Version classiqueVersion mobile

Le libre arbitre

Jean-Baptiste Guillon

I. Conceptions incompatibilistes du libre arbitre

Chapitre 3

Agent-Causation and supervenience: How you can have one without giving up the other

Joana Rigato

Texte intégral

1. Introduction

1According to incompatibilist theories about free will, in a world where determinism is true, there can be no free actions, since, given the past and laws of nature, only one future is possible. On the opposite side of the barricade stand compatibilist theories, according to which people can act freely despite causal determinism.

2Compatibilism is the current favored view among believers in free will. Nevertheless, several incompatibilist attempts to account for free action have been put forward in the past decades: event-causal theories (Kane 1996, Ekstrom 2000, Clarke 2003), agent-causal theories (O’Connor 2000, E.J. Lowe 2008), and non-causal theories (Ginet 1990, McCann 1998, Goetz 2008). Event-causalism, according to which free actions are caused by mental events of a certain sort, is clearly the most popular of all libertarian views. Non-causalist accounts have instead been met with suspicion insofar as they consider free actions not to be caused at all, and it is hard to understand how an uncaused action can remain under the agent’s control and be made for reasons. Finally, agent-causal theories have generally been accused of being unscientific, as they require free actions to be ultimately caused by the agent’s substantial self and thus rely on a non-reductionist account of the agent.

3In this paper, I will attempt to help reevaluate the plausibility of the agent-causalist perspective. I will briefly mention what I consider to be the main argument in favor of it, and then show how the implications of endorsing this type of account are entirely compatible with a scientific worldview.

2. The disappearing agent argument

4Agent-causalists usually criticize the event-causalist mainstream view because of its inability to endow the agent with a sufficiently active role in the production of the action. Derk Pereboom has championed the so called “disappearing agent argument”, which can be stated as follows:

If the decision remains genuinely undetermined up to the very last moment, and if the agent’s neural and mental events are what ultimately brings about that decision, then she has insufficient control over which decision is eventually made.

5This is a very strong consequence of both the first presupposition of any incompatibilist account (that the decision is indeterministically caused) and event-causalism (the idea that everything that happens in the world, including free actions, is caused by occurrences rather than by objects). These two assumptions make it so that the agent as such becomes irrelevant in the action’s etiology, since what causes the action are states and events that are already given at the moment of decision and which, by definition, are insufficient to determine that decision. In the words of Pereboom:

With the causal role of the antecedent events already given, whether the decision occurs is not settled by any causal factor involving the agent. In fact, given the causal role of all causally relevant antecedent events, nothing settles whether the decision occurs. (Pereboom 2014: 61)

6For an action to be freely brought about by the agent, she must be the one who ultimately settles which of the forking paths available to her she will take. However, under an event-causal perspective, the final decision is not settled, neither by the events that precede it, nor by the agent, whose role is exhausted by the role those events play in the causal chain. This perspective renders the agent a spectator, rather than an actor. This is hardly something a libertarian should feel comfortable with.

7It is important to note, however, that full authorship of an action is rendered impossible not only by an event-causal theory of action production, but also by an aggregational account of the agent (cf. Rigato 2015, 2016). If the agent is a collective entity composed of many parts and endowed with properties, such as the ability to deliberate and choose, which are reducible to the properties of those parts and their relations, then the causal events she is involved in are actually grounded in more fundamental interactions at the neural, chemical and physical level. For the agent to be the effective cause of an action, and not merely the locus where sub-personal causings take place, she must have non-derivative causal powers, such as the ability to form an intention to act, to which one can trace the ultimate source of what happens. Given the well-known problems facing substance dualism, I believe a contemporary agent-causal theory should help herself with an emergentist account of the agent’s irreducible causal powers.

8For argument’s sake, let us assume this section was sufficient to inspire the free willist to embrace agent-causalism and opt for emergentism as its metaphysical supporting view. I will now develop such an account and assess its logical and empirical implications.

