causal role of the NS-emergent property is a con- L-supervene on the spatiotemporal microphysical straining role. In virtue of property P, e1 and e2 level. Thus it seems that, when W* is appropriately behave in a correlated way when the spin meas- formulated, The Zombie Argument is valid even urements occur. Let us take it that, in general, if there are NS-emergent properties in W. when it comes to causality NS-emergent proper- ties act as constraints for the behaviour at lower Reply: This is a subtle point involving downward levels53. Now: causation whose thorough discussion exceeds the scope of the present paper. Thus, I will only Take two subsequent times t1 and t2 in the his- advance two remarks about the following scenar- tory of the actual world W, and two subsequent ios: (1) there is no downward causation; (2) there times t1* and t2* in the history of W*, such that: is downward causation. (i) between t1 and t2 there were downward effects (in W) due to the existence of a NS-emergent (1) Suppose the NS-emergent properties in property P; (ii) the evolution of the basic items of W do not have downward causal powers. Then, W* between t1* and t2* mirrors the evolution of it is certainly conceivable for W and W* to dif- the basic items of W between t1 and t254. fer in non-microphysical levels: In W, on top of the basic items, there are further items (the Clearly, the microphysical evolution of W between NS-emergent ones). t1 and t2 is nomologically determined by P: W finished att2 in a particular microstate W(2) However, that a NS-emergent property has partly because of P. By contrast, W* finished at causal powers seems to entail that it has down- time t2* in a particular microstate W*(2) partly by ward effects and, as I said, epiphenomenal emer- chance. In W there is a law of nature, namely the gent properties are not what ontological emer- 51 one that captures the effect ofP , which W* lacks. gentism is usually about . Thus, this first scenario is not really relevant. Objection 5: Forget about microphysicalism. The core of The Zombie Argument is the idea that no (2) Suppose that the NS-emergent properties in phenomenal fact can be metaphysically necessi- W do have downward causal powers: There are tated by physical facts; it points to the existence top-down causal effects that influence the behav- of an ontological gap. Now, to judge that con- iour at the microphysical level. Now, recall that sciousness may be a NS-emergent property is to W*, ex hypothesi, is identical to W at this level. The deny the existence of this gap, since consciousness microphysical level in W* behaves like the one in would emerge exclusively from physical stuff. W despite the absence of NS-emergent proper- ties. In this scenario, is it possible for W and W* Reply: I agree with the idea that phenomenal to differ at high levels? I will focus on a particu- facts cannot result from physical facts in virtue of lar conception of the causal powers of NS-emer- a metaphysical necessitation relation. But, firstly, gent properties and argue that W and W*, despite recall that I take NS-emergentism as a form of having the same microphysical dynamics, differ physicalism. Therefore, phenomenal facts, despite regarding the laws of nature. If one is a neces- our deeply grounded dualist intuitions, are taken sitarian about laws, this might be a decisive as physical facts. Secondly, recall that NS-emer- point. If, on the contrary, one is a regularist or an gence involves a nomological, not a metaphysical, antirealist, this might seem innocuous52. relation, mediated by fundamental laws of nature. It would be in virtue of a law of nature that a 51. The entailment from an emergent property having cau- complex physical entity would have phenomenal sal powers to it having downward effects is a critical step in properties even though none of its physical parts the formulation of the causal exclusion problem. See, e.g., has them. But this last point, clearly, is persuasive (Kim J., 2005). only for someone who holds a realist and necessi- 52. For the regularists, statements of laws of nature only capture the regularities we observe in the natural world. tarian view about laws. Accordingly, laws are nothing over and above these regu- larities. By contrast, for necessitarians the laws of nature are something over and above mere regularities. Their existence explains why some regularities factually obtain, and why they would obtain were counterfactual situations (framed in 53. (Bishop R., Atmanspacher H., 2006) and (Kistler M., physically possible worlds) realised. See, e.g., (Schrenk M., 2009), among others, propose this view. forthcoming). 54. Note that t1* and t2* do exist ex hypothesi.