3. Challenges of an emergentist agent-causal account

9In order to develop my argument, I will start by explaining the concept of emergence and then present what I consider to be its conditions of possibility. In assessing them, I will show that the emergentist agent-causalist will not have to renounce the supervenience of the mental on the physical, which I believe is crucial for the plausibility of her account.

3.1. Defining ontological emergence

10One must never underestimate the important distinction between epistemological and ontological types of emergence. The former type is largely uncontroversial and concerns the relation between theories: according to it, it is impossible to deduce higher-level laws or properties from our knowledge of the lower-level elements and of the laws that govern their behavior and their relations. The limited significance of this position is shown by the fact that this might be only a practical impossibility that derives from the imperfections of our epistemic access to reality.

11In turn, ontological emergence has to do with the relation between entities themselves (beyond the theories that describe them). An emergent entity is a non-fundamental substance or property which depends on a lower-level structure in order to exist but which possesses certain characteristics that are in-principle irreducible to that lower-level domain. This irreducibility does not depend on our temporary ignorance; it is a genuine insufficiency of the lower-level reality (for example, the neurobiological states and events) to provide the conditions of possibility for phenomena that arise when a certain threshold of complexity is crossed (for example, consciousness). In order to account for this novelty, fundamental emergence laws are usually postulated.

12What is most important about the possibility of ontologically emergent entities (properties or substances) is that they carry with them new causal powers that are not derivative, i.e. that are basic (despite belonging to composite individuals). Unlike structural properties, the causal powers of which just amount to the collective entity’s parts having certain properties and relating to each other in a certain way, emergent causal powers are not explainable on the basis of the causal powers of their emergence base. This means that the causal effects the upper-level entity can have are distinct from the effects its emergence base can have, even if we take into account all its parts and their relations.

13To sum up, an emergent entity cannot be epiphenomenal – it has to be able to have causal effects in the world –, and it has to be novel – which means that its causal efficacy as a whole has to amount to more than the sum of the causings of its parts. Hence the following definition:

Ontological emergence is a relation between entities (substances or properties) belonging to different levels of organization of reality (different lengths and time scales), in which the lower-level structure gives rise to the higher-level entity without the novel causal properties manifested at the emergent level being derivable from or causally explainable in terms of the properties of the constituent or subvenient elements that originate them, nor in terms of the laws that govern their behavior and interactions.

  • 1 .Subvenient entities are lower-level properties, events, substances or laws that are in a natural s (...)

14Several clarifications are in order. First, I refer to lower-level elements as “constituent or subvenient”1 because it is important that the entities that form the emergence base be allowed to consist in mereological parts as well as not. The relationship between the brain and the conscious mind, for instance, which is what concerns me in this paper, is for sure not a part-whole relation (thus failing to count as a constitution relation) but it is a very good candidate for an emergence relation nonetheless. Second, note that in this definition there is an epistemological element that I failed to eliminate: explanation. In fact, it is not sufficient to claim that the causal powers of an emergent entity are distinct from those of its emergence base. The causal powers of a composite whole are very often distinct from those of its parts but if they derive from them, in the sense that the physical and chemical properties of the bottom-level structure are sufficient to explain the properties we can find in the compound, they cannot be considered emergent. On the contrary, in cases of emergence, “it is not possible to trace the determinative chain that goes from the emergence basis to the emergent” (Sartenaer 2005: 5). The fact that we have to use the epistemological concept of explanation in the definition of ontological emergence is revealing of our limitations as subjects, of course. We cannot assess the degree to which a certain portion of reality (the emergence base) is a sufficient condition for the coming to be of another portion of reality (the emergent entity), independently from our epistemic access to both of them. Because of this, it is very hard to establish whether a certain entity is only epistemically or also ontologically emergent.

3.2. The agent-causal power

15According to the agent-causalist and emergentist theory of agency, every free action is non-derivatively caused by the agent, who exercises some form of downward causal power over her body. In other words, the agent-cause of the action performed is a certain emergent part of the organism performing the action – a part I will call the “emergent self” – which can cause bodily movements to occur at will with an ability that does not reduce to the neural causings which underlie its activity.

16The emergent self does not have to be a hard core with no structure. It can be a dynamical system that emerges from the interactions between its parts. Here is the difference between an aggregate and a system, according to Alicia Juarrero:

In an aggregate, the properties of the parts do not change depending on whether or not they are part of the aggregate. In a system, on the other hand, the properties of the components depend on the systemic context within which the components are located. (…) Correlation and coordination among the parts confer a peculiar unity on the overall system. (Juarrero 1999: 109)

17Systems are not mere epistemic entities, they are present in our everyday reality, from cells and multicellular organisms whose identity is dynamically grounded on the interactions that keep them together and integrated, to the weather or the food we eat. These are all systems in which all the parts are correlated and depend on the relation they are embedded in, in order to be what they are. The difference between these common non-emergent systems and possible emergent ones is the reducibility of the causal properties of the former, in contrast with the genuine novelty of those of the latter. So the emergent self may be a dynamic system like many others in our body, whose unity derives from it being a system, but which is emergent because it can downwardly cause events in the material substrate that produces it in such a manner that one cannot take those causings as mere macro effects of many micro causings.

18Given the subjective and qualitative nature of the conscious experience of an agent, the whole that emerges from this system’s intrinsic and extrinsic relations is uniquely endowed with a first-person unified perspective on the world, which guides the use of its ability to voluntarily control its own movements – the ability we call the will.

19This happens with human as well as with non-human complex animals. Unlike the paramecium, which moves reactively only, most animal agents are capable of controlling and directing their bodily movements and interactions with the world purposefully. And this faculty is guided by the unified and subjective point of view with which their active self relates to the world and authors these actions. In the words of Helen Steward, who grounds agency in this body/owner distinction:

“Something that can make its body do various things must be a thing of which it at least makes sense to say it ‘has a body’ – something that can reasonably be regarded as an ‘owner’ of its body. And it is only of some sorts of entity that it makes sense to say that they ‘have’ bodies, thereby separating what is moved (a body or a body part) from what is doing the moving (an animal). It is these entities that are potentially sufficiently complex to sustain an owner/body distinction which I call ‘agents’ (…). Which sorts of entity may be said to ‘have’ bodies (…)? Entities with a mind.” (Steward 2012: 17)

3.3. The break of Causal Closure

20The main reason why the emergentist hypothesis is met with suspicion in scientifically inclined circles is the idea that the physical world is closed to causal influences from outside its frontiers. The principle of the Causal Closure of the Physical (CCP), that many mistake for an unquestionable axiom of physical theory, asserts roughly that for every physical event (or its chances) there is a sufficient physical cause, insofar as there is any cause at all. At face value, this seems to entail one of two consequences: putatively non-physical properties and objects are either actually physical after all or they are epiphenomenal. In other words, either the chemical, the biological, the psychological and the social domains are only different levels of description of one same reality that can in principle be understood in terms of the laws of a complete physical science, or the emergent levels are distinct from the physical reality but there is nothing left for them to cause, since all effective causes are physical causes.

21CCP can be read in two different ways: as stating the completeness of the microphysical (as opposed to the chemical, biological, etc.) or affirming the completeness of the physical, intended as the material reality that exists in space-time and can be described or explained in quantitative terms (as opposed to the subjective nature of conscious thoughts, say). The former reading is with no doubt the most common in the philosophy of science literature and asserts that every physical cause is ultimately a microphysical cause. The latter is mostly discussed in philosophy of mind and claims that every cause is a material (hence non-mental) cause. This version of CCP appears as particularly threatening, since people cherish the causal autonomy of mental properties more than that of any other property. If the solidity, mass and speed of a cannonball are nothing but the global effect of the properties of its tiny parts, that does not disturb us much. But if our conscious states (intentions, desires, decisions) are epiphenomenal, then our vision of who we are and how we act in this world is in need of major revision.