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 17/21
Conclusion Bell John (1964), “On the Einstein-Podolsky- Some arguments against the thesis that con- Rosen paradox”, Physics 1: 195-200. sciousness is a physical property, and in particular Bell John (1987), Speakable and Unspeakable in Chalmers’ Zombie Argument, suppose the valid- Quantum mechanics, Cambridge: Cambridge ity of ‘microphysicalism’: roughly, the view that all University Press. the physical properties of composed physical sys- Bell Mary & Gao Shan (Eds.) (forthcoming), 50 tems are metaphysically necessitated by their fun- Years of Bell’s Theorem, forthcoming in Cambridge, damental physical constituents. Through the dis- Cambridge University Press. cussion of the EPR scenario and Bell’s Theorem, I questioned the validity of microphysicalism and Bernal Velásquez Reinaldo José (2012). E-phy- favoured emergentist metaphysics in the sense sicalism. A Physicalist Theory of Phenomenal of the nomological supervenience views. If this Consciousness. Ontos Verlag (2012)/De Gruyter metaphysics happens to be correct, Chalmers’ (2013). Argument loses part of its grounds, and, a possi- Bishop Robert, Atmanspacher Harald (2006), bility opens up for consciousness to be a physical “Contextual emergence in the description of pro- property. perties”, Foundations of Physics 36: 1753–1777. Acknowledgements Block, Ned (1998), “On a confusion about a func- tion of consciousness”, in Block Ned, Flanagan I am indebted to the Fondation Maison des Owen & Guzeldere Guven (Eds.), The Nature of Sciences de l’Homme (FMSH) for the Fer- Consciousness: Philosophical Debates, Cambridge nand Braudel-IFER fellowship that enabled (MA), MIT Press. me to pursue my research from March to Bohm David (1951). Quantum Theory, New York, November 2015, to the Institut Jean-Nicod Prentice Hall. (CNRS/ENS/ EHESS) for hosting me during this period, to Uriah Kriegel for his Broad Charlie Dunbar (1925), The Mind and Its Place in Nature, London, Routledge & Kegan mentoring at this institute and very useful Paul. commentaries on the final drafts of this paper, and to Santiago Echeverri, Ignacio Brown Harvey & Timpson Christopher (forth- Avila and Harriet Levenston for very va- coming), “Bell on Bell’s theorem: The changing luable remarks on previous drafts. face of nonlocality”, in Bell M., Gao S., (Eds.) (forthcoming). References Brown Harvey, Wallace David (2005), “Solving Alexander Samuel (1920), Space, Time, and Deity, the Measurement Problem: De Broglie-Bohm London, Macmillan. loses out to Everett”, Foundations of Physics, 35: Anderson Philip (1972), “More Is Different”, 570–540. Science, New Series 177(4047): 393-396. Campbell John (2010), “Control Variables and Aspect Alain, Grangier Philippe, Roger Gerard Mental Causation”, Proceedings of the Aristotelian (1981), “Experimental Tests of Realistic Local Society 110: 15–30. Theories via Bell’s theorem”,Phys Rev Lett 47: Chalmers David (1996), The conscious mind, 460-463. Oxford, Oxford University Press. Aspect Alain, Grangier Philippe, Roger Gerard Chalmers David (2002), “Consciousness and its (1982), “Experimental realization of Einstein- Place in Nature”, in Stich Stephen, Warfield Ted Podolsky-Rosen-Bohm Gedanken Experiment: (Eds.), Blackwell Guide to the Philosophy of Mind, A new violation of Bell’s inequalities”, Phys Rev Oxford, Blackwell. Lett 49: 91-94. Chalmers David (2010), The Character of Aspect Alain, Delibard Jean, Roger Gerard Consciousness, Oxford, Oxford University Press. (1982), “Experimental tests of bell’s inequalities Clauser John, Horne Michael, Shimony Abner, using time-varying analysers”, Phys Rev Lett 49: Holt Richard (1969), “Proposed experiment to 1804-1807.