22Authors such as Robert Bishop (2006, 2010) and E.J. Lowe (2008) have devoted a considerable part of their work to questioning CCP in its first sense. Their main line of argument relies on the consideration that the principle of Causal Closure is not a scientific truth but rather a metaphysical assumption for which no proof can be given. In the present context, however, it is important to note that one can challenge aggregational accounts of the agent by questioning only the second interpretation of CCP, and leaving the first untouched. By this I mean that one can consider that the whole material world can be understood in reductionistic terms, i.e., that all the hierarchical levels are connected by some sort of “building relation” (Bennett 2011) and that, say, the laws of chemistry and biology describe macro interactions which are entirely driven by micro causings. Still, the mental world can be seen as something new, distinct and genuinely emergent, something that is ontologically irreducible to the non-mental domain and that has new non-derivative causal powers.

23John Dupré is one of the authors that questions CCP in the material-to-mental sense (even though he is an anti-reductionist in the first sense as well), based on the argument that accepting the postulate of causal closure would render mental causation impossible. He uses the example of a person wanting to drink a glass of water (mental domain) and moving her arm accordingly (“physical” domain): if one assumes CCP to be true, the extraordinary coincidence of the co-occurrence of appropriate events at the two levels (the level of the mental intention to drink and the level of the physical causes that produce the actual movement) remains a mystery:

All the physical movements of the agent would have happened even if the mental occurrences had not, if, that is to say, there had been no principles or laws requiring mental processes or events to come along for the ride with the physical ones. (…) Causal completeness at the microlevel must entail reductionism, at the very least in the sense of the supervenience of everything else on the microphysical. And even supervenience, I claim, is sufficient to deny any real causal autonomy to higher structural levels. (Dupré 2001: 161, 162)

24The alternative would be to assume that mental states are actually physical, but that is something neither Dupré nor the agent-causalists are willing to do.

25Dupré associates Causal Closure to the principle of supervenience, intended as the asymmetric covariance relation whereby all macro changes that non-fundamental entities undergo depend on corresponding micro changes, but not vice-versa. At face value, in fact, supervenience seems not to leave room for downward causation, as this relation entails that mental properties depend on neurobiological properties, and not the opposite. When a certain brain state at t1 causes another brain state at t2, there seems not to be anything left for the related mental state to cause, since all the changes taking place at that mental level (thus, the changes from the mental state at t1 to the mental state at t2) are taken to depend on the underlying physical changes.

26I agree with Dupré that Causal Closure entails the epiphenomenality of any state or event which is distinct from the physical states and events on which it supervenes, and that this is extremely unintuitive. However, I do not think supervenience is the problem. Clausal Closure must be broken for the mental to be allowed to cause change at the physical level, but supervenience, or at least natural supervenience, can be kept.

27In his famous 1996 book The Conscious Mind, David Chalmers presents a very useful distinction between logical and natural supervenience. Logical supervenience is a relation that holds between properties A and B such that there cannot be a world in which A (the subvenient property) is present without B (the supervenient property) being present as well. This is the type of relation that holds between the aesthetic properties of a painting and the physical substrate that realizes them. If one wanted to reproduce a certain painting, one would only have to duplicate every single physical particle that constitutes it, in its proper place and entertaining the same relations to all the others as in the original painting, and the aesthetic properties of the painting would come along as a bonus. This is the type of supervenience that Chalmers’ zombie hypothesis stands against and, despite my sympathy towards it, it is not what interests me here. When discussing the plausibility of an emergentist account of the agent, I’m mostly concerned with natural supervenience, the type of asymmetric covariance relation that science assumes to hold between lower-level and upper-level properties in this world. This is the type of supervenience that I contend does not have to be called into question even if emergence is true of some phenomena or levels of organization.

3.4. Why supervenience is not the problem

28Assuming that the material (non-mental) world is not causally closed, that ontological emergence is possible and that at a certain point in the scale of complexity some emergent entities appear, one does not have to imagine that those entities’ autonomy is such that some of the changes they undergo at the upper level might happen regardless of corresponding changes at the bottom level. As I have argued at length (Rigato, forthcoming), even if supervenience holds between the emergent level and its emergence base, such that there is always covariance with a bottom-up entailment, the upper-level entities, which are realized at t1 by the subvenient entities, can affect the world at t2 in ways that are not merely the consequence of how the subvenient elements affect each other. There is nothing a priori incoherent about this, unless one assumes that the bottom-most level obeys causally deterministic rules.

29If every detail of the lower-level is already defined and the laws which guide its evolution in time are deterministic, then there is nothing left for the upper level to cause, neither in a top-down way, nor as a same-level causal event. Imagine a dust particle suspended in a fluid, undergoing Brownian motion. Its space of physical possibilities would obviously be much broader if it were outside this context. However, given its concrete location and its relation with all the other particles around it, that possibility space is limited to what the laws allow in the concrete situation it is in. Do Newtonian laws allow for diverse alternatives in the particle’s next move given this situation? Of course not, since they are deterministic! Now picture an atom inside your brain, instead. If there is no indeterminism at the level of the laws that govern that atom’s behavior, then how can neurons and molecules that are made up of billions of atoms like this have a downward influence over its movement? Only by means of the interactions each atom has with its neighbors, which constrain its motion just like a dust particle in a fluid. The downward constraint exerted by the molecules and neurons that atom is embedded in is only apparent and derivative, for all the determination happens at the level of the parts.

30When the biological system which we call an agent decides to make a bodily movement such as reaching for a glass of water, she is influenced by reasons, motives and traits of character, just as she is causally conditioned by physical affordances and liabilities. The complex interaction among all these elements is what ultimately structures her possibility space. However, if all the abstract degrees of freedom of each particle in the agent’s motor cortex, as well as in her reaching hand were limited by the specific circumstance in which they are embedded (the quantitative values of their mass, charge, location and momentum as well as the values of all the other elements in that physical context), such that there would only be the objective probability of 1 that a certain outcome would become effective, then whatever may be the psychophysical interactions that might take place, there would be no more open alternatives than that one outcome. If the laws were deterministic, the trajectory of each particle could not change, neither could the timing of a neural spike.

31But what would happen if one were to suppose the bottom-level to obey probabilistic laws instead? If it were possible for a certain brain state at time t1 to be compatible with different futures at time t2, then there would be nothing contradictory in the idea of a “supervenience-friendly” downward causation. The irreducible self of the agent, being in a certain conscious state, which in turn supervenes on a certain indeterministic neural state, the instant prior to the decision, can exert her causal influence over which of the different possible alternative decisions comes to be. Those alternatives are left open by the indeterministic nature of the physical laws, and the agent’s intervention is what renders the actualization of one of them non-random. This underlying indeterminism comes as no surprise, of course, since it is precisely what, by definition, incompatibilism assumes to be true of the world.

3.5. The location of indeterminism

32Apparently, one can imagine this downward influence being exercised by the conscious self on any one of the underlying levels (e.g. nerves, neurons) or directly on the bottom-most level (e.g. quantum domain). However, given the success of the methodological reductionism of the natural sciences, I believe it is sensible to adopt a sort of reductionism-by-default attitude regarding the material world. This means assuming the different levels of material reality to be related by some sort of bottom-up determination relation, but still allowing for the conscious level to constitute an exception, given the radical discontinuity we find between the objective quantitative third-person reality of stones, plants, bodies and brains, and the subjective qualitative first-person phenomenology we all experience as conscious persons (cf. Nagel 1974, Jackson 1982, Searle 1992). Considering emergence to enter the picture only in the transition from the neurophysiological to the psychological domain, however, bears as a consequence that the levels of organization of non-conscious reality are regarded as being reducible to one another and thus synchronically determined from the bottom up, which entails that we cannot postulate a direct downward causal effect of the mental on the biological, for example. We should thus opt for a model which endows the conscious self with the power to affect directly the bottom-most level, on which all the other levels supervene.