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 18/21 test local hidden-variable theories”, Phys. Rev. Kistler Max (2009), La cognition entre réduction Lett. 23 (15): 880–4. et émergence. Étude sur les niveaux de réalité. Paris, Thèse d’habilitation. Clauser John, Horne Michael (1974), “Experi- mental consequences of objective local theories”, Kripke Saul (1980), Naming and Necessity, Cam- Physical Review D 10 (2): 526. bridge (MA), Harvard University Press. Crane Tim (2001), “The Significance of Emer- Loewer Barry (2001), “Review of J. Kim, Mind gence”, in Barry Loewer, Carl Gillett (Eds.) in a Physical World”, Journal of Philosophy 98(6): (2001), Physicalism and Its Discontents, Cam- 315–324. bridge, Cambridge University Press. Mainwood Paul (2006), Is more different? Dowe Phil (2000), Physical Causation. New York, Emergent properties in physics, Dissertation, Uni- Cambridge University Press. versity of Oxford. Einstein Albert, Podolsky Boris, Rosen Nathan Macdonald Cynthia & Macdonald Graham (1935), “Can quantum-mechanical description (2010), Emergence in Mind, Oxford, Oxford Uni- of physical reality de considered complete?”, Phys versity Press. Rev 47: 777-780. McLaughlin Brian (1995), “Varieties of Super- Freedman Stuart, Clauser John (1972), “Expe- venience”, in Savellos Elias, Yalcin Ümit (Eds.), rimental test of local hidden-variable theories”, Supervenience: New Essays, Cambridge, Cam- Phys Rev Lett 28: 939-941. bridge University Press. Greenstein George, Zajonic Arthur (1997), The McLaughlin Brian (1997), “Emergence and Quantum Challenge, Modern Research on the Foun- Supervenience”, Intellectica, 2: 25–43. dations of Quantum mechanics, London, Jones and Mermin, David (1985), “Is the moon there when Bartlett Publishers. nobody looks?”, Phys Today 38: 38-47. Howard Don (1985), “Einstein on Locality and Nagel Thomas (1974), “What is it like to be a Separability”, Studies in History and Philosophy of bat?”, Philosophical Review 83: 435-456. Science 6: 171-201. Nida-Rümelin Martine (2006), “Dualist Emer- Howard Don (2007), “Reduction and emergence gentism”, in McLaughlin Brian & Cohen Jona- in the physical sciences: some lessons from the than (Eds.) (2006), Contemporary Debates in Phi- particle physics and condensed matter debate”, losophy of Mind, Blackwell. Evolution and Emergence: Systems, Organisms, Per- sons: 141-157. O’Connor Tim, Wong Hong Yu (2012), “Emergent Properties” in Edward Zalta (Ed.), The Humphreys Paul (2008), “How Properties Stanford Encyclopedia of Philosophy (Spring 2012 Emerge”, in Bedau Mark, Humphreys Paul Edition).
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 19/21
Sewell Geoffrey (2002),Quantum Mechanics and Its Emergent Macrophysics, Princeton, Princeton University Press. Shimony, Abner (1986), “Events and Processes in the Quantum World”, in Penrose Roger, Chris Isham (Eds.) (1986), Quantum Concepts in Space and Time, Oxford, Clarendon Press. Shoemaker Sydney (2002), “Kim on Emergence”, Philosophical Studies 108: 53-63. Schrenk Markus (forthcoming). Metaphysics of Science. A systematic and historical introduction. Forthcoming in Routledge in 2016. Strawson Galen et al. (2006), Consciousness and its place in nature, Exeter, Imprint Academic. Van Cleve James (1990), “Mind-Dust or Magic? Panpsychism Versus Emergence”, Philosophical Perspectives 4: 215-226. Woodward James (2008), “Mental Causation and Neural Mechanisms”, in Hohwy Jakob, Kalles- trup Jesper, (Eds.), Being Reduced: New Essays on Reduction, Explanation, and Causation, Oxford, Oxford University Press.