33Quantum mechanics, which is predominantly interpreted as a truthful description of a genuinely indeterministic realm, fits perfectly into this picture. Put simply, the picture is this: there are parallel possibilities at the quantum level, which are amplified in the brain, leading to alternative neural connections, on which different conscious states supervene. By indeterministically choosing one path rather than another, the conscious self intervenes to determine which of these alternatives becomes actual.

34This diagram may help see this better:

Fig. 1: Synchronically, the conscious self (composed of different mental states M1-Mn but endowed with a new causal power that none of them individually has) emerges out of the neural multi-layered substrate. The self is an effective cause that can determine at t1 the actualization at t2 of one of the different quantum states that at t1 have a certain probability of being partially caused by Q at t2. Depending on the intervention of the conscious self, either the 1st (Q1-C1-B1-SELF1) or the 2nd (Q2-C2-B2-SELF2) psychophysical reality will happen.

35According to this schema, at the bottom-most level, indeterministic laws allow for one cause (Q) to be a necessary but insufficient condition for the system to evolve into different alternative quantum states (Q1 or Q2). These states constraint other states at the chemical and biochemical levels, however, even if the material world is structurally organized in such a way that each hierarchical level determines the next one in an upward manner, the mental domain is emergent and the conscious self simultaneously contains and transcends her diverse mental states. The fact that the physical world is not causally closed and that quantum laws are only probabilistic allows this emergent entity to exercise its causal powers over the material reality, contributing at a given instant t1 to the production of a new state of affairs at t2.

4. Last empirical considerations

36You may now ask: why shall we postulate the truth of neural indeterminism, rather than simply assuming the break of supervenience? Given that the macroscopic world of persons and brains has traditionally been considered to work deterministically, should we not prefer a solution that does not have to question causal determinism? Several authors have recently argued in favor of a compatibilist version of agent-causalism (Markosian 1999, 2012; Pereboom 2015; Franklin 2016). Why should the agent-causalist not opt for this alternative instead?

37There are several reasons in favor of the incompatibilist alternative I suggested. They all have to do with plausibility. On the one hand, to postulate that every free action would break supervenience, so that upper-level changes can take place independently from micro-level changes, would certainly fuel the case against agent-causalism. The assumption that the mind depends on the brain and has a covariance relation with it has been the guiding principle of neuroscientific research, a field in which extraordinary progresses have been made in the past decades. Today it is possible to infer from behavior and subjective reports the existence of chemical imbalances in the brain that can be treated pharmacologically, physical lesions that can be confirmed later via brain imaging techniques, etc. All this must be taken seriously into account when dealing with the associated philosophical problems.

38On the other hand, the truth of ontological indeterminism at the most fundamental level of reality is coherent with the current scientific picture of the world and thus comes surprisingly inexpensive as a solution for the problem of the possibility of downward causation. Certainly, the distance between phenomena such as quantum superposition and synapses is immense and large numbers are usually considered to wash out the effect quantum fluctuations can have on a macro scale. However, this distance can be bridged, and the recent field of Quantum Biology has shown that quantum effects can sometimes persist in biological large and warm systems such as the brain (Lambert et al. 2013). Also, neural interactions are probabilistic processes which can plausibly be grounded on genuine indeterminacy (cf. Glimcher 2005).

39I admit that the idea that the conscious self is not just a product of the neural correlates of its mental states is certainly not popular in contemporary philosophy and much less in neuroscience. However, the immense gap between neurobiology and phenomenology is still as deep today as it was fifty years ago when Nagel published his influential 1974 paper, and that prevents anyone from just dismissing an emergentist hypothesis without further evidence. We simply do not know how the conscious can come out of the unconscious. Several non-materialist eminent scientists over the years have contributed different theories of how a downward causal relation between the conscious self and its body could be the answer to many open questions in science (cf. London & Bauer 1939, Von Neumann 1932/55, Wigner 1967, Popper & Eccles 1977, Eccles 1970, 1994, Penrose 1989, Hameroff & Penrose 2014). All of them regard the possibility that mind and brain interact by means of quantum physics. Their interaction is enabled by the existence of genuinely indeterministic neural processes in the brain, crucial to its functioning, the probabilities of which could be influenced by the conscious self.

5. Conclusion

40As we conclude the itinerary traced in this paper, I hope my main point is clear: the metaphysical requirements of agent-causalism are not excessively demanding. Mental-to-physical interaction can fit into our scientific picture of reality without contradicting what we currently consider to be our best models of the functioning of the world.

41The one condition that agent-causalism cannot do without is the falsity of the principle of the causal closure of the physical lato sensu. But given that causal closure is nothing more than a metaphysical assumption, the agent-causalist can take emergentism as her ally and choose between two theoretical alternatives: a) keeping supervenience or b) keeping causal determinism. The first choice implies assuming that indeterminism is true of the bottom-most level of reality, which leaves room for the agent to intervene in the actualization of one amongst several alternative possibilities in the unfolding of reality, despite the asymmetric covariance relation between her conscious self and her strictly physical brain. The second alternative implies assuming the structure of reality to be such that upper levels of complexity may sometimes evolve in ways that are independent from the evolution of the levels below; only such a hypothesis would allow for downward causation in a deterministic world.

42I hope to have made a good case for the first of these alternatives and hence contributed to the consideration of agent-causalism as a legitimate incompatibilist alternative to event-causal accounts of agency and reductionist accounts of agents.



Bennett, K. (2011) ‘Construction area (no hard hat required)’. Philosophical Studies, 154, 79–104.

Bishop, R.C. (2006) ‘The hidden premiss in the causal argument for physicalism’. Analysis, 66, 44-52.

Bishop, R.C. (2010) ‘Free will and the causal closure of physics’. In Chiao, R.Y., Cohen, M.L., Leggett, A.J., Phillips, W.D., and Harper, C.L. (eds.), Visions of discovery: new light on physics, cosmology, and consciousness (Cambridge: Cambridge University Press), 601-611.

Chalmers, D. (1996) The Conscious Mind: in Search of a Fundamental Theory. New York: Oxford University Press.

Clarke, R. (2003) Libertarian Accounts of Free Will. New York: Oxford University Press.

Dupré, H. (2001) Human nature and the limits of science. Oxford: Clarendon Press.

Eccles, J.C. (1970) Facing Reality. Philosophical Adventures by a Brain Scientist. Heidelberg: Springer-Verlag.

Eccles, J.C. (1994) How the SELF Controls Its BRAIN. Berlin: Springer-Verlag.

Ekstrom, L.W. (2000) Free Will. A philosophical study. Boulder, CO: Westview Press.

Franklin, C. (2016) ‘If Anyone Should Be an Agent-Causalist, then Everyone Should Be an Agent-Causalist’. Mind, 125, 1101-1131.

Ginet, C. (1990) On Action. Cambridge: Cambridge University Press.

Glimcher, P. (2005) ‘Indeterminacy in Brain and Behavior’. Annual Review of Psychology, 56, 25–56.

Goetz, S. (2008) Freedom, Teleology, and Evil. London: Continuum.

Hameroff, S., Penrose, R. (2014) ‘Consciousness in the universe. A review of the ‘Orch OR’ theory’. Physics of Life Reviews, 11, 39–78.

Jackson, F. (1982) ‘Epiphenomenal qualia’. In Lycan, W.G., Prinz, J.J. (eds.) (1999) Mind and Cognition. An Anthology, 2nd Edition (Oxford: Blackwell Publishers), 440-446.

Juarrero, A. (1999) Dynamics in Action. Intentional Behavior as a Complex System. Cambridge, MA: MIT Press.

Kane, R. (1996) The Significance of Free Will. New York: Oxford University Press.