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 20/21
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120 An Emergentist Argument for the Impossibility of Zombie Duplicates 21/21
Working Papers parus en 2015 et en 2016
Andrea Lanza Un organicisme de Matthieu Renault, Préface à Stefania Tarantino, La liberté la complexité. Notes pour un cha- la révolution. C.L.R. James, et l’expérience politique des pitre sur le socialisme et les sciences lecteur de Melville, FMSH- femmes face à la crise : les fémi- naturelles (France, première moi- WP-2015-104, juillet 2015. nismes italiens et leurs prolonge- tié du XIXe siècle), FMSH- ments au XXIe siècle, FMSH- Yang Dongping, Enseigne- WP-2015-96, juin 2015. WP-2016-112, avril 2016. ment supérieur, équité et société, Vincent Duclos, Le design du FMSH-WP-2015-105, octobre Camille Devineau, BwEni, un monde. De McLuhan à Sloter- 2015. mot pour dire ce qu’implique être dijk, vers une anthropologie de griot chez les Bwaba, FMSH- Olivier Galland, Un pacte l’espace en devenir, FMSH- WP-2016-114, septembre 2016. implicite entre les générations WP-2015-97, juin 2015. pour le statu quo, FMSH- Anne Marie Moulin, La Mathias Grote, What could the WP-2015-106/GeWoP-9, « vérité » en médecine selon son ‘longue durée’ mean for the his- novembre 2015. histoire, FMSH-WP-2016-115, tory of modern sciences?, FMSH- octobre 2016. Anaïs Albert, Consommation de WP-2015-98, june 2015. masse et consommation de classe à Priya Ange, Ethnographie Philippe Steiner, Comte, Altruism Paris des années 1880 aux années des bijoux de Franco-pondi- and the Critique of Political Eco- 1920 : bilan d’une recherche, chérien-ne-s. Au cœur d’une pro- nomy, FMSH-WP-2015-99, FMSH-WP-2016-107, janvier duction des relations de genre et GeWoP-8, june 2015. 2016. parenté, FMSH-WP-2016-116, décembre 2016. Pierre Salama, Argentine, Bré- Philippe Rousselot, Les forces spé- sil, Mexique entrent dans la ciales américaines : vers une refon- Dirk Rose, « L’époque polé- tourmente. Quo vadis Amérique dation?, FMSH-WP-2016-108, mique », FMSH-WP -2016- latine ?, FMSH-WP-2015-100, mars 2016. 117, 2016. juin 2015. Priscilla Claeys, The Right to Pierre Salama, L’Argen- Ayşe Yuva, Die Historisierung der Land and Territory : New Human tine marginalisée, FMSH- Philosophie in Deutschland und Right and Collective Action WP-2016-118, décembre 2016. Frankreich nach Kant, FMSH- Frame, FMSH-WP-2016-109, Marie-Paule Hille, Le dicible et WP-2015-101, october 2015. mars 2016 l’indicible. Enquête sur les condi- Elisa Brilli, Mettre en image les Matias E. Margulis and Pris- tions d’écriture d’une histoire reli- deux cités augustiniennes (ms. cilla Claeys, Peasants, Smallhol- gieuse au sein d’une communauté Florence, BML, Plut. 12.17), ders and Post-Global Food Crisis musulmane chinoise, FMSH- FMSH-WP-2015-102, octobre Agriculture Investment Programs, WP-2016-119, décembre 2016. 2015. FMSH-WP-2016-110, avril 2016 Julie Patrier, Les dépôts alimen- taires dans les tombes d’Anatolie Antoine Kernen et Antoine centrale au IIe millénaire av. J.-C., Guex, Partir étudier en Chinee FMSH-WP-2015-103, octobre pour faire carrière en Afrique, 2015. FMSH-WP-2016-111, avril 2016.
Retrouvez tous les working papers et les position papers sur notre site, sur hypotheses.org et sur les archives ouvertes halshs http://www.fmsh.fr/fr/ressources/working-papers http://halshs.archives-ouvertes.fr/FMSH-WP http://wpfmsh.hypotheses.org
Fondation Maison des sciences de l’homme - 190 avenue de France - 75013 Paris - France http://www.fmsh.fr - FMSH-WP-2016-120