Lambert, N., Chen, Y.-N., Cheng, Y.-C., Li, C.-M., Chen, G.-Y., Nori, F. (2013), ‘Quantum biology’. Nature Physics, 9, 10-18.

London, F., Bauer, E. (1939) La théorie de l'observation en mécanique quantique. Paris: Hermann.

Lowe, E.J. (2008) Personal Agency. New York: Oxford University Press.

Markosian, N. (1999) ‘A Compatibilist Version of the Theory of Agent Causation’. The Pacific Philosophical Quarterly, 80, 257-277.

Markosian, N. (2012) ‘Agent Causation as the Solution to all the Compatibilist’s Problems’. Philosophical Studies, 157, 383-398.

McCann, H. (1998) The Works of Agency. Ithaca: Cornell University Press.

Nagel, T. (1974) ‘What is it like to be a bat?’. Philosophical Review, 83, 435-450.

O’Connor, T. (2000) Persons and Causes. New York: Oxford University Press.

O’Connor, T., Wong, H.Y. (2005) ‘The Metaphysics of Emergence’. Nôus, 39, 658–678.

Penrose, R. (1989) The Emperor's New Mind: Concerning Computers, Minds and The Laws of Physics. Oxford: Oxford University Press.

Pereboom, D. (2014) ‘The disappearing agent objection to event-causal Libertarianism’. Philosophical Studies, 169, 59-69.

Pereboom, D. (2015) ‘The Phenomenology of Agency and Deterministic Agent Causation’. In Altman, M., Gruenig, H. (eds.) Horizons of Authenticity in Phenomenology, Existentialism, and Moral Psychology: Essays in Honor of Charles Guignon (New York: Springer), 277-294.

Popper, K.R., Eccles, J.C. (1977) The Self and its Brain: an argument for interactionism. London: Springer.

Rigato, J. (2015) ‘Reductionism, Agency and Free Will’. Axiomathes, 25, 107–116.

Rigato, J. (2016) ‘The Agent as Her Self: How Taking Agency Seriously Leads to Emergent Dualism’. Rivista Internazionale di Filosofia e Psicologia, 4, 48-60.

Rigato, J. (forthcoming), ‘Downward causation and supervenience: The non-reductionist’s extra argument for incompatibilism’, Philosophical Explorations. DOI:

Sartenaer, O. (2015) ‘Synchronic vs. diachronic emergence: a reappraisal’. European Journal for Philosophy of Science, 5, 31-54.

Searle, J.R. (1992) The Rediscovery of the Mind. Cambridge, MA: MIT Press.

Steward, H. (2012) A Metaphysics for Freedom. Oxford: Oxford University Press.

Von Neumann, J. (1932/55) Mathematical Foundations of Quantum Mechanics. Princeton: Princeton University Press.

Wigner, E. (1967) ‘Remarks on the Mind-Body Question’. In Wigner, E.P. Symmetries and Reflections (Bloomington: Indiana University Press), 171-84.


1 .Subvenient entities are lower-level properties, events, substances or laws that are in a natural supervenience relation with upper-level entities. I will expand on the supervenience relation in sections 3.3. and 3.4.

Table des illustrations

Légende Fig. 1: Synchronically, the conscious self (composed of different mental states M1-Mn but endowed with a new causal power that none of them individually has) emerges out of the neural multi-layered substrate. The self is an effective cause that can determine at t1 the actualization at t2 of one of the different quantum states that at t1 have a certain probability of being partially caused by Q at t2. Depending on the intervention of the conscious self, either the 1st (Q1-C1-B1-SELF1) or the 2nd (Q2-C2-B2-SELF2) psychophysical reality will happen.
Fichier image/jpeg, 132k


Center for Philosophy of Sciences of the University of Lisbon and Champalimaud Research

Le texte et les autres éléments (illustrations, fichiers annexes importés) sont sous Licence OpenEdition Books, sauf mention contraire.


Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search